As a publisher-first company, we constantly develop, test, and implement new products and features that can maximize the revenue per session. It’s not just us. Every publisher will hit the wall after implementing header bidding.
Think about it for a minute. You are selling the premium placements via direct deals and PMPs and you have header bidding in place to sell the remaining ad impressions at the best possible CPMs.
What else can be done? Of course, you can increase the number of ads on the page, but it’ll likely affect the user experience. So, publishers end up trying to maximize revenue per session. And, more often than not, ‘ad refresh’ saves the day.
Well, does it?
Active Exposure Time:
If you use a time trigger (refresh every 60s), the ads are being refreshed without considering many other factors. Even when the user isn’t active on the page, ads will keep on refreshing after a minute. This means the likelihood of a user engaging with the ad is minimal. So, the value of your refreshed impressions (CPMs) will decrease over time.
Sidenote: You can read more on ad refresh triggers here.
For example, your CPM for the first impressions can be $2 and the CPMs for the subsequent impressions will cascade down as long as the user stays on the page. From $2, you may end up with $.90 after the third or fourth refresh.
That’s not all. Your viewability might take a hit as well.
“Anecdotally, there is a correlation between bad auto-refresh practices and poor viewability; it’s a common pattern. For a site that has a low average for viewability, [say] under 10%, typically the reason is poor auto-refresh activities.”
– Ian Trider, Director, Centro (Digiday).
So, when ad refresh has become the de facto method for improving revenue/session, we noticed what’s wrong with time-based refresh and came up with Active Exposure Time (AXT). To put it simply, AXT is an advanced ad refresh product that recreates viewable impressions by tracking the ad viewability and user activity on the page.
Only when an active user views an ad for a period of X seconds, then the ad will be recreated — i.e., a new ad will be served. In other words, you, a publisher, can guarantee user attention to the advertiser.
If advertising is all about trading user attention, you are essentially doing it via AXT. You ensure that an active user (who’s engaging with the content on the page) looked at the ad for X seconds before replacing it with a different ad.
Frankly, that’s all a media buyer is looking for. What’s more? Your viewability and CPMs aren’t going to be impacted much.
In fact, we studied how AXT impacts CPM, ad viewability, and revenue over the three-month window. Results?
- Ad viewability increased substantially. 49% increase overall and all the publishers saw 20% to 50% improvement based on their average viewability.
- Ad revenue increased by up to 30% for some publishers. Note that this is additive revenue.
- CPMs didn’t go down. It stayed more or less the same. In some cases, we just saw a 10 to 20 percent drop in eCPMs.
But we don’t stop there. We optimize the setup further to maximize revenue via AXT. Let’s see how.
Note: Regardless of the type of ad refresh you do, you can replicate some of the strategies here.
We’ll start with the easiest one – sticky ads. Publishers have been running sticky ads as it stands to capture user attention and guarantees viewability. This means you can recreate more viewable impressions with AXT on sticky ads. Imagine a scenario where you have an ad unit that’s at the end of the article and the other one’s a sticky unit on the sidebar.
Apparently, the one at the sidebar tends to stay in-view for a longer period of time and you can refresh the ad units more than the other units on the page.
UX tip: Run sticky ads on sidebars for desktops and on the footer for mobile devices. These placements provide better reading experience for your users.
As a publisher, this isn’t surprising — users tend to differ from one another. If you are targeting a specific segment with your content, then they tend to be engaged more than the fly-by users. For instance, if you are specifically catering to U.S. audiences, users from Canada may spend less time on your site.
In such cases, we enable AXT for specific geographic regions. After all, users need to stay on the page for refresh to work well.
Target Engaging Content:
You can’t expect the readers to spend the same amount of time on every page. You might have a category where you publish long-form content and you might have categories that don’t retain users for long.
We identify such categories (and the layouts) and run AXT to capitalize on the time spent by the users. Depending on the site and its user behavior, we also adjust the time (according to the categories) that users need to stay active to recreate ads.
Dynamically load ads:
When we mentioned ‘long-form content’, you might have come up with the question – Can I introduce ad units based on the content depth? Yes, you can and we do it here.
Typically, publishers define ad spots and then render ads on those spots. But in case of the long articles, it is best to dynamically load ads based on scroll depth and content length.
Optimize Ad Placements:
On top of these, we let AXT to recreate valuable (and viewable) ad impressions automatically based on user activity on the page. We’ll end up finding the best placements on the page. For instance, if a particular ad unit is refreshed more than the others, then active users typically see the ad units more than the others.
A publisher can use the learnings to optimize the setup further.
As we always say, think long-term. Rather than focusing only on maximizing revenue per session, focus on the means to hit your target as well. Because if you make the necessary moves to deliver a better experience for users and value for advertisers, you’ll stay “lucrative” in the open marketplaces (ad exchanges).
So, what type of refresh are you experimenting with? Want to see AXT in action? Give us a shout.