When operating within paid search, one of the greatest benefits we have is our data. We can slice and dice our performance by target, term, location, device, message, time, and on and on. And when we find ourselves incapable of accessing this magnitude of information, it can be a bit of a shock.

One Google report that I love to reference is the This vs Other ad extension report. It can be found as a segment in your Ad Extensions tab, and shows you not only how many impressions, clicks, and so on your specific extension receives, but also how many times something other than the extension received these things.

As the Bing-Google rivalry continues, Bing has earned another point for its team by offering a similar report:

The Ad Extension Details report

It offers the following:

• Segmented data for sitelinks, location extensions, and call extensions
• Clicks, impressions, cost
• Conversions, revenue and ROAS details
• Device type
• Top vs Other positioning
• Performance at ad group & campaign level

This means that you are able to segment the success of your ad extensions for a clearer picture of best performers.

What does this mean?

You are effectively able to run trials of ad extensions, such as sitelinks, and judge them by their true data. You can also identify areas of inefficient spend by noting those receiving many clicks but minimal conversions.

How might I do this?

As you might expect, the process begins in the BingAds Reports tab:

Upon selecting this report, you’ll have the usual questions about what all you’d like to include in your report.

Suggestions: Ad extensions details, of course, but also spend and revenue volume, Top vs. Other details, and Device performance. All 3 of these sets of data will give you clear ideas of what is performing best in your account and what needs reshaping.

The resulting report will be a thing of beauty. Depending on the size of your account and how specific your ad extensions are, you might need to run this report on just a campaign level, or perhaps segment certain aspects of the report once it’s downloaded.

In the example below, we have all of our Branded campaign sitelinks collected:

How to I clean up my data?

From this, I create a pivot table overlapping all the performance associated with these similar campaigns, broken down by device.

Because I’m able to specify which sitelinks I want to appear to my mobile devices, I always evaluate my ad extensions by device. Don’t make the mistake of over-generalizing your audience!

When I create the pivot table, I’m able to see my top performing sitelinks. Here we compared off-line savings, product-specific inventory, and a general New Arrivals sitelink.

In the able above, the New Women’s Arrivals sitelink received the highest direct CTR, with a solid ROAS as well. The Roxy sitelink had a standout response once visitors arrived on site, with the highest conversion rate and ROAS.

*Author’s note: Both the aforementioned sitelinks had nearly the exact same number of sales, which allows for confidence in using such metrics as the Conv Rate and ROAS.

What to do with this information?

Optimize! Test! Test & Optimize!

Truly, you can do so much with this data. My first step would be to seek out clear winners and losers. In the set of data above, there are clear top performers. The question to ask yourself is “What made this sitelink special?”

• Did you phrase your information differently?
• Was it inventory or price specific?
• Are there are other similar pages we could also incorporate into this type of sitelink?

Another place to focus is upon cost inefficiency. A sitelink may have a strong CTR, but is it resulting in the on-site behavior you’d like to see, such as converting?

Just like keywords, device targeting, and ad performance, ad extensions deserve a threshold for cost or CPL in order to keep them performing at their best.

A third suggestion for further insights is to compare your ad extensions performance between publishers Google & Bing.

Just as we reviewed the top Bing ad extensions, let’s take a quick look at the final outcome for some similar links in Google:

Interesting, right? The outcomes are actually quite different from what we saw in Bing over the same time period. Note: Those sitelinks in grey have low conversion data, and aren’t in the running for Top Performing.

Google has better conversion performance from Sperry than any other sitelink, and a strong ROAS as well. The Women’s New Arrivals sitelink has the second strongest engagement but a poor conversion rate. Do Google visitors use the sitelink in their browsing phases, but not when they’re prepared to buy?

An example test of this theory would be to assign specific top performing product sitelinks to general search terms. If the search term is “womens Brooks running shoes” is generating a Brooks-specific ad, you may want to draw your shoppers in with broader sitelinks, such as “New Brooks Arrivals,” “Men’s Best Sellers,” “Women’s Best Sellers,” and “Final Sale” – all of which continue to escort your Brooks-minded customers to the category that best suits them, without selecting their style for them.

Pro tip: Callout extensions are a great place to utilize your top styles. While your sitelinks may be a bit broader to these shoppers, your callouts may notify them that you carry their favorite style.

Whatever way you slice it, more data can give you more insight to the successes and pitfalls of your paid search accounts. The ability (and common practices urging you) to segment sitelink activity is a nice clean step in increasing account efficiency and improving your testing.