SAN FRANCISCO, October 08, 2024--LiveRamp (NYSE: RAMP), the leading data collaboration platform, today revealed the first Artificial Intelligence (AI) connections to its network, enabling marketers to drive even better outcomes from LiveRamp’s premier global ecosystem, by easily tapping the ...

LiveRamp Brings Artificial Intelligence-Powered Marketing to LiveRamp’s Network, Unlocking More Ways for Customers to Drive Value

<strong>LiveRamp</strong>® (NYSE: RAMP), the leading data collaboration platform, today announced that its CEO Scott Howe and CFO Lauren Dillard will present at the Morgan Stanley Technology, Media &amp; Telecom

LiveRamp to Present at the Morgan Stanley Technology, Media & Telecom Conference

SAN FRANCISCO, February 20, 2024--<strong>LiveRamp</strong>® (NYSE: RAMP), the leading data collaboration platform, today announced that its CEO Scott Howe and CFO Lauren Dillard will present at the Morgan Stanley Technology, Media &amp; Telecom Conference in San Francisco on Tuesday, March 5th at 8:45 a.m. PT ...

LiveRamp

liveramp

This article was , an independent, nonprofit organization that works side by side with consumers for truth, transparency and fairness in the marketplace. . copublished with Consumer Reports Learn more here By now most internet users know their online activity is constantly tracked. No one should be shocked to see ads for items they previously searched for, or to be asked if their data can be shared with an unknown number of “partners.” But what is the scale of this surveillance? Judging from data collected by Facebook and by non-profit consumer watchdog Consumer Reports, it’s massive, and examining the data may leave you with more questions than answers. newly described in a unique study Using a panel of 709 volunteers who shared archives of their Facebook data, Consumer Reports found that a total of 186,892 companies sent data about them to the social network. On average, each participant in the study had their data sent to Facebook by 2,230 companies. That number varied significantly, with some panelists’ data listing over 7,000 companies providing their data. The Markup helped Consumer Reports recruit participants for the study. Participants downloaded an archive of the last three years of their data from their Facebook settings, then provided it to Consumer Reports. By collecting data this way, the study was able to examine a form of tracking that is normally hidden: so-called “server-to-server” tracking, in which personal data goes from a company’s servers to Meta’s servers. Another form of tracking, in which Meta tracking pixels are placed on company websites, is visible to users’ browsers. Because the data came from a self-selected group of users, and because the results were not demographically adjusted, the study does not “make any claims about how representative this sample is of the U.S. population as a whole,” Consumer Reports noted. Participants were also likely more privacy conscious and technically inclined than typical users and more likely to be members of Consumers Reports. Despite its limitations, the study offers a rare look, using data directly from Meta, on how personal information is collected and aggregated online. Meta spokesperson Emil Vazquez defended the company’s practices. “We offer a number of transparency tools to help people understand the information that businesses choose to share with us, and manage how it’s used,” wrote Vazquez in an emailed statement to The Markup. While Meta does provide transparency tools like the one that enabled the study, Consumer Reports identified problems with them, including that the identity of many data providers is unclear from the names disclosed to users and that companies that provide services to advertisers are often allowed to ignore opt-out requests. One company appeared in 96% of participants’ data: the San Francisco-based data broker LiveRamp. But the companies sharing your online activity to Facebook aren’t just little-known data brokers. Retailers like Home Depot, Walmart and Macy’s all were in the top 100 most frequently-seen companies in the study. Credit reporting and consumer data companies such as Experian and TransUnion’s Neustar also made the list, as did Amazon, Etsy and PayPal. LiveRamp did not respond to a request for comment. What Exactly Does This Data Contain? The data examined by Consumer Reports in this study comes from two types of collection: events and custom audiences. Both categories include information about what people do outside of Meta’s platforms. Custom audiences allow advertisers to upload customer lists to Meta, often including identifiers like email addresses and mobile advertising IDs. These customers, and so-called “lookalike audiences” made up of similar people, can then be targeted with ads on Meta’s platforms. The other category of data collection, “events,” describes interactions that the user had with a brand, which can occur outside of Meta’s apps and in the real world. Events can include visiting a page on a company’s website, leveling up in a game, visiting a physical store, or purchasing a product. These signals originate from Meta software code included in many mobile apps, their tracking pixel, which is included on many websites, and from server-to-server tracking, where a company’s server passes data to a Meta server. The Markup has written extensively about the and how it has been used to surveil people as they , , , , and . Meta Pixel dial suicide hotlines buy their groceries take the SATs file their taxes book appointments with their doctors Website owners can configure the pixel to track user website interactions such as searches or filling out a form, sending each action to Meta, even if the user doesn’t have an account on Facebook. Although research tools like The Markup’s “Pixel Hunt” can detect the Meta pixel or SDK tracking, there is no way for a consumer to monitor the traffic between a company’s server and Meta’s. This Consumer Reports study looks at server-to-server data along with the rest. How Can I See My Data? Facebook users can browse through the list of companies that have sent their data to Facebook by going to: https://accountscenter.facebook.com/info_and_permissions From this menu, users can either download or access their information. To see the companies outside of Meta that have been sharing your information, select “Your activity off Meta technologies,” then “Recent activity.” You might be prompted to re-enter your password at this point. This view will show you the number of recent connections between various third-party businesses you visited and Meta, and examples of the activity that was shared. You can choose to prevent future sharing by the company by selecting “Disconnect.” To see detailed information about these interactions, . request a copy of your data But after jumping through the many hoops required to actually get your hands on this data, you may still be left with lingering questions. For example, in addition to easily recognizable company names, more than 7,000 companies in Consumer Reports’ study were named using unreadable gibberish, or just numbers that don’t mean anything to most users. Even those companies with readable names often did not include links to the company website. Some company names were ambiguous, such as “Viking,” which could be a number of different businesses. “This type of tracking which occurs entirely outside of the user’s view is just so far outside of what people expect when they use the internet,” Caitriona Fitzgerald, Deputy Director of the Electronic Privacy Information Center, told The Markup in an interview. Fitzgerald said that while users are likely aware that Meta knows what they are doing while they are on Facebook and Instagram, “they don’t expect Meta to know what stores they walk into or what news articles they’re reading or every site they visit online.” In the report, Consumer Reports calls for a number of policy proposals covering data collection practices, some of which could be part of a national digital privacy law, something that the organization has long advocated for. Among the recommendations specifically aimed at Meta’s technology and the advertisers who use it: Require companies to adopt “data minimization” strategies which call for the collection of the absolute minimum amount of data needed to provide the service being offered. Expand the powers of “authorized agents” to act on behalf of consumers to act on their rights. is a mobile app by Consumer Reports that allows users to quickly opt-out of and delete data from a long list of companies and data brokers on behalf of users; the app takes advantage of state privacy laws’ authorized agent provisions. Permission Slip Increase ad transparency by creating ad archives that allow the public to see all ads that have been served to users on a platform, following the lead of the EU's . Digital Services Act Existing Meta transparency tools should improve the quality and readability of the data that is made available so consumers can actually act on the information they review. Fitzgerald echoed these recommendations, saying the problem lies with the fact that the burden is on the consumer to take action to prevent this data collection. Even a “global opt-out” mechanism allowing users to avoid having their data shared is not enough because “that still requires the user to take an action to protect their privacy. A lot of people are not going to have the time or knowledge to do that,” said Fitzgerald. Vazquez, Meta’s spokesperson, said the company would “continue to invest in data minimization technologies to keep up with evolving expectations. As we cover in our terms, businesses are responsible for getting permission to share people’s information with companies like ours.” For now, the lack of a federal privacy law leaves consumers in most states with few options. “I think people should be encouraging their elected officials to pass privacy laws that require businesses to change some of these business practices to stop this ubiquitous tracking of our every click and every movement,” said Fitzgerald. Credits: , , , , , Jon Keegan Carlo Cadenas Maria Puertas Gabriel Hongsdusit Ryan Tate Michael Reilly Also published here Photo by on Amanda Dalbjörn Unsplash

If You're a Facebook User, Thousands of Companies Are Watching You

আপনি যদি একজন ফেসবুক ব্যবহারকারী হন, হাজার হাজার কোম্পানি আপনাকে দেখছে

Wenn Sie ein Facebook-Nutzer sind, werden Sie von Tausenden Unternehmen beobachtet

Si eres usuario de Facebook, miles de empresas te están observando

Si vous êtes un utilisateur de Facebook, des milliers d'entreprises vous surveillent

यदि आप फेसबुक उपयोगकर्ता हैं, तो हजारों कंपनियां आप पर नजर रख रही हैं

あなたが Facebook ユーザーなら、何千もの企業があなたに注目しています

당신이 Facebook 사용자라면 수천 개의 회사가 당신을 지켜보고 있습니다.

Se você é um usuário do Facebook, milhares de empresas estão observando você

Если вы пользователь Facebook, за вами наблюдают тысячи компаний

Facebook Kullanıcısıysanız Binlerce Şirket Sizi İzliyor

Nếu bạn là người dùng Facebook, hàng nghìn công ty đang theo dõi bạn

如果您是 Facebook 用户，成千上万的公司正在关注您

How phases out support for third-party cookies resembles how James Cameron filmed . There were multiple delays and more than a decade of waiting until both the publisher and the movie world were ready. Google Avatar 2 But is the world of advertising ready for the planned death of third-party cookies by the end of 2024? Since it’s the second delay in a row, it seems that Google aims to give the ad industry enough time to prepare for it. Why the wait? Why is it even a challenge to get rid of third-party cookies? How do these hold-ups affect ad businesses? Let’s try to figure it out. Controversy Behind Third-Party Cookies While first-party cookies store your data to make the faster and easier, 3rd-party cookies are used solely for advertising purposes. These web cookies allow the ad software to retarget potential customers with personalized ads as they surf the web. user experience With that, you might’ve already guessed what’s the controversy here. The users don’t like when someone messes with their privacy and spams their screen with ads. The ad business requires third-party cookies as it’s one of the pillars of digital advertising. Why Use Third-Party Cookies? To clarify things, here are some of the main reasons why the cookie-land exists. Publishers depend on third-party cookies less than advertisers, but their addiction is still unhealthy. Some reports state pubs due to cookie death. Why? Publishers heavily rely on data sharing to fill ad inventory. could lose up to $10 billion Advertisers make the most of third-party cookies, benefiting from targeting and optimized conversions. According to a global survey, in 2021, third-party cookies as their main data source. 51% of senior marketers use Cookies have been a part of the industry for a very long time; the very concept of programmatic advertising relies on third-party data. Without them retargeting, frequency capping, and measurement will either become more difficult or require a new approach. Personalization has its benefits for regular visitors as well. For 90% of the users, irrelevant ads are "annoying." For 53%, irrelevant ads are the top annoying things in their web surfing. So if a user sees an ad, it'd better be relevant. Why Refuse Third-Party Cookies So what's the deal with third-party cookies, and why does the Internet want to eliminate them that badly? There are three main reasons: 1. Consumer privacy To be completely honest, cookies are only a part of the global problem. A huge chunk of US consumers have trust concerns regarding how their data is protected.  Not only do third-party cookies contain sensitive information, but they are also vulnerable to different exploits and personal theft. Moreover, to share data to improve their experience, and 51% are uncomfortable sharing third-party info, like what sites they visit. The market aims to please the trend for user privacy, and in a world like that, there is little place for third-party cookies. 71% of consumers agree 2. Irrelevancy Imagine buying a toaster on the web. You browse different websites and watch reviews to buy the toaster of your dreams; once you buy it – your demand for toasters is closed for several years. At least, you think so. That little banner at the bottom of each web page you visit will disagree. Third-party cookies are often irrelevant in advertising, as they can’t give the marketer a bigger picture. In 2023 annoying toaster ads are outdated, and both the users and advertisers understand that. having an ad blocker installed only proves the point. 42% of Internet consumers 3. Lack of cross-platform tracking Modern users switch from device to device multiple times during the day, but third-party cookies are only capable of storing information on one device. So despite having the same Google account on both mobile and laptop, you won’t get the same ads. As a result, everyone suffers. Advertisers get incorrect view performance, publishers' ad inventory de-values, and users see outdated ads. Elimination of third-party cookies is a slow process that started years ago. Firefox has blocked third-party cookies since 2013; Safari did this completely in 2020. But Google Chrome holds , so the funeral only starts when the big browser brother decides so. History of Delays 66% of the desktop browser share Initially, Google planned to phase out third party-cookies by Q2 2022. In June 2021, Google postponed it to 2023, and in a year, we have the final (we hope so, at least) hold-up along with the approximate date – the second half of 2024. Why the delays? Well, officially, we have a blog post from Anthony Chavez, Google's VP of the Privacy Sandbox, stating: "The most consistent feedback we've received is the need for more time to evaluate and test the new Privacy Sandbox technologies before deprecating third-party cookies in Chrome." If we dive a little deeper, we'll see how far the root of the problem actually lies. The Chaos of Crossing As 81% of companies depend on third-party cookies, some speculate that Google will never really phase them out. But these claims have little grounds to be true. Firstly, Google is actively working on its Privacy Sandbox – a system that should ultimately cover the need for cookies (at least everyone hopes so). Secondly, it's not just browsers preparing for the death of third-party cookies. In October 2019, the Court of Justice of the European Union made active consent obligatory for tracking cookies. According to the new GDPR ruling, a website can't apply tracking or analytics to a user unless they tick "agree to all cookies" beforehand. Companies that shifted away from third-party cookies have prepared themselves for the inevitable. While they may experience inconveniences due to delays, we can only congratulate them. The situation is different for those who did not. Challenges for the Companies Bidding farewell to third-party cookies is not as easy as it may sound. By the start of 2023, 55% of companies are not ready to say goodbye. 42% of them expect lower ROIs and worse campaign efficiency management. The best-prepared countries are Italy, Spain, and Brazil due to their tight privacy regulations. Most advertisers and publishers have these particular challenges to overcome: Lack of the right tools The majority of demand-side platform do not support people-based identifiers. they can't get the required data with their current tools. 60% of them lack proper data management as well. 73% of companies state Lack of confidence in their own data Only 38% of marketers are confident that their data systems could win and retain customers. Solely 11% of respondents said that data is readily accessible to them. Lack of money The heavy reliance on third-party cookies messes up acquiring customers in new ways. But 44% of advertisers lack proper funding to do that, with 55% lacking access to data silos. 44% of marketers expect to increase their spending by 5-25% to cover the gap that the cookieless world brings in 2024. What's Happening Inside Google But not just the data and technologies concern advertisers so much. Say "Hello" – to Federated Cohort of Learnings(FLoC), Google's first-born alternative to third-party cookies. It was first introduced in 2019 as a part of the Privacy Sandbox project. The concept is simple – the algorithm anonymously groups people with similar interests into cohorts shared with advertisers afterward. However, testing has shown that despite the company's efforts, FLoCs were vulnerable to reverse-engineering that makes the user deanonymized. Naturally, this resulted in a delay as Google evolved FLoCs into Google Topics – a successor to FLoC that has a similar concept but works differently (don’t worry, we’ll explain how a bit later). Along with this system, Privacy Sandbox includes FLEDGE, Attribution Reporting API, and CHIPs. Testing all of these takes time. Moreover, there is a growing concern that the death of third-party cookies will only benefit Google, making its walled garden the only place to go for most brands. Being infamous for wanting power, to the Competition and Markets Authority(CMA), UK's anti-monopoly watchdog. Google has committed Thus, they can only release Privacy Sandbox when CMA approves that there are no competition concerns. Alternatives to Third-Party Cookies All of the above takes a lot of time, so delays are natural. But the vengeance will come upon third-party cookies. If you are a brand or a publisher, postponing the cross towards alternatives is foolish. The main issue is that there's yet to be an obvious alternative. Among the available options the ad industry has: First-Party Data With around $600B spent on digital ads in 2022, 77% of revenue accounts for the world's largest "walled gardens" that belong to Google, Facebook, LinkedIn, Amazon, and TikTok. This is great for advertisers, as these systems can provide them with first-party data and cross-device targeting. As for publishers, 71% think they while working with walled gardens. 51% of publishers consider walled gardens to be their most significant competitors. need to be in higher priority Universal IDs Universal identification is a concept where multiple sources of information about a user merge to create a single user identifier. Of course, the data is anonymous and can be passed onto approved partners through a supply chain. LiveRamp, TradeDesk, DigiTrust ID – this solution has many pioneers. As wonderful as it may sound, most of these systems rely on first-party data; thousands of publishers and advertisers must systematically collect and share user data. Furthermore, nobody knows if Google will support universal IDs and how that will work with Google's Privacy Sandbox. Zero-Party Data Zero-party data is information that the customer intentionally shares with the brand. While first-party data comes from the user's activity, zero-party data could come from surveys and polls. The great thing about zero-party data is that it all comes voluntarily with the expectation that providing it will give them a better user experience. With information like that, you don't have to question the source, worry about covering cross-device usage or doubt its accuracy. It's a win to win for both parties. What’s better,  if you’re a publisher that decides which ads to show up, you could potentially use zero-party data for . Another win, but this time for publishers. targeting through an ad server Forbes calls it "the new oil," and it's hard to disagree. Collecting zero-party data is a no-code task involving quizzes, surveys, forms, and more. It has to be engaging, creative, and benefit the user. Google Topics Google's alternative for cookies determines topics representing your top interests for the week based on your browsing history. Topics are kept only for three weeks and stored entirely on your device without involving extra servers(even Google's). When you visit a resource, Topics picks just three topics, one topic from each of the past three weeks, to share with the site and its advertising partners. Goodbye, Third-Party Cookies The hasty cat makes blind kittens, and we can't say postponing third-party cookies' death is necessarily bad. Yeah, there is no universal alternative for publishers and advertisers; Google constantly changes its course to apply the industry and the law, while users are thirsty for more privacy. Third-party cookies have been a part of the Internet for a long time, so the chaos surrounding their elimination is unsurprising. We can only advise procrastinators: "Start acting now." The wind of changes will probably ruin a lot of companies, so it's better to be prepared. Good luck! The lead image for this article was generated by HackerNoon's AI Image Generator via the prompt "cookies"

The Swan Song of Third-Party Cookies: What Do We Hear in 2023?

El canto del cisne de las cookies de terceros: ¿Qué escuchamos en 2023?

Le chant du cygne des cookies tiers : qu'entendons-nous en 2023 ?

तृतीय-पक्ष कुकीज़ का हंस गीत: हम 2023 में क्या सुनते हैं?

サードパーティ Cookie の白鳥の歌: 2023 年には何が聞こえるでしょうか?

O canto do cisne dos cookies de terceiros: o que ouvimos em 2023?

Bài hát thiên nga của cookie bên thứ ba: Chúng ta nghe gì vào năm 2023?

第三方 Cookie 的绝唱：我们在 2023 年听到了什么？

One of the largest pediatric networks in the country was sending personal information about children and their parents to Facebook, The Markup found. Nemours Children’s Health, which serves nearly half a million families in the U.S., had a Facebook tracking tool on its appointment scheduling website that shared details about the appointment with Facebook. The site also had a handful of other third-party trackers that share potentially sensitive information with data brokers. The Markup could not always determine what information the trackers were sending to data brokers, simply that the trackers were present. With Meta’s tracker, however, we found the Nemours site sending Facebook visitors’ IP addresses, information about the specific doctor and specialty the patient was scheduling an appointment with, and in some cases the first and last name of the child the appointment was for. That information, in combination with other data available to Facebook, could be used to link health conditions to individuals and to the Facebook profiles of their parents. And experts say the data-sharing with Facebook could violate patient privacy laws. The Health Insurance Portability and Accountability Act (HIPAA) covers health care providers like Nemours and services they provide, like scheduling appointments, said Charlotte Tschider, an assistant professor at Loyola University with a focus on information privacy and the health care industry. After The Markup reached out to Nemours, many of the trackers on the scheduling site were removed, but trackers from Facebook, Google, and Salesforce remained. Last week, after The Markup published a story about dozens of hospitals with Facebook, Nemours removed the Facebook tracker as well. sharing the sensitive health information of patients Not all health data is covered under HIPAA. The law specifically applies to health care providers, health insurance providers, and health care data clearinghouses. It also applies to companies that do businesses on behalf of those organizations, Tschider said. “When you are going to a covered entity’s website, and you’re entering information related to scheduling an appointment, including your actual name, and potentially other identifying characteristics related to your medical condition, there’s a strong possibility that HIPAA is going to apply in those situations,” Tschider said. The Markup found trackers on the Nemours site through our website scanning tool, . Blacklight A scan on May 9 showed the main website of the children’s health network, which boasts more than 95 locations in four states, had nine ad trackers and 10 third-party cookies. But its had more than double the number of trackers, with 25 ad trackers and 38 third-party cookies, including from companies like Facebook, Amazon, and Google. scheduling site There were also trackers from data brokers like Oracle, a tech giant that boasts data insights on more than , the advertising platform , and LiveRamp, which . 80 percent of the U.S. internet population MediaMath operates a data marketplace Blacklight also detected a session recorder on the page from the company Mouseflow. Session recorders can potentially track what people click on a page. Mouseflow’s website the company will sign a Business Associate’s Agreement, a legal agreement that needs to be in place to allow a HIPAA-covered organization to transfer covered health data to a third party. states When we asked whether such an agreement was in place between Nemours and Mouseflow, Mouseflow did not respond to the question. “Mouseflow does not allow the recording of PII or PHI. Mouseflow automatically masks IP addresses and all information website visitors are typing into form fields and search fields. This information is not being recorded,” Jakob Ohlsen Baagø, director of enterprise sales at Mouseflow, wrote in an email. “According to our terms, our customers are not allowed to record any PII or PHI and Mouseflow offers all the necessary tools to easily comply with this.”' Nemours Children’s Health did not respond to our requests for comment. Facebook collects data from websites , an analytics and marketing tool that developers can install on their websites. In 2018, that its pixel was embedded in more than two million websites. through the Meta Pixel Facebook told U.S. lawmakers The pixel can log specific actions on a page, like if a person on a retail site or signs of a product. The pixel is also present on non-commercial sites. adds an item to their shopping cart up for a free trial The Markup recently found . In that case, we found that Facebook was receiving such information from the FAFSA site as an applicant’s first name, last name, country, phone number, and email address. Facebook trackers on the Department of Education’s federal student aid application site Meta spokesperson Dale Hogan told The Markup that the company has systems that are supposed to detect and block sensitive health information coming from the pixel and other tracking tools. “Advertisers should not send sensitive information about people through our Business Tools as doing so is against our policies. While our system is designed to filter out potentially sensitive data it is able to detect, we are reaching out to Nemours Health directly,” Hogan said in an emailed statement. The Markup could not determine whether the Nemours data was in fact removed before storage, and Hogan did not comment on whether or not the data was retained or used by Facebook. A recent joint investigation by Reveal and The Markup found that the about appointments a reporter requested with crisis pregnancy centers. filters did not block information Pixels are primarily used by marketers who want to run targeted ads based on the data that Facebook collects—and hospitals aren’t immune to advertising needs. Health care companies spent more than $2.5 billion on marketing in 2018, . according to a study in the Journal of Medicine and Life The study noted that digital marketing gives the health care industry the “potential to significantly increase its coverage.” But privacy advocates worry that such marketing campaigns could end up leaking sensitive information about patients. “The hospitals have been incredibly greedy in trying to take advantage of social and digital media to capture potential new patients and to retain existing ones,” Jeff Chester, executive director of the Center for Digital Democracy, said. “They mindlessly adopted Facebook’s pixel at the beginning of Facebook’s expansion without thinking of the consequences.” AddThis, a marketing company that Oracle bought in 2016, has trackers on the scheduling page but not the main website. The company offers a free tool that loads tracking cookies and pixels, claiming to collect “up to 30 data points” from each visitor, . The Markup reported in 2020 “Once companies know there’s a child that’s sick in the household, you can market to that child quite differently, you can appeal to emotions better, you can market to the family,” Katharina Kopp, the deputy director of policy at the Center for Digital Democracy, said. “The data that is out there undermines our rights to move around without being profiled and compared to people like us.” Experts said HIPAA requires covered entities like Nemours to protect health information that is linked to 18 kinds of personally identifiable information, including IP addresses. “So using a name, an appointment date, or an IP address in connection with health services of a covered entity would likely qualify for HIPAA [protected health information] status,” Tschider, the Loyola professor, said. The law provides an exemption for or in specific instances where patients consent to sharing their information. de-identified health information While the Meta Pixel hashes patients’ names before it collects that data from Nemours’ website, hashing would not prevent Facebook from being able to link a patient to their Facebook profile. In addition, the data can be sent alongside Facebook cookies that allow Facebook to link a user of the scheduling site to their Facebook profile and an IP address. “It is particularly unfair to do this about children,” Kopp said. “This is going to stick with the family and the child, potentially, for the rest of his or her life.” Credits: , Alfred Ng Simon Fondrie-Teitler Also published here Photo by on Brandon Holmes Unsplash

Facebook Has Your Children's Info Thanks to This Hospital Network

In February 2022, I finally moved to the US on O1 work visa. This was preceded by 2.5 years of work on strengthening the case, waiting for approval. 1.5 years before this date, I started preparing for an interview for a Front End position. Currently working as a Senior Front-End Engineer at Roku. This post and the creation of the were motivated by the lack of detailed information about interviews for Front-End engineers in the USA. telegram channel What I did before Before relocation, I worked for 7 years in outsourcing companies in Belarus. This allowed me to use different technologies, approaches and work on various projects. Prior to relocating, I already held a position as Engineering Manager with 26 JavaScript engineers (this strengthened my CV, especially for a Staff position). Over the past year, I have conducted about 50 technical interviews in the department. But in the USA, I realized that my experience is not entirely relevant, and I need to start from a Senior position in order to figure out how the processes are built here, the culture, and how to conduct interviews, which are completely different here. I started preparing for the technical interview 1.5 years in advance while I was waiting for my visa. Every morning of my working week was spent for preparing. It became a habit over time and did not cause any discomfort. After a while, I was promoted to the Engineering Manager position as I was growing at a rapid rate. Preparation 1. Leetcode I solved 245 problems on leetcode. I mostly focused on arrays, matrix, trees, linked lists, strings, hash tables, sorts, and two pointers. In my , you can see the topics that I solved. profile are all the tasks that I solved. Here To familiarize and refresh data structures and algorithms in JS, I used the resource. javascript-algorithms To quickly identify a Leetcode problems patterns, read this article — . It helps identify the problem. 14 Leetcode problems patterns To determine the complexity of the code, there is a on Big O notation. cheat sheet For examples of ​​what an interview looks like, I used the resource. interviewing 2. Big Front End ( ) https://bigfrontend.dev/ I solved 50% of BFE problems (easy and medium). I published all these solutions on my t . This resource is most useful for interviews at Meta, but other companies are more or less likely to ask similar questions. elegram channel 3. System design To prepare for system design, I used videos. He covered most aspects of systems design that I encountered in real interviews. Evgenii Ray’s A very useful resource for system design and for real work (Web Performance) — 3perf All about accessibility 4. Behavioral interview for preparation and examples of answers. List of questions For the mock interview, I used the telegram channel and . FAANG Pramp I also used this and and resolved everything from there, reading it all the way through. telegram channel Evgenii Ray’s Notion On glassdoor, you can find some examples of questions and tasks that companies ask. And finally, I’m in the USA… Upon arrival, I had a 2-week jet lag, there was no way of passing the interview, I could not sleep normally. But after I adapted, I began to actively set up calls with recruiters, writing on LinkedIn. There were 20–30 messages with opportunities per day, but most of them are no-name companies or startups. Before replying to the message, I analyzed the company, investments, it is a public or private company, what they do and what technologies they work with (I basically focused on React.js, JS / TS stack, as well, the Front End position). This was the original filter. During the day I had 3–4 phone calls with recruiters, then 2 technical interviews. In total, I participated in about 20 mocks to understand how to behave in an interview and what my weaknesses are, but this does not compare with real interviews. In real interviews, I already had a noticeable growth, already after 1–1.5 months, I began to solve problems more efficiently, find out requirements, sell myself and answer behavioral questions, because after each interview I did work on mistakes, and corrected stories. I started with small startups and companies and I was ready for rejection (this is important!). Every day for 2 months I had phone calls with recruiters, technical screenings, on-site interviews. It is very convenient that now everything is online. After almost a month of technical screenings, I even said that I liked the interview process, there are so many interesting problems and people, but then the onsite time came … after the first 5-hour interview, my opinion changed in a completely different direction, it’s sooo hard! In the beginning, I had rejections and it was frustrating of course. But as soon as I received the first offer, my motivation immediately grows .. until the next rejection. I was ready for rejection, but it’s still unpleasant to hear, especially when you spent 5–6 hours and they send you a standard reply that we decided to move with another candidate without any feedback. This motivated me to reflect and analyze what went wrong. I took notes, and tried to find answers on how to solve the problem or what needed to be clarified and how to design something. This helped in future interviews to combine different solutions into one and produce a high-quality solution. Also, communication with recruiters helped me improve my English, at the beginning I might not even understand half of what they were saying to me, but then I got used to it and it was much easier. In the end, I had really good interviews, but I kept getting rejected for some of them. And the worst thing is, you don’t understand what went wrong. The interview process My interview stats: : Meta, Amazon, Apple, Google, Microsoft, Lyft, Uber, LinkedIn, Roku, Salesforce, PayPal, DoorDash, Robinhood, Roblox, Intuit, Zoom, Grammarly, Imply, Disco, Square, SigmaComputing, LiveRamp, Zoox, Palo Alto Networks, Snowflake, Aurora Tech Screens (26) : Meta, Amazon, Google, Microsoft, LinkedIn, Roku, Salesforce, PayPal, DoorDash, Zoom, Grammarly, SigmaComputing, LiveRamp, Zoox, Aurora Onsite interviews (15) : Google, Roku, Microsoft, PayPal, SigmaComputing, LiveRamp Offers When applying for a Front End position, you will likely be asked only questions about the frontend, but at least 1 algo-problem of the easy-mid level will be asked, mainly about the data structure and its processing (there are not a lot in JS :)). But some like to ask algorithms the old-fashioned way. All of the preparation materials I mentioned above covered the interview questions. The process is pretty standard: Call with recruiter Technical screening Onsite (virtual) interview: Algorithms Tasks for coding a widget or a JS function System design, with a focus on Front-End Behavioral interview 4. Negotiation 1. Call with recruiter On a call with a recruiter, you discuss general questions, learn more about the company and the team you are interviewing. They ask about your work permit. In conversation, they ask about your experience and check your English. They find out what stage of your job search you are at and can build on this, for example, to speed up your interview process (In Aurora, I said that I completed all my interviews and if we can quickly go through all the stages, I agree. As a result, all the stages were done to me in 4 days, and onsite in just 3 rounds). Calls often happen before you have sent your CV. Only if everything went well and there is a match, you will be asked to send your CV. They may ask about your salary expectations, but I answered truthfully that I have no idea how much I am worth on the market since I just arrived. This was enough for them to pass over the question. You can ask for all the information you need in the same call, including the technical stack of the team. If you have questions that are important to you, this is the time to ask them. For me, it was the terms for the green card and the tech stack. Recruiters respect your time, so you coordinate each call according to the calendar, choosing from the recruiter’s one, or recruiter can choose from your calendar. Do not specify your salary expectations in any case, if they are very demanding (it has only happened once), specify a wide range. Turo’s recruiter repeatedly asked me what I expect, and when I gave him the range he said our compensation was 2 times less and we wouldn’t be able to match. 2. Technical screening So you have passed the first stage and you have been assigned a technical screening. Usually, it is 45 minutes. On it, again, you will tell a little about yourself, you will be asked questions about your experience and will be given to solve some small problem (in 90% it will either make a small widget in JS and a framework, or an easy problem for algorithms, or rather data structures that is in JS). Occasionally, there were problems on Hackerrank with HTML, CSS, and JS, 2–3. These are fairly simple to solve. Sometimes you can use any editor of your choice. But most of the time I send a link to my online editor (even with code highlighting). The most important thing is not to be silent, to think out loud so that the interviewer follows your thoughts. Related to code execution, it’s 50/50, sometimes you need to run the code when you want and as much as you want, sometimes I don’t even need to run the code (there’s not even such a possibility). There was one time when I was asked to design an elevator system at a technical screening (without code), it didn’t end well :) 3. Onsite(virtual) interview The most difficult stage of the interview, as it duration 3–6 hours. Many companies allow you to split it into 2 days, this is up to you. It consists of coding, system design, and behavioral interview. You can ask the recruiter about each round of onsite to understand what to expect from them. are similar to technical screening: find out the requirements, discuss the solution, code, and test it. Coding rounds System design Always focused on front-end issues, no need to talk about highly loaded systems, databases, etc. You just need to describe some functionality in the browser, ways to optimize, customize and expand, describe how to transfer data to the server, how to store it, where to store it, and why. Sometimes you need to design several endpoints with request and response structures. Don’t forget accessibility, it’s very important! Behavioral interview Here you can expect questions about your experience, projects, and teamwork. Basically, these questions are “Tell me a situation when you ..” In general, the questions asked about the most challenging project, the conflict in the team, what we did when deadlines were approaching and we didn’t have time, and what our biggest failure was. All stories should be told according to the STAR (situation — task — action — result) methodology. The story shouldn’t be very long. You will be asked clarifying questions if you didn’t explain something or the interviewer misunderstood you. In most cases, the hiring managers conduct the interviews. In the end, this is a great opportunity for you to ask all your questions about the team and the product. Negotiation There is no less stressful or important stage. It took me almost a month. With some companies, there were even several rounds, on average, I raised offers for 70–100k. Ask for feedback about you to better understand how good you were and use it as a negotiation tool. But nothing works better than a counteroffer. But it is important not to go too far, look at the compensations on the levels.fyi website beforehand in order to understand the range. Some conclusions about companies In all companies recruiters do a very good job and support you every step of the way, tell you what to expect in each round, how to prepare, and tell you what the interview process looks like in general. : I had the worst interview ever. At all rounds I had problems in understanding what the interviewer wanted from me, I clarified the requirements most of the time and explained my solution because for some reason they couldn’t understand it. And when they started asking questions about JavaScript during the coding process, it turned out that they didn’t know it and had to spend more time explaining why I use it and how it works. As a result, there were 5–7 minutes left for coding, but this was enough to finish. In the last round, I got the hard problem with trees, that I didn’t fully solve. Amazon : Some of the interviewers again didn’t know JavaScript, only asked algorithmic problems and asked questions about how JavaScript works. It was more like an interview for a full-stack position. Microsoft : Only math problems and algorithms were asked. Zoox : Again, only algorithms were asked, although the position is pure Front-End. However, my friend’s interview focused only on front-end questions. Aurora : I didn’t expect so much attention from this company. I had a lot of calls with a recruiter, the duration of some was an hour, where he told everything in detail about the interview process, gave preparation materials, asked how the preparation was going and gave advice, including how to behave in an interview, what to ask, what better not to ask. The questions were on the frontend focus. After the onsite, the hiring freeze started right away… Meta : They have an Ember.js framework, so questions are only about pure JavaScript and the frontend. LinkedIn : Lost my feedback after the technical screening. Still searching. Apple : The interview process was with front-end focus. After getting approval from the Google Hairing Committee, your next step is team matching. I had mixed feedback between L5 and L4 levels. As a result, the committee gave me L4 without additional interview stages. Only after a successful team match, they will send you an offer with numbers, but how long this process can take depends on many factors. Google I had 2 team calls a week, although I could do every day. The process took 2 weeks. The criteria by which I compared the teams: Team size. Frontend, backend, infrastructure engineers. Tech stack (all Angular). Promotion to L5 (time and opportunity). Product. Responsibilities. Front end or full stack position. There was even a backend only. Development process. Who are the users of the product, how many. Perhaps this is an internal project or a large one for a large number of users. Evaluated how comfortable communicate with the manager or team leader for me. As a result, I matched with the Google Analytics team, which is responsible for the performance of this product. In this team, I saw many opportunities for growth and metrics for promotion. How do I accept the offer I compared offers and decided which one to accept according to the following criteria. It is very important that the product is of interest to you. I have many years of experience in developing streaming platforms and know the needs of users and developers. I can learn a lot and see many challenges for myself in such products. Roku gives me this opportunity. Product. I don't like Angular, I prefer React.js. This was one of the most important reasons why I chose Roku. Tech stack and tools. This is a very important thing. I don't want to burn out. During the interview process, you can ask questions about it and listen carefully to the answers. Work-life balance. Growth in technical terms and within the company to the lead position (Staff) Opportunities for growth. After the position of engineering manager, I was not ready for middle positions (yes, there were those too). Position level. Base and stock. Roku pays 85% in cash for senior positions, which is more profitable for me, as a beginner in this country, since at this stage money is needed here and now. Total compensation. Some companies start the green card process from the 1st working day, some after 3, 6, 12 months. I filtered companies based on 1 day and 3 months. But as it turned out, these terms can also be negotiated, I was able to negotiate from 3 months to 1 working day.
An important nuance, Meta does not issue green cards, I asked them about it at every call, at every stage of the interview, but I never received an answer, no one at all. But according to information on the Internet, everything is bad with them. Green Card Process. I don't like working in a team where you are one Front End engineer, or even 2. I want to grow and learn. The level of engineers in the team was also important for me. Team size and responsibilities. I prefer office work. Working from home is not comfortable for me. Some companies now offer to work 3 days in the office, 2 days remotely, and some offer full remote work. Remote work or office work. I live in San Jose and some companies have offices in San Francisco. It will take you at least 1.5 hours to drive here. Do you want to spend 3 hours a day on this? Office location. Tips Do not schedule more than 3 (but preferably 2) onsites per week. Last week I had 4 onsites in a week and it was hell! But I got 2 offers :) In an interview, your skills grow exponentially, so do not immediately pass the interview with the company of your dreams. After the interview you can talk to someone, it helps to relax. Do work on the mistakes after each interview to understand what could have gone wrong, look for answers to questions and tasks that were in the interview and try to figure it out. Prepare the text for your self-presentation for 1–1.5 minutes, you will tell it at each stage of the interview. For a behavioral interview, prepare and practice as many STAR stories as you can, I had about 20 that covered almost all questions. It took about a month to prepare them. Buy a paid subscription to Calendly (or a similar service) to make it easier to manage your interviews. Sending the dates in an email means they can respond within a week, and demand 5 dates (respectively), so these 5 days are blocked for you and you cannot schedule other interviews on them. If you have any questions or topics that I have not covered, ask in the comments here or on my ! telegram channel

How I Relocated to the US and Found a Job at Roku

Cómo me mudé a los EE. UU. y encontré un trabajo en Roku

Comment j'ai déménagé aux États-Unis et trouvé un emploi chez Roku

कैसे मैं अमेरिका में स्थानांतरित हुआ और Roku . में नौकरी पाई

私がアメリカに移住してRokuに就職した経緯

Como me mudei para os EUA e encontrei um emprego na Roku

Làm thế nào tôi chuyển đến Hoa Kỳ và tìm một công việc tại Roku

我如何搬到美国并在 Roku 找到工作

Last year, nearly visited the website of Planned Parenthood, a nonprofit that many people turn to for very private matters like sex education, access to contraceptives, and access to abortions. 200 million people What those visitors may not have known is that as soon as they opened plannedparenthood.org, some two dozen ad trackers embedded in the site alerted a slew of companies whose business is not reproductive freedom but gathering, selling, and using browsing data. See our data here: GitHub The Markup ran and found 28 ad trackers and 40 third-party cookies tracking visitors, in addition to so-called “session recorders” that could be capturing the mouse movements and keystrokes of people visiting the homepage in search of things like information on contraceptives and abortions. The site also contained trackers that tell Facebook and Google if users visited the site. Planned Parenthood’s website through our Blacklight tool The Markup’s scan found Planned Parenthood’s site communicating with companies like Oracle, Verizon, LiveRamp, TowerData, and Quantcast—some of which have made a business of assembling and selling access to masses of digital data about people’s habits. Katie Skibinski, vice president for digital products at Planned Parenthood, said the data collected on its website is “used only for internal purposes by Planned Parenthood and our affiliates,” and the company doesn’t “sell” data to third parties. “While we aim to use data to learn how we can be most impactful, at Planned Parenthood, data-driven learning is always thoughtfully executed with respect for patient and user privacy,” Skibinski said. “This means using analytics platforms to collect aggregate data to gather insights and identify trends that help us improve our digital programs.” Skibinski did not dispute that the organization shares data with third parties, including data brokers. A a localized website specifically for people in the Gulf region, including Texas, —churned up similar results. Blacklight scan of Planned Parenthood Gulf Coast— where abortion has been essentially outlawed Planned Parenthood is not alone when it comes to nonprofits, some operating in sensitive areas like mental health and addiction, gathering and sharing data on website visitors. , The Markup scanned more than 23,000 websites of nonprofit organizations, including those belonging to abortion providers and nonprofit addiction treatment centers. Using our Blacklight tool The Markup used the to identify nonprofits that have filed a tax return since 2019 and that the agency as focusing on areas like mental health and crisis intervention, civil rights, and medical research. IRS’s nonprofit master file categorizes We then examined each nonprofit’s website as publicly listed in . We found that about 86 percent of them had third-party cookies or tracking network requests. By comparison, when The Markup did a in 2020, we found 87 percent used some type of third-party tracking. GuideStar survey of the top 80,000 websites About 11 percent of the 23,856 nonprofit websites we scanned had a Facebook pixel embedded, while 18 percent used the Google Analytics feature. “Remarketing Audiences” The Markup found that 439 of the nonprofit websites loaded scripts called session recorders, which can monitor visitors’ clicks and keystrokes. Eighty-nine of those were for websites that belonged to nonprofits that the IRS categorizes as primarily focusing on mental health and crisis intervention issues. “As a user of this website, by sharing your information with them, you probably don’t assume that this sensitive information is shared with third parties and definitely don’t assume that your keystrokes are recorded,” Gunes Acar, a privacy researcher who copublished a , said. “The more sensitive the website is, the more worried I am.” 2017 study on session recorders Tracy Plevel, the vice president of development and community relations at Gateway Rehab, one of the nonprofits with session recorders on its site, said that the nonprofit uses trackers and session recorders because it needs to stay competitive with its larger, for-profit counterparts. The more sensitive the website is, the more worried I am. Gunes Acar, privacy researcher “As a nonprofit ourselves, we are up against for-profit providers with large advertising budgets as well as the addiction treatment brokers who grab those seeking care with similar online advertising tactics and connect them with the provider who is offering the greatest ‘sales’ compensation,” Plevel said. “Additionally we know user experience has a big impact on following through on treatment. When someone is ready to commit to treatment, we need to ensure it [is] as easy as possible for them before they get frustrated or intimidated by the process.” Other nonprofits had a significant number of trackers embedded on their sites as well. The Markup on The Clinic at Sharma-Crawford Attorneys at Law, a Kansas City legal clinic that represents low-income people facing deportation. found 26 ad trackers and 50 third-party cookies Rekha Sharma-Crawford, the board president of The Clinic, wrote in an emailed statement, “We take privacy and security concerns very seriously and will continue to work with our web provider to address the issues you have identified.” Save the Children, a humanitarian aid organization founded more than 100 years ago, had 26 ad trackers and 49 third-party cookies. March of Dimes, a nonprofit started by President Franklin D. Roosevelt that focuses on maternal and infant care, had more than 29 ad trackers on its site and 58 third-party cookies. City of Hope, a Californian cancer treatment and research center, had 25 ad trackers and 47 third-party cookies. Results of Blacklight scans of the homepages of Save the Children, March of Dimes, and City of Hope performed on Oct. 19, 2021. Paul Butcher, associate vice president of global digital strategy at Save the Children, said in an emailed statement that the organization “takes data protection very seriously.” Butcher also wrote that Save the Children collects some data through ad trackers “to improve user experience” and that the organization is in the process of revamping its data retention policies and recently hired a new head of data. March of Dimes and City of Hope did not respond to requests for comment. State-Level Privacy Laws Miss Nonprofits While health data is governed by HIPAA, and FERPA regulates educational records, there are no federal laws governing how websites track their visitors. Recently, a few states—California, Virginia, and Colorado—have enacted consumer privacy laws that require companies to disclose their tracking practices and allow visitors to opt-out of data collection. But nonprofits in two of those states, California and Virginia, don’t need to adhere to the regulations. Sen. Ron Wyden (D-OR), who has proposed his , said that nonprofits accrue a large amount of potentially sensitive data. own federal privacy legislation “Nonprofits store incredibly personal information about things we’re passionate about, from political causes and social views to which charitable causes we care about,” Wyden said in an emailed statement. “If a data breach reveals someone donates to a domestic violence support group or an LGBTQ rights organization or the name of their mosque, any of that information could be incredibly private.” Nonprofit leaders, however, argue that they lack the infrastructure and funding to comply with privacy law requirements and must gather and share information on donors in order to survive. “One of the most substantive and impactful uses of data by nonprofits has been our fundraising,” said Shannon McCracken, the CEO of The Nonprofit Alliance, an advocacy group made up of nonprofits and businesses. “Without the ability to cost-effectively reach prospective new donors and current donors, then nonprofits can’t continue to be as impactful as they are today.” But purposeful or not, privacy experts say, nonprofits are feeding personal information to data brokers and tech giants like Facebook and Google. “A nonprofit might share your phone number and name with LiveRamp. Tomorrow, a for-profit entity can then reuse that same data to target you,” said Ashkan Soltani, a privacy expert, and former chief technologist at the Federal Trade Commission. “The data flows that go into these third-party aggregators and data brokers come often from nonprofits as well.” Soltani, who was appointed executive director of the California Privacy Protection Agency on Oct. 4, helped draft the California Consumer Privacy Act, which was originally introduced with the nonprofit exemptions. A nonprofit might share your phone number and name…. Tomorrow, a for-profit entity can then reuse the same data to target you. Ashkan Soltani, California Privacy Protection Agency Many major nonprofits work with data brokers to help organize and analyze their data, Jan Masaoka, CEO of the California Association of Nonprofits, said. “People that have big donor lists use them extensively, pretty much all of them use one of the services,” Masaoka said. “They don’t keep it in-house, pretty much everybody keeps it with one of these services.” She noted that Blackbaud is a company that nonprofits often turn to. The ’s marketing material promotes a that combines donor data from more than 550 nonprofits with public information on millions of households. registered data broker co-op database Blackbaud didn’t respond to a request for comment. Because of a lack of funds, nonprofits also rely on third-party platforms—which also happen to be data brokers—to manage their data’s security and privacy, McCracken said. But these kinds of companies aren’t immune to cyberattacks either: Blackbaud in which hackers stole passwords, Social Security numbers, and banking information, according to a . disclosed a ransomware attack in 2020 Securities and Exchange Commission filing Hundreds of were affected, along with more than 13 million people, according to the Identity Theft Resource Center. charitable organizations, schools, and hospitals “They rely on this kind of problematic ecosystem to achieve their work, and as a result, they share number lists, email addresses, or browsing behavior with third-party advertising companies and subject their members to risk,” Soltani said. The Exception Unlike its predecessors in California and Virginia, Colorado’s privacy bill doesn’t have an exemption for nonprofits. In both California and Virginia, the bills’ main supporters gave nonprofits an exemption as a political maneuver. Alastair Mactaggart, a real estate developer and founder of Californians for Consumer Privacy, who was the driving force behind the California Consumer Privacy Act, said his proposal was already facing opposition from tech giants and didn’t want a political showdown with nonprofits, too. “You gotta take the first step, so we figured this was the one that would be the easiest to bounce off,” Mactaggart said. “Eventually, I hope that the big nonprofits are included as well.” David Marsden, the state senator who introduced the Virginia Consumer Data Protection Act, echoed that sentiment, reflecting that the law wasn’t perfect but still a good start. “Does this pick up everybody that it should, or exempt everybody who needs an exemption? Probably not, but it comes pretty close,” Marsden said. “We were able, with this bill, to get it passed without people getting up and objecting to what we were trying to do.” Colorado state senator Robert Rodriguez, who co-sponsored the state’s privacy bill, said he didn’t include an exemption for nonprofits because he felt that any entity that had data on more than 100,000 people should have to follow privacy protections. He also didn’t understand why other states had exemptions. “Someone that has over 100,000 records is a good size,” he said in an email. “They should have some protections or requirements to follow.” Correction This story has been updated to correct the title of Paul Butcher at Save the Children. He is the associate vice president of global digital strategy. Written by: and Alfred Ng Maddy Varner This article was and was republished under the license. originally published on The Markup Creative Commons Attribution-NonCommercial-NoDerivatives

How Come So Many Nonprofit Websites Have Ad Trackers?

Image by: Sébastien Thibault Data brokers’ millions of dollars in lobbying spending in 2020 rivaled that of some Big Tech firms By Alfred Ng and Maddy Varner The data brokers who’ve made fortunes from collecting and sharing millions of people’s personal information tend to fly under the radar. Names like LiveRamp or RELX might not be familiar to most Americans, but they’re making themselves known on Capitol Hill. Collectively, data broker spending on lobbying in 2020 rivaled the spending of individual Big Tech firms like Facebook and Google. The Markup searched lobbying disclosures in the U.S. Senate’s database and the watchdog Center for Responsive Politics’ tool for the names of companies that registered as data brokers in or . Those states are the only two that require companies to annually disclose that they collect, sell, or share people’s personal information without having a direct relationship to them. Lobbying Disclosure Act OpenSecrets Vermont California All in all, we found 25 companies whose combined spending on federal lobbying totaled $29 million in 2020. Many of the top spenders were not pure data brokers but companies that nonetheless have massive data operations. Oracle, which has spent the past decade acquiring companies that collect data, spent the most by far, with disclosure documents showing $9,570,000 spent on federal lobbying. For comparison, of the Big Tech firms with heavy lobbying presences, Facebook spent $19,680,000, Amazon $18,725,000, and Google $8,850,000 in the same period, according to the Center for Responsive Politics. Public Citizen, a consumer advocacy group, found that Big Tech spent . $108 million collectively on lobbying in 2020 Oracle has its own data collection arm but has also built its portfolio by buying up companies like , Compendium, and Crosswise. The companies, which were acquired in 2012, 2013, and 2016, respectively, take data from a variety of sources. DataRaker gets data from millions of smart meters and sensors for utilities companies, while Compendium delivers targeted ads. Crosswise allows Oracle to track people across devices, claiming to process data from billions of devices every month. DataRaker Oracle also acquired , in 2014, which connected offline purchases to online profiles. Additionally, Oracle combines datasets from more than 75 other data brokers, which it calls “the world’s largest collection of third-party data.” Datalogix Our report comes as the data broker industry is not only growing but also facing scrutiny for the first time. California recently passed a statewide privacy law that establishes an agency focused on regulating data privacy issues. Virginia and Maine have also passed regulations to protect people’s online information. California’s and Virginia’s privacy laws hit at the core of what data brokers do by requiring companies to delete data collected about people upon request and allowing people to prevent their personal information from being used for targeted advertising. Maine’s privacy law prevents internet service providers from sharing personal information with data brokers until people give “express, affirmative consent.” And the past year also saw concerns raised about how well these brokers protect their massive troves of data—Oracle, for instance, after security researchers found were left exposed on a server. suffered a data breach billions of records from its BlueKai data collection The Markup contacted all 25 companies for comment on their lobbying activities. Several companies, like Inmar Intelligence and LiveRamp denied being data brokers, though they had self-identified as such to California and/or Vermont regulators. “Inmar Intelligence does not generally consider itself a data broker though one of our entities, Inmar-OIQ, LLC, is registered as such in a couple of states,” Holly Pavlika, a corporate marketing senior vice president for Inmar Intelligence, said. The industry itself is hard to define—many companies, including tech giants, make money off of personal data, though the technical details of how they use that data vary. So The Markup relied on companies that self-reported to and as members of the industry. The list, 480 companies long, shows just how pervasive it has become for companies to traffic in personal information. Vermont California The list includes businesses that are primarily data brokers, like CoreLogic, which claims to have collected data on 99 percent of property and homeowners in the U.S. and spent $215,000 on lobbying in 2020; and Acxiom, which spent $360,000 on lobbying in 2020 for issues related to data security and privacy. Acxiom’s InfoBase boasts of datasets on more than 250 million people and collects information including location data, purchases, interests, life events, and behaviors, according to its marketing material. Our list also includes credit monitoring services like TransUnion, Equifax, and Experian, which each spent about $1.4 million lobbying in 2020 on issues related to credit score reform, like the of 2020, the Credit Access and Inclusion Act of 2019, and the of 2019. TransUnion also owns subsidiaries like Callcredit, Iovation, and Signal, which has been used for collecting and profiling users for gambling apps, . Protecting Your Credit Score Act Clarity in Credit Score Formation Act according to The New York Times Notably, our tally also does not include lobbying by trade associations that data brokers are a part of, such as the Interactive Advertising Bureau and the Digital Advertising Alliance. Of the data brokers who spent the most on lobbying, Oracle far outspent its peers Oracle, the biggest spender, used lobbyists to advocate on issues like annual defense and intelligence budgets and “competition and antitrust in the mobile telecommunications and digital advertising industries,” according to . public records The company didn’t respond to requests for comment. The second largest spender was Accenture—a technology and consulting company that boasts a marketing and analytics branch called . According to its , the company can combine datasets from sources including people’s purchase history and location to help its clients build customer profiles for advertising. Accenture Interactive marketing materials In 2020, the company spent $3,250,000 lobbying on issues like artificial intelligence, the , and COVID-19 contact tracing, according to public records. Digital Dollar When The Markup reached out to Accenture, the company pointed us to two statements CEO Julie Sweet made on data privacy legislation in 2018. The statements called for a federal privacy law that would preempt state laws. PricewaterhouseCoopers, a major accounting firm, spent $2,820,000 lobbying the federal government in 2020, third most among registered data brokers. The total amount 25 data brokers spent on federal lobbying in 2020. $29M PWC collects data from advertising networks and data analytics partners including personal information like names and addresses, which it uses to help clients personalize advertising campaigns, . according to the company’s privacy statement Spending went to monitoring privacy legislation and accounting and auditing issues. PWC didn’t respond to a request for comment. Other big spenders included Deloitte, which was awarded a $106 million contract to build a development environment for the Joint Artificial Intelligence Center this summer. The company spent $2,400,000 on federal lobbying in 2020, including on bills like the and the . Department of Defense Artificial Intelligence in Government Act Artificial Intelligence Initiative Act Deloitte is an auditing and advisory company but provides services like , which uses information including people’s hobbies, interests, and financial data to generate a health risk prediction score and help life insurance companies figure out how likely people are to buy their products. PredictRisk provided by data brokers And RELX, which spent $2,375,000 on federal lobbying in 2020, including on data privacy bills like the , the , and the ; and on bills related to data privacy and COVID-19, including the and the . Data Accountability and Trust Act Information Transparency & Personal Data Control Act Data Broker Accountability and Transparency Act Covid-19 Consumer Data Protection Act Exposure Notification Privacy Act RELX is a major data broker that owns companies like LexisNexis and ThreatMetrix, which has customers in law enforcement, insurance, and financial services. ThreatMetrix alone boasts tracking on 4.5 billion devices, from RELX in 2018. according to a statement Deloitte and RELX didn’t respond to a request for comment. In addition, many other data brokers spent lower amounts on lobbying efforts An Industry Increasingly Under Fire Some of the bills targeted by lobbyists would have regulated the data broker industry, though disclosures do not specify whether they lobbied for or against the bills. The Data Accountability and Trust Act looked to establish security standards and require postbreach audits for data brokers and also prohibit collecting information under false pretenses. The Information Transparency & Personal Data Control Act would have required data brokers to get consent to collect sensitive data and go through an annual privacy audit. The Data Broker Accountability and Transparency Act of 2020 wanted to mandate opt-outs from data brokers and for the FTC to create a national list of data brokers. While lobbying records don’t always list specific bills, filings show that RELX paid lobbyists to address all three bills. Deloitte records show that some of its lobbying efforts went toward addressing the Data Accountability and Trust Act. Both COVID-19 data privacy bills looked to provide stronger controls over data related to contact tracing. None of the bills named in this story passed Congress, except for the National Defense Authorization Act, which was enacted. Some of the companies that showed up in lobbying records faced other sorts of pressure during 2020. For instance, Apple and Google from their app stores last December following a series of reports from on how the location data broker was providing information to contractors who passed that information to the U.S. military. X-Mode spent $30,000 on lobbying in the last quarter of 2020, during the height of the public pressure. banned X-Mode’s trackers Motherboard Both and , another location data broker, which spent $160,000 on lobbying, are facing scrutiny from Congress over their location data sales. Sen. Ron Wyden, a Democrat from Oregon who signed on to , called the data broker industry “out of control,” in a statement to The Markup. X-Mode Venntel a letter calling for an investigation on Venntel “Americans are learning more every day about the secretive and shady data broker industry and they’re demanding new laws to protect our privacy,” Wyden said in the email to The Markup. “It’s no surprise data brokers are trying every avenue they can think of to ward off the common sense protections Americans desperately need.” -Sen. Ron Wyden X-Mode and Venntel didn’t respond to a request for comment. Some companies said they were lobbying to have a voice on legislation, including potential federal laws on privacy. “With increased momentum, attention and legislative activity in multiple states, we support a federal data privacy law that can rebalance the system and set standards that rebuild trust with the people providing the data—consumers,” Christine Travis, a senior communications director for LiveRamp, said in an email. “A federal approach to data privacy and security is better than a patchwork of state laws for all stakeholders.” When somebody shows up on the lobbying records, it just tells me they see a real threat to their business model. -Hayley Tsukayama, Electronic Frontier Foundation LiveRamp was the top lobbying spender among companies whose for advertising purposes. The company, which connects people’s activity across the web and from offline purchases for advertisers, spent $630,000 on lobbying in 2020 on issues such as the and Privacy Shield. primary focus is collecting data Augmenting Compatibility and Competition by Enabling Service Switching (ACCESS) Act Privacy Shield was a framework for transferring data between the European Union and the U.S. but was declared . invalid last July for failing to meet the EU’s privacy regulations Travis denied that LiveRamp is a data broker despite being listed on , insisting instead that the company is a “data connectivity platform.” California’s data broker registry Experts said scrutiny and oversight is fairly new to the data broker industry. “When somebody shows up on the lobbying records, or in meetings or in trade associations, it just tells me they’ve probably recently woken up to this issue and they see a real threat to their business model by privacy regulations,” Hayley Tsukayama, a legislative activist for the Electronic Frontier Foundation, said. Others, however, said Congress is not doing nearly enough to regulate the unwieldy industry. “What is widely understood now in Congress and among independent agencies like the FTC are the ways in which large tech companies like Facebook and Google are extracting data from consumers and using that to monopolize markets like the advertising industry,” Jane Chung, a Big Tech accountability advocate at the consumer advocacy group Public Citizen, said. “What a lot of people don’t understand is how data brokers fit into that ecosystem.” Originally published as " “ with the Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license The Little-Known Data Broker Industry Is Spending Big Bucks LobbyingCongress

Who are the Data Brokers Spending Big Bucks to Lobby Congress?

In early March 2021, we announced our $5.2M seed round with Accel for - our open-source data integration platform; 2 months later, we announced our $26M Series-A round led by Benchmark. Airbyte In this article, we want to tell you about what happened behind the scenes, including the deck we presented to the Benchmark team. We hope this will give you some insights about the fundraising process, even though our case might be atypical. If you like what we do, don’t hesitate to subscribe to our or star ! newsletter our GitHub project If you're mostly interested in the deck, here's a . Slideshare link to the Series-A deck The Timeline Some additional context: We actually signed the term sheet for our seed round with Accel in December 2020, but we just closed the round (after the due diligence and all the negotiations on the long form contract) in early February. It then took us about a month to prepare the , which happened on March 2, 2021. announcement for the seed The First Call Benchmark contacted us at the end of February, and we had our first call with them on March 10. Although our seed round wasn’t public information, they already knew about it, as most VCs did. Private news travels fast in the VC world. We witnessed that again during the Series-A, as we got contacted by a lot of VCs after we signed the term sheet, even though we kept it confidential. One hypothesis is that the information leaks during the We’ll come back to that later. vetting process. How did the first contact happen? Chetan Puttagunta and Peter Fenton both reached out to Michel and John, via different mutual connections. Having both of them try to reach out to us was a strong signal about how serious they were about talking with us. Leading up to the Partnership Meeting After the first call on March 10, we had several video calls over the next month. Each one ranging from an hour to several hours. These long form conversations allowed us to get to know Benchmark and, at the same time, allowed Benchmark to get to know us as founders as well as our plans for Airbyte. We knew that we were going to take on a board member with the Series A and it’s critical to have clarity before starting the 10+ year journey together. We felt that this discussion had to happen over several weeks. We presented to the Benchmark partnership on April 12. That’s when we presented the deck below. We sent it the day before so they could prepare their questions and so we could leverage that hour the best. After that, we had calls on April 13 and 14 to negotiate the term sheet, and we got it signed on April 15. During that time, we didn’t interact with any VCs other than Accel. We didn’t need to raise money. And Accel was a great sounding board and was putting our company first. Vetting Process While you can be sure the VC will contact people you have worked with in the past, you want to talk to companies have invested in as well. We try to get feedback from companies in various states of progress: hugely successful, successful but in a slow growth mode, and some that didn’t work out. The goal is to obtain a well-rounded view of the VC’s support and behavior in all scenarios. We made all these calls between the 7th and 14th of April. they It might be possible the information leak happens at this stage. After all, the CEOs that you’re talking to are well connected with other VCs. The Term Sheet A piece of advice here: Try to get a maximum amount of information about the terms of the round written in the term sheet. Once you have signed the term sheet, most of the negotiation you will do will be with the VC’s lawyers rather than the VC partner. So, if you get it written in the term sheet, it’s one less thing that is open to negotiation. And don’t accept a one-page term sheet. Try to get the full set of terms clearly written. This is the only way to know as entrepreneurs that there will not be an egregious term in the final documents. Benchmark clearly spelled out the terms of their financing in their term sheet, which made the closing process smooth and on schedule. Post-Term Sheet As founders, you have two jobs to do after the term sheet is signed: Negotiate the details of the long-form investment agreement with the VC’s lawyers. That’s when having great lawyers helps, as they can make sure to represent your best interests. Complete the remainder of the round. You might have pro-rata rights to abide by for your earlier investor, and you might want to involve some high-profile business angels who can bring a ton of value, too. In the end, we got about a dozen more investors in the round, including Shay Banon (co-founder & CEO of Elastic), Dev Ittycheria (CEO of MongoDB), Auren Hoffman (co-founder of LiveRamp and Safegraph), and SV Angel. Why We Decided to Raise Even Though We Didn’t Need the Money Entrepreneurs often think in terms of valuation and the equity they own in the company. But there is a third factor that is the most important of all: the which goes from 0 (bankruptcy) to 1 (IPO or huge exit). success factor, success factor * ownership * valuation = value you own We did YC in January-March 2020 because we thought it would have a significant impact on our success factor. We chose to raise with Accel (we didn’t need that money then as we still had a full year of runway) because we thought they were the ideal partner to bring us closer to the 1 for our success factor. This is what happened again with the Benchmark team. We were not convinced at all about raising a Series-A now at the start, but each interaction with Chetan and the Benchmark team was bringing immense value and extremely relevant advice. After the third call, we could no longer see Airbyte without them working closely with us to bring us closer to the 1. Now that we knew whom we wanted to partner with, it was all about getting terms (valuation, etc.) that worked for both us and Benchmark. And that would be our advice in fundraising: Optimize first for whom you want to partner with. And then, optimize the terms of the fundraising (valuation included) for that partner. Our Deck As the was public, we couldn’t just reuse it. And a lot of progress had been made since we signed the seed term sheet in December, and we wanted to reflect that. Airbyte seed deck So here’s the Series-A deck we presented to the Benchmark partnership team: https://www.slideshare.net/jeanlaf/airbyte-seriesa-deck Our deck was pretty standard, though some might consider it a bit short. That’s why we include the appendices along with the deck. The Structure and Our Pitch on Each Slide Here are more details on our slides - why we included them and what we were saying during our pitches. 0. Cover Although you might only spend 10 seconds discussing the cover slide, it is important. It’s about showing that you have identified your positioning. In our case, we identify EL(T) as . So, essentially, we want to identify ourselves as the future of the ETL industry here. the future of the ETL industry 1. Industry Context This is the only slide that is common to our seed deck. Thanks to Snowflake’s IPO and Segment’s acquisition, we knew that data infrastructure was hot from an investor point of view. We decided to start with some context on the industry and how we fit into this ecosystem. What we were saying: When you look at the data infrastructure industry, there is often a new category that emerges through a commercial product. Once the market matures, an open-source alternative gets created and ends up taking over the category. This behavior is often seen because data infrastructure requires privacy, security and scale - which cloud-based solutions can’t offer as well as open-sourced ones. There are many examples, such as Kafka, Spark, and now dbt. We want to be the open-source solution for data integration. You might wonder why an open-source approach would also win the format for data integration; sometimes a closed-source cloud-based approach works. This last sentence is a transition to the next slide. 2. Problem We’re Solving This is a rephrasing of our seed round slide with more details. What we were saying: In June and July 2020, we started reaching out to 250 of Fivetran’s, StitchData’s and Matillion’s customers. We ultimately managed to talk to 45 of them. We wanted to know whether an open-source approach would make sense to address data integration. What we learned is that a cloud-based closed-source solution will never be able to fully address the data integration problem. It has several inherent issues. 100% of the companies we talked to were using Fivetran, StitchData or other solutions, while also building and maintaining their own connectors. They did so because either (a) the ETL solution didn’t support the connector they wanted, or (b) the solution supported it, but not in the way they needed. When you look at Fivetran, for instance, you’ll see that after 8 years, they only support 150 connectors. The hard part about ETL/ELT is not about building the connectors, but maintaining them. It is costly, and any cloud-based closed-source solution will be restricted by a ROI (return on investment) consideration. It isn’t profitable for them to support the long tail of connectors, so they only focus on the most popular integrations. During those 45 interactions, we also identified a third issue. Some of the companies were about to stop using Fivetran for some connectors because it was becoming too pricey. The value of an ELT solution is about replacing a paid data engineer who builds and maintains a connector in-house. The amount of work required from an engineer is about the same, whether the volume of data being moved is low or high. So with volume-based pricing, at some point it just stops making sense to use an external solution. And the last inherent issue with a cloud-based approach: Although cloud data warehouses are winning the enterprise market, it is because they are considered part of the data infrastructure. All other solutions must go through a rigorous privacy compliance process that will take several months. 3. Our Solution That’s the slide where we start to get deeper into our differences from our seed deck. What we were saying: There are two main differences with our approach. 1. Being open-source: Enables us to address the long tail of integrations: We built a (CDK) that enables our users to build new connectors in a matter of hours (instead of days!) for our community. The idea is to standardize the way connectors are being built, thus making them much easier to maintain by us and the community. Connector Development Kit Enables our users to customize any connector to their specific needs, and to leverage Airbyte the way they need to, along with other tools in their data stack. Enables our users to start testing Airbyte without asking upper management for permission. So the speed to value in a testing use case is way better than any closed-source solution. Enables us to address the use case of all companies that can’t use a third-party vendor to handle their sensitive data. This includes healthcare, fintech and insurance, as well as enterprises. 2. Current solutions have adopted a -based pricing, which has always been counterproductive for data. Having a value-based pricing. volume We won’t detail much yet what our intentions are here, as this is something that we would like to keep under the hood. 4. Our Product What we were saying: Airbyte addresses two different audiences. With our , we want to empower , such as data analysts and scientists, to start replicating data in a no-code way. Airbyte essentially makes them autonomous; they are no longer dependent on the priorities of the data or software engineering team. It takes literally for a data analyst to replicate data from Salesforce to Snowflake, including the deployment through our Docker Compose. easy-to-use UI data consumers 2 minutes 5. Our Product [2] What we were saying: The second audience we’re addressing is With our and soon our , we enable them to leverage Airbyte to power the data replication they need in their software and workflow. Some will even leverage Airbyte to offer data integrations on their own platform to their customers. data and software engineers. API CLI Airbyte integrates with dbt, Airflow and Kubernetes for now. But our intention is to integrate with the whole stack by the end of the year. Our goal is that Airbyte just works - whatever your data volume, architecture, infrastructure, or connector needs - by the end of the year. 6. Our Team In this slide, we review the expertise of the team. In our case, we have six people coming from Liveramp, where they built and maintained more than 1,000 connectors and were moving more than 100TB of data every day. Our goal is to make investors understand that we’re experienced and have already done it in the past, but now the goal is to enable company to do it. every Since we presented these slides we’ve hired three more people: Natalie K. (Director of Operations), Jenny B. (Senior Software Engineer) and George C. (Senior Software Engineer). By the way, we are hiring! 7. Our Velocity This is possibly the most important slide, as it validates everything we’ve said with facts. Please keep in mind that these metrics are the ones we had on April 11, when we presented the deck to the Benchmark partnership team. The point we wanted to make is how far we’ve come since the seed round’s term sheet was signed. We deliberately didn’t choose the seed closing date, because Accel made the decision to invest in Airbyte in December, not in February. The number of companies using you is most likely the most important data point for investors. What we were saying: We actually started working on Airbyte at the end of July. We soft launched a MVP 2 months later (at the end of September), with only 6 connectors and no support for incremental replications. We wanted to have feedback as early as possible. Since then, we’ve been used by X companies to sync data, we have Y PR contributors and Z issue contributors, and we’ve ramped up the number of connectors to 60. 8. Some Growth Metrics Investors will always ask for the names of companies that are using you. So here, we’re just anticipating the question. Our point on this slide is to show that the conversion rate from users having synced data to users who are using us in prod every day is growing every week! 9. Our Roadmap Now that we’ve shown our velocity up to this point, it’s time to show what we can accomplish in the next few quarters. Investors want to see where you will be when it is time to raise the next round. That is this slide’s goal. 10. Our Ask This is the last slide of the pitch. The goal here is that you shouldn’t need to say much! It shouldn’t come as a surprise, and is a direct consequence of the roadmap slide. 11. Appendix! You need some sort of slide to separate the main deck from the appendix. No surprise here! We only showed the appendix slides if asked relevant questions, but we sent the deck with the appendix. 12. The Competitive Landscape Here are a few choices we made that distinguish Airbyte from other open-source solutions: What we were saying: So the velocity in terms of execution, team and community growth are not comparable. Airbyte’s connectors are usable out of the box through a UI and an API, with monitoring, scheduling and orchestration. It takes literally 2 minutes to start replicating data from Salesforce to Snowflake. Airbyte runs connectors as Docker containers, so they can be built in the language of your choice. Airbyte’s components are modular, and you can decide to use subsets of the features to better fit in with your data infrastructure (e.g., orchestration with Airflow or Kubernetes or Airbyte…) We integrate with dbt for the transformation piece and will soon let the community contribute normalization schemas for all connectors. Unlike Singer, Airbyte uses one single open-source repo to standardize and consolidate all developments from the community, leading to high quality connectors. We built a compatibility layer with Singer so that the Singer community can easily migrate their work to Airbyte. The point for us here was to demonstrate that we are well on our way to becoming the industry’s open-source standard. 13. Some Growth Metrics [1] Even though GitHub stars are essentially vanity metrics, they are an indicator of the awareness you create. And it adds more support in demonstrating growth to your investors. 14. Some Growth Metrics [2] These metrics are much more important to show, as our defensibility will come through the community. The exponential growth was what was exciting to investors. 15. Some Growth Metrics [3] We anticipated that Series-A investors might need more metrics than we included in the seed deck. We showed the growing usage per user. If we make our prod users successful, they will naturally put more and more of their connectors on our platform when moving away from other solutions. 16. Some Growth Metrics [4] Even though we never actually presented this slide, it was important for us to include it in the deck. As we sent the deck before the meeting, it was already answering the question, “Who is using you in prod?” 17. Our Investors This is probably the slide we showed the least often, as serious VCs will already know who your main investors are. But you should still have it ready just in case. That’s it, folks! Now you know why we included the slides we did, and what we were trying to accomplish with each one. Hope this helps you understand us better, and maybe helps you prepare for your own investor encounters. Let us know what you think or if you have any questions in the comments below! Previously published at https://airbyte.io/blog/how-airbyte-raised-its-series-a-round-2-months-after-its-seed

Valuable Insights About The Fundraising Process

In May 2018, the most far-reaching legislation for the protection of personal data ever conceived took effect: The European Union's (GDPR). General Data Protection Regulation The law was created by the European Parliament and Council to unify privacy regulation within the European Union and provides individuals extensive control over their personal data. To accomplish this goal GDPR can impose severe penalties on violators. Companies not complying with the provisions of the law, face fines of up to 4% on worldwide turnover or EURO 20 million, whichever is higher. The following month Governor Brown signed , now known as the California Consumer Privacy Act of 2018, which grants Californians similar rights with respect to the collection of their personal information. Assembly Bill 375 A lot has been written about how GDPR might affect not only European businesses but also the operations of non-EU companies and international corporations. For the following analysis, we will assume that – outside of notable exceptions - the law will be enforced within and outside the EU and will govern how US-based companies treat personal data. Personal Data Personal data is generally understood as any information relating to an identifiable individual; an identifiable person is one who can be identified, directly or indirectly, by reference to an identification number (i.e. social security number) or one or more factors specific to a person’s physical, physiological, mental, economic, cultural or social information. Usage of personal data by third parties pretending to be the rightful data owner is considered “identity theft”. As such GDPR protects the individual’s identity. GDPR prohibits the transfer of data outside of the European Economic Area to a third-party country unless the recipient country provides an “adequate level of data protection” Though the United States has worked extensively with the European Commission on data security standards, it is currently not considered to be an Adequate Jurisdiction by the Commission. Fiat Identity Like fiat currencies, which enables governments to declare which money is legal tender, “fiat identities” are tied to government decrees pronouncing which documents it considers to be legitimate for the verification of a person’s identity. In the United States, the government began issuing Social Security Numbers in 1936. To-date nearly 454 million different numbers have been assigned, according to the . Social Security Administration While the original objective was to track U.S. workers’ earnings to determine their Social Security benefits, government agencies and companies soon found many new use cases for the number, gradually morphing it to a “profiling system”. When someone is asked for his/her "last four digits of your social" the answer is literally the government assigned serial number. One of the more enlightened identity solutions thus far has successfully been deployed by the tiny European nation of Estonia. E-identity cards have already been issued to 98% of the countries citizens. The card is a cryptographically-secured digital identity powered by a blockchain-like infrastructure on the back-end and enables Estonians to access public services, financial services, medical and emergency services as well as to pay taxes online, e-vote, provide digital signatures, and travel within the EU without a passport. Profiling "Product YOU" The Internet, and to a greater extend the emergence of the world wide web, have made personal data accessible to an ever-increasing number of data brokers, including those offering . social engineering as a service (SEaaS) Simultaneously, applications build on the web led to an explosion of the type of personal data that can be observed and recorded - i.e. click-throughs on web pages - which led to the creation of online identities (sometimes referred to as 'digital identities'). While most European nations have identified and battled the dangers of leaving online identity management to for-profit corporations for some years, it took a nationwide scandal - now simply referred to as - to wake up U.S. citizens to the fact that their identities had been turned into products which are sold to advertisers and bad actors alike. Cambridge Analytica More troublesome, shockingly data brokers do not limit themselves to rent personal data to advertisers but trade this information among each other. Companies such as (LiveRamp Holdings, Inc.) openly promote their prowess to sell personal data with statements such as Acxiom “ LiveRamp IdentityLink is an identity resolution service that ties data back to real people and makes it possible to on-board that data for people-based marketing initiatives across digital channels.” The primary (commercial) nature of Google, Facebook, LinkedIn and many of the most popular web and mobile applications is that of an advertising marketplace. The "free service" of these companies are paid for in cash at a rate of more than $10 million each hour by marketing companies and loss of privacy by the user. Blockchain to the Rescue? Blockchains combine a number of features that make it desirable for the protection of sensitive data: decentralization, encryption, transparency and access controls. Many – and sometimes all – of these features are missing from most systems currently controlling personal data. GDPR requires companies to secure personal data via multiple levels of protection to ensure that data is neither lost, destroyed, or disclosed to unauthorized individuals. This principle is shared with a characteristic of blockchain technology referred to as pseudonymization. Consequently, a growing number of blockchain projects are attempting to tackle the topic of personal data and identity management, including Civic, SelfKey, Evernym, uPort and Shocard. The importance of blockchain-powered identity solutions has also caught the attention of traditional technology companies such as IBM which advertises its commitment “to creating a secure, blockchain-enabled decentralized identity for everyone on the internet”. Most often these are focused on improving processes around legal and regulatory burdens involving financial institutions – such as ‘know your customer’ (KYC) laws – using government-issued credentials. applications Others subscribe to the idea that consumers might be interested in being their own data broker and thus profit from selling their personal information. As shown above, current government systems and the treatment of personal data as commodities by commercial entities are at the center of the current problems surrounding identity management. As Franklin Foer in his book “World without mind” explains, Facebook and other social media applications owe their astonishing growth rates to the fact that they removed friction from the process of personal data exchange and defaulting to ‘opt out’ paradigms which required users to act to protect their data, while often signaling greater utility by unrestricted sharing of information. In the light of GDPR, all future applications will have to adhere to privacy by design, requiring users to opt into the use of data while adding usage transparency and anonymization to their processes. Preliminary Conclusions There can be little doubt that blockchains can be part of the solution to the current crisis in identity management. However, outside of the decentralization of data handling, any approach that does not address problems created by legacy systems is not likely to find widespread adaptation. After all, there is little value in protecting a 'digital identity' (profile) which has already been copied by data brokers and unauthorized users. Scalability is also a challenge for those companies that seek to enable users to profit from selling their personal information as data sets with low volumes have little or no commercial value for data brokers. A truly innovative solution is more likely to emerge from those groups addressing the needs of the almost one billion people on the planet without government-issued IDs. As seen with the adaption of mobile phones and later smartphones, users in the developing nations might leapfrog solutions encumbered by legacy technology debt. Finally, just like email and voice required dedicated protocols to provide end-to-end control mechanism (SMTP; VoIP), identity will require its own protocol to be owner controlled at last. Further reading: Blockchain & Sentiment

Blockchaining the Internet of Identity 

There are quite a few decks already available online, and we found them all useful, each in different ways. That’s why we thought the deck we used for our seed round could be helpful to some companies, especially those that are open source or developer tools. For context: on that seed deck, we started working on our , , in the end of July, and raised our seed round with Accel in December, so only 5 months in. We’re assuming the deck was acceptable, as we raised the seed after only 13 days in the fundraising process. open-source data integration platform airbyte.io We will write a dedicated article about the process within a few days, so this article is really about the deck itself. Our Deck Here’s the Slideshare . link Our deck was pretty standard, though some might consider it a bit short. That’s why we always sent the appendixes along with the deck after our meetings. Here are more details on our slides – why we included them and what we were saying during our pitches. 0. Cover Although you might only spend 10 seconds discussing the cover slide, it is important. It’s about showing that you have identified your positioning. 1. Industry Context Thanks to Snowflake’s IPO and Segment’s acquisition, we knew that data infrastructure was hot from an investor point of view. We decided to start with some context on the industry and how we fit into this ecosystem. What we were saying: When you look at the data infrastructure industry, there is often a new category that emerges through a commercial product. Once the market matures, an open-source alternative gets created and ends up taking over the category. This behavior is often seen because data infrastructure requires privacy, security and scale, which cloud-based solutions can’t offer as well as open-sourced ones. There are many examples, such as Kafka, Spark, and now DBT. We want to be the open-source solution for data integration. You might wonder why an open-source approach would also win the format for data integration; sometimes a closed-source cloud-based approach works. This last sentence is a transition to the next slide. 2. Problem We’re Solving What we were saying: In June and July, we started reaching out to 250 of Fivetran’s, StitchData’s and Matillion’s customers. We ultimately managed to talk to 45 of them. We wanted to know whether an open-source approach would make sense to address data integration. What we learned is that a cloud-based closed-source solution will never be able to fully address the data integration problem. It has several inherent issues. 100% of the companies we talked to were using Fivetran, StitchData or other solutions, while also building and maintaining their own connectors. They did so because either (a) the ETL solution didn’t support the connector they wanted, or (b) the solution supported it, but not in the way they needed. When you look at Fivetran, for instance, you’ll see that after 8 years, they only support 150 connectors. The hard part about ETL/ELT is not about building the connectors, but maintaining them. It is costly, and any cloud-based closed-source solution will be restricted by a ROI (return on investment) consideration. It isn’t profitable for them to support the long tail of connectors, so they only focus on the most popular integrations. During those 45 interactions, we also identified a third issue. Some of the companies were about to stop using Fivetran for some connectors because it started to become too pricey. The value of an ELT solution is about replacing a paid data engineer that builds and maintains a connector in-house. The amount of work required from an engineer is almost the same whether a low volume or a high volume of data is being moved. So with volume-based pricing, at some point it just stops making sense to use an external solution. And the last inherent issue with a cloud-based approach: although cloud data warehouses are winning the enterprise market, it is because they are considered part of the data infrastructure. All other solutions must go through a rigorous privacy compliance process that will take several months. 3. Our Solution What we were saying: We’re building an open-source ELT platform that syncs data from SaaS apps, APIs and databases to data warehouses, data lakes and other databases. Our solution can fully integrate with your data infrastructure and stack, if you are using Kubernetes or Airflow for orchestration, or DBT for transformation. Our goal is to become the open-source standard for anything ELT by the end of 2021. 4. How it addresses the problem What we were saying: By making it trivial to build and maintain connectors using Airbyte rather than doing it in-house, we will become the new standard for building connectors. This will help us support the long tail of connectors. And since connectors run as Docker containers, you can build them in the language of your choice. As connectors are open-sourced, any team can edit a pre-built connector and tune it to their needs. If a connector breaks, anyone can jump on the code, submit a PR and, once approved, the change can be propagated across all the existing users of that connector. Airbyte focuses first on a self-hosted offer;  pricing will be based on the feature and number of connectors used. The pricing will not be indexed on the volume of data. Being open sourced enables us to have a bottom-up approach with a frictionless adoption from data teams, without going through privacy compliance, as we handle data security as a first-class citizen. 5. Who We’re Targeting / Our Go-To-Market What we were saying: With Airbyte, we wanted to address 2 different audiences: Data consumers, including data analysts and scientists.Data engineers who were building and maintaining the connectors themselves, or managing the data infrastructure. What made Fivetran successful is that they enabled data consumers to leverage the data without the help of data engineers – they made them autonomous (as much as Segment made product teams autonomous). Airbyte provides a UI that makes it very easy to start replicating data for non-technical users. It takes literally for a data analyst to replicate data from Salesforce to Snowflake, including the deployment through our Docker Compose. 2 minutes In addition, Airbyte will offer data integration connectors through an API that data engineers can leverage to build their own workflow and applications. This is also a way for Airbyte to address SaaS businesses that want to offer their own integrations to their customers through Airbyte. 6. Our Team In this slide, we review the expertise of the team. In our case, we have 4 people coming from Liveramp, where they built and maintained more than 1,000 connectors and were moving more than 100TB of data every day. Our goal is to make investors understand that we’ve already done it in the past, but now the goal is to enable every company to do it. 7. Our Velocity This is possibly the most important slide, as it validates everything you’ve said before with facts. The number of companies using you is most likely the most important data point for investors. What we were saying: We actually started working on Airbyte in the end of July. We soft launched a MVP 2 months later (at the end of September), with only 6 connectors. We wanted to have feedback as early as possible. It’s been 6-7 weeks since we soft launched, and we are now used by X companies, we have Y contributors, and we’ve ramped up the number of connectors to 43. 8. Our Roadmap Now that we’ve shown our velocity up to this point, it’s time to show what we can accomplish in the next few quarters. Investors want to see where you will be when it is time to raise the next round. This slide’s goal is to accomplish this. 9. Our Ask This is the last slide of the pitch. The goal here is that you shouldn’t need to say much! It shouldn’t come as a surprise, and is a direct consequence of the roadmap slide. 10. Appendix! You need such a slide to separate the main deck from the appendix. No surprise here! We only showed the appendix slides if asked relevant questions. 11. Some growth metrics [1] The point for us here was to insist that we hadn’t done a hard launch yet, only a soft launch on our social networks on 09/24. The rest of the growth is organic. 12. Some growth metrics [2] Investors will always ask for the names of companies that are using you. So you should always have such a slide ready. The more recognizable the names of the companies are, the better. 13. The competitive landscape What we were saying: Here are a few choices we made that distinguish Airbyte from other open-source solutions: Airbyte’s connectors are usable out of the box through a UI and an API, with monitoring, scheduling and orchestration. Airbyte runs connectors as Docker containers, so they can be built in the language of your choice.Airbyte’s components are modular, and you can decide to use subsets of the features to better fit in with your data infrastructure (e.g., orchestration with Airflow or K8s or Airbyte’s…)We intend to integrate with DBT for the transformation piece soon and let the community contribute normalization schemas for all connectors. Unlike Singer, Airbyte uses one single open-source repo to standardize and consolidate all developments from the community, leading to higher quality connectors. We built a compatibility layer with Singer so that Singer taps can run within Airbyte. 14. Our investors This is probably the slide we showed the least often, as serious VCs will already know who your main investors are. But you should still have it ready just in case. That’s it, folks! Also published here .

All 14 Slides from the Deck We Used to Raise Seed with Accel in 13 Days

The art of building a large catalog of connectors is thinking in onion layers. We’re building an open-source data integration platform at . We launched our MVP about a month ago. We were thrilled by the amount of feedback and support we got from the community. We even got our first big pull request from a contributor this week (2,000+ lines of code). But during this full month, we didn’t release any new connectors. You might wonder why we didn’t build on that momentum. If people were excited with our MVP even though it had only 6 connectors, you might think we should have ramped up on the number of connectors as fast as possible. We didn’t do that for two very important and differentiating reasons. Airbyte First, we were defining exactly what the best data protocol would be if we wanted to solve data integration once and for all, and this for companies. You can learn more about our specification . Even though it’s not final yet, you will have a glimpse of our vision for the future. all here Second, and just as important, we were building a real manufacturing plant for data integration connectors. See, our team led data integration at Liveramp, which has more than 1,000 data ingestion connectors and 1,000+ distribution connectors. So we have the experience of abstracting what can be abstracted and simplifying the manufacturing of new integration (very often without code). We haven’t fully built our manufacturing plant, but engineers can already add one new connector every day. This article describes how we built this connector manufacturing plant. What you need to think about when building a large number of connectors When building a large catalog of connectors, there are several things that you need to think through. Initial build This is when you start from a blank page. This step usually requires a little bit of planning since it involves communication with external teams/companies. The initial build step involves: Access to the source/destination documentation Access to test accounts, test infrastructure, etc. Using golden path encoding good practices Using the best language for the task: today, we support both Java and Python, but anyone can add their own languageC reating documentation Defining the necessary inputs Tests Tests are essential to make sure that any code or protocol change won’t affect the connectors. They need to run before every merge. They also ensure that the connector behaves as you expect. For that you need to run your connector against the actual production service. For example, if you’re working on the Salesforce connector, you must make sure that Salesforce actually behaves the way you expect. It is not unusual that an API or service documentation doesn’t fully reflect the reality. We currently have the foundation of our test framework; it allows developers to focus solely on providing inputs and outputs, and the rest is taken care of by the framework. These tests give us 90% certainty that the connector is fully functional. If there are edge cases, it is always possible to add more custom tests. Liveliness & Change detection It is essential to ensure that the source or destination continues to behave as it was encoded  during the initial build phase and to ensure that the source or destination is still alive for monitoring purposes. These verifications must be run at a cadence, and any failure needs to be investigated and fixed, leading to the maintenance phase. Maintenance We need to define how we are going to update the connector, push changes and propagate the changes to all the running instances of Airbyte. The art of building connectors is thinking in onion layers Segmenting cattle code To make a parallel with the pet/cattle concept that is well known in DevOps/Infrastructure, a connector is cattle code, and you want to spend as little time on it as possible. Anything you can do to prevent yourself from doing work in the future, you need to do. This will accelerate your production tremendously. Abstractions as onion layers Maximizing high-leverage work leads you to build your architecture with an onion-esque structure: The center defines the lowest level of the API. Implementing a connector at that level requires a lot of engineering time. But, it is your escape hatch for very complex connectors where you need a lot of control. Then, you build new layers of abstraction that help tackle families of connectors very quickly. Today, we’ve built one of these abstractions to support existing Singer integration. Building an integration leveraging Singer takes us less than 3 hours, and our goal is to bring it down to less than 10 minutes. We have the same ambition for every other family of sources and destinations. As we continue to improve our manufacturing plant for connectors, we will build tools that will allow us to handle 95% of integrations with no or very little code. This is how we are going to address the long tail of integrations and how we’re going to make integrations a commodity. What Airbyte has built up to now We’ve built the following: The center of the onion The golden path in Java & Python to build new connectors The first version of the integration test framework Connectors:10 sources with a rate of 1 new source per day, and 4 destinations A layer to quickly support Singer integrations What our ambitions are with this connector manufacturing plant We want to reach a rate of 5 connectors per day and accelerate even beyond that. We also want to provide the community with more tools to build and contribute their own connectors. Ideally, 95% of connectors can be added to Airbyte with no code. We hope this gives you a better understanding of what we’ve been up to and what our real ambitions are. If you see any ways to improve this architecture, we’re all ears. Don’t hesitate to join our to discuss any questions or suggestions with the team. Slack Previously published at https://airbyte.io/articles/data-engineering-thoughts/how-to-build-thousands-of-connectors/

Stories, News & Mentions about LIVERAMP

Liveramp’s Stories on HackerNoon

Liveramp's News & Mentions