Tools & Tech

Reload, don’t refresh

Sovrn Publisher Advocate // May 1, 2019

blog signal sovrn smart ads reload adtech

Recently, we covered some of the well-deserved reasons that old ad refresh tools have a bad reputation. Today, we’ll explore a few of the ways Signal differs from these out-of-date products.

Signal is not an auto-refresh

One of the questions we get the most is “Doesn’t Google Ad Manager already refresh ads?”

Well, sort of.

Google’s Ad Manager (GAM) has a setting which enables auto-refresh of specific ad units. It’s a similar feature to other legacy refresh solutions in that the ad unit is reloaded based purely on a pre-set timer that starts on initial page load. Ad Manager doesn’t consider whether the ad is or has been in view, or whether or not the user is or was active.

This means that GAM creates new inventory without a minimum quality standard, which then harms viewability rate and viewable time. That hurts publishers in the long run, because poor viewability and engagement scores lead to buyers paying a low rate for your inventory—if they buy at all. Bad refresh technology can potentially even get your inventory banned from ad exchanges, as this type of unintelligent ad refresh can be considered ad fraud.

So yes, while GAM can refresh ad units, it’s a blunt instrument.

Broadly speaking, this is true for most refresh features. Most of the refresh systems built into ad servers or employed by websites are unintelligent legacy refresh products. This means that the impression, or in some cases the page, refreshes on a predetermined time (called refresh rate) from initial page or ad load, as opposed to taking into account ad viewability and user engagement.

Again, these tools focus purely on short-term benefits without considering long-term health. They’re built around boosting impression volume, which is itself a flawed metric. Impressions don’t consider or reflect real reader behavior, and Signal does.

Signal tracks real user engagement

Signal only recognizes that a user is active if they’re ‘focused’ on the tab. They will be deemed inactive if they:

  • click the URL bar
  • move to another tab in the browser
  • cover any portion of the browser window with another app
  • move to another screen (if using multiple screens)

Once the ‘focused’ condition has been met, Signal tracks 45 unique engagement triggers (such as mouse movement, keyboard movement, scrolling, etc.) to determine if a user is active. Here are a few of the events that Signal tracks:

  • Mouse movements (such as  mouse up, mouse down, mouse click)
  • Keyboard movements (such as key up, key down)
  • Focus & Blur events (such as tab changes)
  • Touch events (specific to mobile and tablet environments)

Signal sets the user status as ‘inactive’ if no engagement events occur for a period of pre-defined time, or once the disengagement threshold is met (for example, if a reader clicks into the URL tab, they’re immediately labeled ‘inactive’). This threshold determines the period of time a user can fail to produce any engagement events before they’re labelled inactive. This setting is customizable, but our standard is 5 seconds.

Here’s a chart that demonstrates how a user changes from an active to an inactive state via the disengagement threshold:

As long as the user is active, Signal is active as well. But unlike auto-refresh features, Signal constantly tracks how long two conditions have been met:

  • Viewability
    A given ad is ‘viewable’ by the user (51% of the ad is in view for 1+ second, as per MRC standards)
  • Engagement
    The user is active and engaged, based on Signal engagement tracking.

The Signal reload timer starts when both of the above conditions are met, and stops if either of the conditions is not met. The timer will start and stop to mirror user behavior. Because Signal is constantly tracking user engagement, you’re never mindlessly refreshing ad units.

/Signal goes beyond viewability and dwell time to deliver value to publishers that actively reflects the engaging content they create. While legacy ad refresh tools have been around for some time now, they don’t mirror the realities of online behavior, or of online publishing. In some cases, legacy tools can actively harm the publishers they were built to help. We built Signal to be intelligent, so that you can benefit in the short term with powerful boosts in revenue in impressions, while at the same time focusing on the long-term health and growth of your business.


Want to learn more?

Share this article