Everyone is in agreement about the importance of extensions on your ads. In fact, they are so important that if you do not include extensions on your ad, Google will dynamically generate certain extensions for you so that you do not miss out.
The trouble with extensions comes in when attempting to analyse the impact they are having on your account, and this is simply due to how credit is attributed.
It is possible for a campaign to have 10 conversions, while your sitelink extensions may record anywhere from 0-60 conversions if you only look at the basic extensions reporting tab.
This view can be seen from the extensions tab:
Here we can see some discrepancies. For example, if you add the conversions recorded for the first two callouts, they already exceed the number of conversions attributed to all callouts. This can seem like an error until you realise that a conversion is attributed to all callouts that were shown on the ad, despite there being only one conversion from that click.
So for example, if there were three callouts shown with your ad, and the user converts after clicking that ad, then the campaign will have one conversion, and all callouts will have one conversion, but each callout that was shown will record a conversion against itself. There will be three conversions listed but only one counted in the total.
This occurs across all extensions that can show more than one of that extension type, and can take some getting used to.
It is for this reason that there are two other ways I recommend looking at extensions before you make any decision about whether your sitelinks are performing well! You can see these through segmentation:
Click type is a segmentation that can be found under the "segment" menu and can be applied on most tabs or views that you are looking at (eg. campaign, ad group, keyword, extension, etc.).
This segmentation will show you how people are engaging with the ad at the level you are looking at. To make this easier, let's consider that we are just looking at one ad. Click type will tell you the stats of the main ad (headline) as well as the stats for the various extension types (all grouped & de-duped). You will be able to see how many people clicked on your ad or on an extension, for example. It is worth noting here that only the "Headline" will match the impressions of the current view, since this is the only part that is guaranteed to be shown 100% of the time.
This last way to analyse your extensions is on the extensions tab, and uses a different segmentation: "This extension vs. Other":
This Extension vs. Other
This segmentation is most useful when looking at clickable extensions, as it will give you no additional information when looking at non-clickable extensions (such as callouts).
This final segmentation is also what you will use to see the specific performance related with that particular extension. For example, you are now able to see exactly how many people clicked on this particular sitelink and then converted, without having an aggregated view of the sitelink.
All three views are needed for a complete analysis of the extension you are interested in, as they all give you different information and ways of looking at the extension.
The “This Extension vs.Other” gives the actual extension statistics but does not take into account the residual benefit of simply listing the extension. You may find that a particular sitelink receives very few actual clicks on it, but when you then look at “click type” or the overview, you may find that having the extension is more beneficial than perhaps favouring a more clicked extension. The reasoning behind this is that it is possible that the extension conveys information or assurances that that the user appreciated at the beginning of the conversation.
There are many factors that need to be taken into consideration, and for this reason, analysis and testing of extensions tends to be a bit more complex than simple A/B testing of ads. Replacing an extension with another that has a higher CTR may actually decrease the CTR of your whole ad (same applies for any other KPI) and thus negatively affect your performance.
It is worth noting that there are a number of other segmentations (such as device, search partners, etc.) that can be applied to extensions just as with your campaigns and other groupings. Lastly, be sure not to compare mobile-specific extensions with desktop extensions.
Despite the added complexity of analysing extensions, they play an integral part of your account performance and need to be given fair consideration when looking at account improvement.
Please post any questions you have below and I'll do my best to answer them.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.