Measuring Traffic Quality in Mobile Applications: Standards and Best Practices

In today's world, mobile devices have become indispensable companions, connecting us to mass media, entertainment, and essential information. They transformed the way we communicate, access information, and engage with the world around us. According to Statista, as of January 2023, approximately 311 million people in the United States accessed the internet, with a significant portion using mobile phones. In 2022, the country saw an estimated 85% mobile internet user penetration. A 2022 survey revealed that nearly 50% of U.S. women and 46% of men considered mobile internet access important anytime and anywhere, while another 45% could not imagine their everyday life without the internet.

This surge in mobile device usage has profoundly impacted the advertising landscape. The global mobile advertising market  was valued at USD 117.97 billion in 2021 and is projected to reach USD 144.08 billion in 2022. It is expected to reach a staggering USD 621.63 billion by 2029, demonstrating a CAGR of 23.2% during the forecast period. The rapid adoption of mobile campaigns and the integration of VR & AR with mobile advertising platforms are anticipated to drive industry growth. The mobile segment has also fueled the expansion of programmatic advertising, with over 80% of all advertising being purchased programmatically.

While mobile app advertising has been around for quite some time, it presents specific risks and limitations for advertisers seeking to measure the quality of ad placements. One primary concern is fraudulent traffic, which can inflate performance metrics and waste advertising budgets. Fraudsters employ tactics, such as fake installations, click spamming, and bot traffic, to deceive advertisers and exploit their advertising dollars.

Another challenge is the accuracy of location settings. Inaccurate location data can lead to misdirected ad placements and reduced return on investment (ROI) for advertisers. This can happen for several reasons, including spoofed GPS coordinates, low-quality location data providers, and users intentionally manipulating their location settings.

Legal and financial risks are also important to consider. With data privacy regulations such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA), advertisers must be cautious about collecting, storing, and processing user data. Non-compliance with these regulations can lead to hefty fines and damage an advertiser's reputation.

Furthermore, ad placements in mobile applications can also be affected by ad blockers, which prevent ads from being displayed on users' devices. This can significantly impact an advertiser's reach and campaign effectiveness, leading to lower engagement rates and reduced ROI.

Lastly, brand safety risks pose another challenge for mobile app advertising. Inappropriate or controversial content in mobile apps can negatively impact an advertiser's image and customer perception, potentially damaging the brand's reputation.

Indeed, these risks and challenges and a lack of understanding of ad placement quality in mobile applications can deter advertisers from fully utilising this channel. This limitation can result in missed opportunities to reach and engage with a vast audience relying on mobile devices for daily needs.

In this article, I aim to enhance your understanding of the complexities of measuring traffic quality in mobile applications. By sharing insights, strategies, and real-world experiences, I aim to equip readers with the information they need to tackle challenges, make informed decisions, and unlock the full potential of mobile app advertising.

Main mobile apps traffic quality metrics 

Viewability

Simply displaying an advertisement does not guarantee that users actually saw it. While a hypothetical reach of 20 million may sound impressive, it is crucial to determine how many users actually viewed the ad. This is where the viewability metric comes into play.

Viewability is a digital advertising metric that quantifies the number of times users have seen an ad. It is expressed as a percentage and calculated using the following formula: the number of viewable impressions divided by the total number of impressions multiplied by 100%.

According to the Interactive Advertising Bureau (IAB) standards, an impression is considered "viewable" if:

• For banner advertising, 50% of the banner is visible on the screen for at least 1 second.
• For a video, 50% of the player is visible on the screen for at least 2 seconds.

Viewability is essential for evaluating ad placements, as it helps advertisers choose platforms that effectively display ads to users. Both extremely low and near-perfect viewability rates can indicate potential issues.

Viewability should be considered when designing creatives. Advertisers should expect users to refrain from engaging in ads for extended periods. Each slide of a banner or video's opening seconds should communicate the advertised brand and product. If users cannot identify the brand, they are unlikely to remember it, leading to a diminished media impact.

VTR

View-through rate (VTR) is a metric that reflects the appeal of advertisements on digital platforms. It considers both the ad format and the message it conveys to users. The measurement is based on the number of ad impressions and views.To determine VTR, a special tracking code can be placed on the web page to collect information about the traffic for both the source and promoted resources. 

When evaluating the VTR for a source website, consider the following data:

• Placement plan
• Presence of infographics
• Number of ad displays and clicks on the banner (can be grouped by characteristics, such as publication method, materials, and days)
• CTR (click-through rate) metric
• User data (survey results, contact information, IP addresses, etc.)

To evaluate the VTR of a promoted resource, hosting logs provided by specialised software can be used instead of tracking codes.The following parameters are considered when evaluating VTR for a promoted resource:

Impressions: Represents the number of times the ad was displayed to users. Users notice the banner but have yet to take any action.

Views: Users pay attention to the content of the banner, become interested, and move their cursor over a specific element of the ad (photo, link, video, etc.). A view is recorded after the cursor hovering over the element for more than 2 seconds. This is possible when the advertisement contains an engaging call-to-action or a captivating headline.

Clicks: Users click the link and visit the advertiser's promoted resource.

VTR is calculated using a simple formula: the ratio of the number of views to the number of ad impressions multiplied by 100%. The resulting metric demonstrates the effectiveness and appeal of a specific advertisement, indicating whether adjustments are necessary to enhance its performance.

CTR

Click-Through Rate (CTR) is a critical performance indicator. It is used to evaluate an advertisement's effectiveness and reflects the extent of audience engagement and the number of users motivated to take action.Though CTR is a quantitative metric, qualitative factors form its foundation. The following criteria can influence CTR:

Traffic quality. CTR may be more significant if it reaches the key segment of your target audience. Users are more likely to take action. Click alone does not guarantee a conversion. The landing page's effectiveness will determine the outcome.

Ad quality. The ad's visual aspect plays an essential role in influencing CTR. Additionally, the size and placement of the advertisement on a user's page can affect click-through rates.

Relevance to the topic. The context in which the ad is displayed matters dramatically. For example, placing an advertisement for a beauty salon on a machinery website would likely result in a low CTR.

To calculate CTR, divide the total number of clicks by the total number of impressions, then multiply by 100%.

Understanding that a "good" CTR varies depending on the advertising platform is essential. For instance, consider advertising on YouTube. Pop-up notifications may distract users from watching videos, resulting in fewer clicks and a lower benchmark for a "good" CTR on this platform. Conversely, if relevant ads are displayed on search engine results pages (SERPs), users may find them more helpful, leading to higher expectations and a higher CTR threshold.

IVT

Invalid traffic (IVT) is a technical term referring to ad impressions generated by bots or any form of non-human traffic. Although often confused with ad fraud, not all invalid traffic is malicious. Nonetheless, detecting all forms of IVT is essential for safeguarding ad spend and ensuring that impressions are valid and visible.

To assist advertisers in protecting against IVT, the Media Rating Council (MRC) introduced IVT accreditation in 2015. This accreditation validates the effectiveness of IVT measurement providers for desktop, mobile, web, video, and in-app ads. Two mutually exclusive categories exist: General Invalid Traffic (GIVT) and Sophisticated Invalid Traffic (SIVT).

GIVT is detected through routine list-based filtering or other standard parameter checks. It includes:

• Known data centre traffic
• Bots, spiders, and other crawlers
• Action-based filtering
• Non-browser user agent headers or unknown browsers
• Prefetch or browser-prerendered traffic (unless it counts as a gross impression)

SIVT refers to hard-to-detect scenarios that require advanced analytics, multi-point confirmation/coordination, or substantial human intervention for analysis and detection. Examples include:

• Bots and crawlers posing as legitimate users
• Hijacked devices and user sessions
• Invalid proxy traffic
• Adware and malware
• Reward-driven measurement manipulation
• Falsely presented websites and advertisements
• Cookie spoofing
• Location data manipulation or falsification

Google, for instance, offers specific recommendations. Publishers should first analyse how users interact with the app, ensuring ads do not obstruct functionality or provoke accidental clicks. Google warns that violating this guideline may result in a publisher's account being disabled.

Standards for Measuring Traffic Quality

MRC: the Media Rating Council

The Media Rating Council (MRC) is a non-profit organisation aiming to ensure audience measurement services' validity, reliability, and effectiveness. The MRC establishes and maintains measurement standards and accredits measurement services that adhere to those standards. The MRC works closely with the advertising industry, including advertisers, agencies, publishers, and media research organisations, to promote transparency and quality in audience measurement. Its primary focus is on media rating research, including various media forms, such as television, radio, print, digital, and out-of-home advertising.

In recent years, the MRC has expanded its focus to address new challenges in digital advertising, such as viewability, invalid traffic (IVT), and ad fraud. It developed standards and accreditation programs to help protect advertisers and maintain the integrity of the digital advertising ecosystem.

IAB: Interactive Advertising Bureau

The Interactive Advertising Bureau (IAB) is a trade association representing the digital advertising and marketing industries. The IAB serves as a central resource for its members, which include media companies, publishers, advertisers, technology firms, and marketing agencies.

The IAB's primary mission is to develop and promote industry standards, best practices, and guidelines that foster growth and innovation in the digital advertising space. It achieves this by conducting research, providing thought leadership, and offering training and certification programs. By doing so, the IAB helps ensure that digital advertising remains an effective and efficient means of reaching consumers in a rapidly evolving media landscape.

Some of the key focus areas for the IAB include:

Digital Advertising Standards: The IAB works to create and maintain technical and creative standards for digital ad formats, ensuring consistency and interoperability across different platforms and devices.

Ad Measurement and Metrics: The IAB collaborates with organisations like MRC to develop guidelines for measuring ad performance, viewability, and audience engagement.

Privacy and Data Security: The IAB addresses consumer privacy and data protection concerns by developing guidelines and advocating for responsible data practices within the industry.

The Open Measurement Software Development Kit (OM SDK) is a standardised solution introduced by the IAB Tech Lab in 2017 to address the challenges and standardise traffic measurements in mobile applications. Its primary goal is to streamline the measurement process, reduce discrepancies between measurement providers, and increase transparency and trust by standardising the measuring ad viewability in mobile apps.

By integrating the OM SDK, publishers and SSP platforms enable advertisers to measure ad impressions using codes from all independent verifiers. To work with the OM SDK, verifiers must undergo mandatory IAB certification, a procedure completed by international verifiers such as Adloox, DoubleVerify, IAS, MOAT, and others.

Before the introduction of OM SDK, advertisers and publishers often had to implement multiple SDKs from different measurement providers, which led to increased complexity and potential conflicts between different tools. The OM SDK simplifies this process by providing a single SDK that can be integrated into an app.

Assessing Ad Visibility Metrics

App environment: Unlike websites, mobile applications run in a closed background, making accessing and measuring ad performance more difficult. Using SDKs (Software Development Kits) like the OM SDK helps overcome this limitation by providing a standardised way to measure viewability across various apps.

Variety of ad formats: Mobile apps offer a wide range of ad formats, including native ads, interstitials, rewarded video ads, and playable ads. Each format has unique characteristics and viewability standards, making measurement more complex.
Example: A rewarded video ad requires the user to watch the entire video to receive an in-app reward. In this case, viewability would depend on the video being on-screen and the completion of the video.

Device fragmentation: The mobile ecosystem is diverse, with multiple operating systems, screen sizes, and device specifications. This fragmentation can affect how ads are displayed and interacted with, influencing viewability metrics.

Example: An ad might be fully viewable on a high-resolution device but partially cut off on a smaller screen. This would impact viewability measurements across different devices.

User behaviour: Mobile users often interact with content differently than desktop users, with shorter attention spans and quicker scrolling habits. This can impact ad viewability, as users may scroll past ads before they can fully load or be viewed.

Connection quality: Mobile users can experience varying internet speeds and connectivity levels, impacting ad loading times and viewability. Slow-loading ads may not have the opportunity to be fully viewed before the user scrolls past or leaves the app.

Fraudulent Traffic

While many users have become experts at handling and identifying invalid traffic in their statistics, advertisers must know more about the various forms of fraudulent traffic within mobile applications.

Device and platform diversity: The mobile ecosystem is fragmented, with multiple operating systems, device types, and app stores. This fragmentation can create opportunities for fraudsters to exploit vulnerabilities in specific platforms or devices.

Example: Fraudsters may target less secure app stores to distribute their malicious apps, leading to increased fraud traffic from these sources.

In-app ad fraud: In mobile apps, there is a higher prevalence of in-app ad fraud, where fraudsters create fake apps or manipulate existing ones to generate fraudulent ad impressions or clicks.

Example: A fraudster may create a fake app with the sole purpose of generating ad impressions. Unsuspecting users download the app, and the fraudster profits from the illegitimate ad revenue.

SDK spoofing: In mobile environments, fraudsters can engage in SDK spoofing, which involves mimicking the SDK of a legitimate app to generate fake ad impressions or clicks. This type of fraud is challenging to detect, as it appears to come from a genuine source.

Example: A fraudster spoofs the SDK of a popular app, causing the advertiser to believe that their ads are being shown in the legitimate app, while in reality, they are being displayed in a fraudulent environment.

Click injection and click spamming: Fraudsters can manipulate mobile-specific features, such as device identifiers and app install tracking, to engage in click injection or click spamming. These tactics involve generating fake clicks or attributing organic installs to fraudulent sources.

Example: In click injection, a fraudster monitors device broadcasts to inject fake clicks right before an app install is completed. This makes it appear that the fraudulent source drove the installation, resulting in illegitimate attribution and payouts.

Mobile botnets: Fraudsters can use mobile botnets, which involve a network of compromised devices controlled by a single entity, to generate fraudulent ad impressions, clicks, or installs.

Example: A fraudster may compromise many devices through malware and use them to generate fake ad impressions, making it appear that real users are viewing the ads.

Brand Safety Measurement

Brand Safety is a policy that protects a brand's performance and reputation from negative, questionable, or inappropriate content and its environment when advertising in the digital space. Compliance with Brand Safety when promoting brands in mobile applications is also an essential factor in the quality of communication with users.

The main factors for compliance with the Brand Safety criteria are:

Store Moderation: The basic measure of compliance with the Brand Safety criteria is the passage of store moderation by mobile applications (Google Play and App Store). Moderation occurs continuously and ensures compliance with store policies throughout the application's life. Thus, the in-app environment is characterised by the predictability of content and its surroundings.

Sound On: The programmatic ecosystem involves the transfer of important parameters between process participants, including the sound-on parameter on mobile devices. Selecting apps for advertising only on devices with the sound turned on guarantees audiovisual reproduction of advertising in mobile applications.

Filtration System: This system auto-checks titles, descriptions, and comments according to word masks, targeting by themes and age restrictions and blocklisting applications with stop words based on masks.

In addition to these main factors, brand safety measurement in mobile advertising can include app content and category filtering, keyword and sentiment analysis, real-time monitoring and blocking, allowlists and blocklists, and third-party verification.

Hitting the target audience

Precisely reaching a brand's target audience based on users' socio-demographic characteristics and interests is a fundamental requirement for ensuring advertising campaigns’ effectiveness.

Device and Platform Targeting: Advertisers can target specific devices, operating systems, and app environments to reach the most relevant audience segments. For instance, a brand might want to target only iOS users to promote an iPhone-specific app or product.

Geo-Targeting: Mobile advertising allows advertisers to target users based on their geographic location. Real-time location data helps brands deliver contextually relevant messages, such as promoting a nearby store or event. For example, a restaurant chain may use geo-targeting to show ads to users within a certain radius of their locations, encouraging them to visit.

Behavioural Targeting: Advertisers can target users based on their in-app behaviour, interests, and past actions. By analysing app usage patterns, brands can identify and target users more likely to engage with their ads. For example, a fitness app may target users who frequently use other health and wellness apps.

Lookalike Audiences: Ad platforms can create lookalike audiences by analysing the characteristics of a brand's existing customers and finding users with similar traits. This helps advertisers expand their reach and target potential customers who are likely to be interested in their products or services.

In response to Apple's requirements restricting data transmission to mobile audiences, such as the App Tracking Transparency (ATT) framework introduced with iOS 14.5, ad platforms had to adapt their targeting and measurement strategies. Apple's changes limit the use of the Identifier for Advertisers (IDFA) for ad tracking and targeting purposes unless users explicitly grant permission. As a result, platforms and advertisers have been exploring alternative solutions, such as:

Contextual Targeting: Focusing on the app's context or content where the ad is being served rather than user-level data. This approach helps deliver relevant ads based on the environment rather than individual user behaviour.

SKAdNetwork: Apple introduced the SKAdNetwork, a privacy-focused framework that allows advertisers to measure the success of their campaigns without accessing user-level data. The network provides aggregated and anonymised ad performance data, helping advertisers optimise campaigns while maintaining user privacy.

First-Party Data: Brands increasingly focus on collecting and leveraging first-party data (data they collect directly from their users) to create personalised and targeted ad experiences. This data can inform targeting strategies and maintain ad relevance without IDFA.

Probabilistic Attribution: Some platforms and advertisers are exploring probabilistic attribution methods that use aggregated, anonymised data to estimate the likelihood that a specific ad led to a conversion or other desired action. This approach helps maintain attribution and campaign optimisation without relying on individual user data.

Conclusion

In-app advertising presents significant opportunities for advertisers to effectively reach and engage their target audience. 

However, advertisers must stay updated with industry standards and best practices to maximise the success of in-app advertising campaigns, such as the OM SDK technology developed by IAB. This facilitates a transparent and fair evaluation of inventory quality and ensures that campaigns are optimised for in-app environments.

Maintaining brand safety and reaching the target audience effectively requires a multifaceted approach that combines careful app selection, age targeting, and interest-based settings. By considering the specific requirements and nuances of in-app advertising, advertisers can create highly effective campaigns that resonate with their intended audience and deliver impactful results.