Engagement is at the heart of editorial, advertising, and marketing strategy. Engagement drives revenue, and smart publishers know that increasing engagement is the best way to boost your audience—and your earnings.
We’ve talked to a lot of publishers over the last year, and it’s clear that although engagement and increasing engagement are often at the top of the priority list, publishers struggle to define engagement. Variously, we’ve seen engagement defined as site visits, clicks, pageviews, and dwell time, which means a) few publishers are on the same page when they talk about engagement, and b) there’s no industry standard for monetizing off of engagement. And, as a consequence, they struggle to take advantage of the engagement they do have.
We have a horse in this race, obviously. We’ve spent a lot of time talking about Signal, and how it helps publishers make more money by rewarding engagement. But we haven’t talked about how we define and measure Signal engagement. We’re here to set the record straight.
Signal Engagement Markers
Internally, we start our measurement of user engagement by looking at engagement markers, which are tracked via Signal. Engagement markers are singular metrics such as:
- Site visits
- Page views
- Page and site dwell time
- User actions (such as clicks, scrolls, etc.)
In sum, we track 45 engagement markers via Signal, including device-specific markers such as taps and scrolls. We’re confident that these markers cover every user interaction that occurs on a page. Individually, they give us (and our publishers) a metric that we can track over time, but this is still an incomplete view of any given user or audience segment. That’s because they don’t tell us anything about contextual or historical behavior. However, when these markers are combined and tracked, we can start to build an in-depth image of a user, an audience, or even a piece of content.
If engagement markers are the brush strokes, engagement is the completed portrait. Engagement is when a reader (or visitor, or viewer) is engaged, that means they’re occupied with your content. They’re involved, they’re interested, they’re interacting.
When we say “engagement,” we’re referring specifically to engaged time, rather than to any individual engagement marker (such as a click) or metric (such as a click-through rate). And we define engaged time as the amount of time that a reader is occupied with your content, based on tracking user actions—engagement markers—over time.
A definition of engagement necessitates a definition of disengagement—and that’s also dependent on engaged time. User engagement markers keep our engaged time counter running, and we start a simultaneous disengagement counter after each input. After 5 seconds without an engagement marker, the user is no longer engaged. We’ve shown this chart before, but it bears sharing again:
Let’s use a cooking website as an example. A reader navigates to a recipe, reading more of it (and scrolling or moving the mouse) as they progress. Over the course of their total dwell time (the total amount of time spent on a page before bouncing), a certain percentage of that time is “engaged time.” Every mouse or keyboard action is a marker that tells us the reader is engaged, and adds to the engaged time counter. If a reader opens a recipe, but then wanders away for an hour without interacting with the recipe any further, they’re not engaged with the content—even if it remains open on the screen.
We’re strict about our measurements because we know that clarity and specificity matter to our publishers, and to our demand partners as well. Using Signal data, publishers can create a more detailed image and analysis of any given user or audience segment—and by tracking the types of content that make for greater engagement, they can adjust strategy accordingly. Of course, they can also use that engagement data to reload or inject ads and show them to engaged readers.
Ultimately, publishers will always be responsible for identifying which Signal engagement markers matter most to them, whether it’s clicks, dwell time, or any other metric. However, we believe it’s important to give context to these markers, and to give publishers the data they need to create a more informed, more intelligent business strategy, and ultimately help them—and the demand partners that rely on their content—capitalize on the engagement they generate.