Earlier in my career, about 10–15 years ago, digital experiences (at that time, just webpages) had what was known as the “eight-second rule,” meaning an eight-second load-time max per webpage.
At the time, eight seconds was about as fast as you could get. But today, the aspirational load time is more like one sub-second. Anything that takes longer is going to drive users elsewhere. Immediacy is the new normal.
Part of the reason for this shift is obvious: Technology is more advanced, and we now have the ability to reach more consumers on more devices than we previously could.
But the other part of this equation may be less readily-apparent.
Consumer expectations have evolved. Immediacy isn’t a nice to have. It is a must-have.
So, retailers, publishers, and brands must now provide content in real time.
This shift has largely been driven by social media. Today’s internet users are accustomed to opening their Facebook, Twitter, or Instagram feeds and seeing fresh, personalized content updated moment-to-moment. In light of this, decisioning engines now need to decide what to show a customer or reader in less than one second.
Real-time marketing is no longer on its way — it’s a fact of the digital age. To keep up, online publishers must assess readers’ preferences moment-to-moment and provide the most relevant content available. And they need to try to keep users on their own sites, rather than driving users off to third-party sites.
Here’s what online publishers need to know:
The other day I took my son to see the Marvel movie Venom. In the movie, a space alien arrives on Earth and can only survive by combining with a human host. The alien unites with a human and creates this amazing fusion: The human is only powerful with the alien’s power and the alien can only survive on Earth with the human.
Similarly, savvy tech leaders are partnering with machine learning technology to enhance customer experiences across a range of platforms. Real-time marketing is a departure from the status quo approach to personalization, which involves making suggestions based on a users’ declared preferences and/or on privacy-intrusive profiling.
For example, when you sign up for a media site, you’ll likely be asked what topics you’re interested in. (This might be politics, business, or pop culture, for example.)
But often, what people actually click on ends up being vastly different from their stated preferences. This means that publishers need to be aware of the disconnect and suggest content accordingly. This is the user’s observed preferences, which are perfect for a machine learning system to detect and make content decisions based upon.
Readers’ preferences change moment-to-moment.
This is unlike an online store where the user is typically going to buy something and may look at a related product. In the case of online media, the reader will be a target for a wide range of content. Picking up on various signals that the user gives off in the moment to determine their interests is critical, while also leveraging external factors such as stock trends, weather, news events, or trending articles can also help generate relevant recommendations.
The future of online publishing is all about understanding what readers want in the moment. If you don’t have a super-responsive site that can pick up on trends, you should consider implementing technology that picks up on readers’ individual preferences in real time. Otherwise, you’ll lose them.
If you’re able to capture a reader’s interest and suggest trending articles at the bottom or side of the page, that’s a solid start.
But any recommendation service that’s sending readers to a different site has several flaws.
First, while the publisher might perhaps make some ad revenue, they will be losing that user for at least a moment — maybe that day, or even forever. From a customer service standpoint, it’s always preferable to drive users to more content on your site as opposed to someone else’s.
On a recent trip to New York, I met with a major news media company who confided that they’d been losing website viewers since they can’t effectively respond to real-time trends across their entire site.
Here’s what happened: They got a scoop on a story and published it on their main site. As expected, the main site was getting a lot of clicks and was driving traffic from search engines to their site to find the article. But they still had a problem with getting that hot content recommended at all locations across their site to aid in recirculation.
Their existing system could not react in time to put the scoop on the other pages of their site. And you can’t capitalize on a trend if you’re not making sure it’s not just prevalent on your homepage but everywhere on your site. To do this, you need to have a way to make sure your site is fully dynamic and responsive.
If you aren’t, you’re missing out on a major business opportunity.
When you drive traffic to another site you have no control over the quality of that content. It could be good content, or it could be horrible. Either way, that experience will reflect on the brand of the referring site.
And if that content doesn’t align with your core values, you’re losing brand affinity. And that will hurt you in the long-term. When you leverage a real-time marketing solution on your site, you can showcase your brand’s ideals and values by aligning your brand with certain causes valued by your customer base.
Nike is a great example. This fall, Nike designed a campaign around Colin Kaepernick, the NFL player famous for refusing to kneel during the national anthem at games. Despite a major boycott, the campaign earned Nike $6 million dollars in additional revenue.
When you know what your readers are seeing, it’s easiest to deliver content you can vouch for. Anything else is a risk for your brand.
In the real-time age, publisher sites need to keep up with the customer expectation that the content is going to be just as relevant and immediate as what they are accustomed to on their social media feeds.
Pretty soon, readers won’t settle for less.
This piece was originally published on DCN.