Bottom Sponsored Ads on Google – How effective are they?

Last week, Google officially rolled out sponsored ads at the bottom of the SERP (after testing for a couple of months). They claimed that in testing, these ads performed better than side sponsored ads because they fall directly in the flow of the searcher.

Google also pointed out that bottom sponsored ads would be mutually exclusive from side sponsored ads – that is, a searcher will only see one or the other, not both.

What this suggests to me, then, is that Google has figured out which searches are more likely to have searchers clicking through to a second page of results and is including bottom sponsored ads for those searches. For searches where it’s unlikely that the searcher will click through to a second page, they will still see side sponsored results.

But how effective will the bottom sponsored ads really be?

We ran a quick eye tracking study with 12 searchers performing 5 different search tasks (for a total of 60 searches). Each of these search tasks returned bottom sponsored results, so presumably Google has decided that searchers are more likely to interact with those ads at the bottom as opposed to on the side. The tasks were randomize to account for order effects.

The five searches were:

Take a look at the heatmaps below. Each heatmap shows the number of fixations (a “count” heatmap – leave a comment if you want to know more about how heatmaps are generated), with red being parts of the page that receive the most fixations.

As you can see, the Google Golden Triangle is still alive and well. Gaze behaviour was pretty typical of what we see when we get participants to do search tasks. On the 1280×1024 resolution display that the participants used, there was very little scrolling below the fold.

Oh, and the bottom sponsored ads? One participant looked at them in one search (“are flu shots effective?”). That’s 1.7% of searches in which searches “engaged” with the bottom sponsored ads. Typically we see anywhere from 20% to 50% engagement with side sponsored ads, depending on the search task.

(Aside: it’s important to note that looking at something does not necessarily equal attention, and that you can pay attention to something without looking at it. So when I say “engagement”, I mean strictly that participants look at the sponsored ads. Whether or not they pay attention to them is something we have to examine through recall metrics and branding questionnaires.)

Google Bottom Ad

“are flu shots effective?”

Google Bottom Ad

“grocery flyers toronto”

Google Bottom Ad

“ps3 game reviews”

Google Bottom Ad

“remembrance day toronto”

Google Bottom Ad

“thai restaurants toronto”

Does that mean they won’t work? No. Google doesn’t make changes like this without doing their own extensive testing, and they have the benefit of big sample sizes and actual click through data (as you can probably guess, there were no clicks on the bottom sponsored ads in our little study). And we only look at the first exposure/click – we might expect there to be more engagement with the bottom sponsored ads as searchers bounce back to the SERP for the second or third time.

This also comes shortly after Google announced that they are expanding the local results that will be shown in the right rail (you can see this in the “thai restaurants toronto” search). This increase means that side sponsored results just wouldn’t work over there for the 20% of searches that return local results.

It’s unfortunate that Google will not be providing data in AdWords to distinguish between performance on the side versus the bottom. Instead, they’ll be lumping both of those into an “Other” category, so you’ll see “Top vs Other” in your AdWords reporting.

Bottom sponsored ads are also a bit at odds with Google’s testing of infinite scrolling. We’ll have to wait and see how that pans out.

How do you feel about bottom sponsored ads? Let us know by leaving a comment!