Testing PPC Sitelinks with Some Surprising Results
Since first learning about PPC Sitelinks earlier this year, I’ve become a BIG fan.
There are so many ways that you can use sitelinks to improve the targeting of those broader search terms and get people to where they want to be a lot quicker. For me, using sitelinks over the past six months has driven twice as many clicks at a fraction of the cost, compared to just using one bog-standard link. And post click, the conversions have been a lot higher too.
But the question is: How do you decide what sublinks to use? There are so many approaches and options to choose from, yet Google will only display four sublinks at a time (granted you can load in up to 10 and over time Google will display the best performing). But what if you want to know up front which sublinks will drive the best results? The answer — as always — is to test.
For searches in the US, I wanted to find out which would do better: A product focus (using product categories as sublinks) vs. targeting intent (using learn/buy/use related sublinks). What better way to find out than through an A/B test.
Prior to the test, my thoughts were that targeting intent would trump a basic product approach hands down. Boy, was I wrong. Results showed that the product focus outperformed the intent focus on the front-end, and then went on to crush it on the back-end with:
- 5% more clicks
- 8% better engagement on the site (measured by a weighted quality score)
- 30% higher conversion
Now I’m not saying that these results ring true for everyone, not by any means. I guess my point is, there are theories out there as to what works and what doesn’t, and sometimes it can be dangerous to generalize.
I was convinced the intent approach was best, but simplicity won in the end. It just goes to show that nothing is certain until you test and tailor your approach specifically for your own business.
SEM Booty Blog 
Nice article regarding sitelinks. Since the inception of this feature I have been a huge fan. I’m wondering if you can clarify this sentence a little more:
“Results showed that the product focus outperformed the intent focus on the front-end, and then went on to crush it on the back-end with”
Specifically, can you provide examples of what type of text and sitelinks (and landing pages) that you used? No need to provide the actual text of the ad, but some examples might help to better explain the data.
Also, a technique I’ve tried with sitelinks is to create campaigns for the sole purpose of using sitelinks:
http://theppcblog.com/2010/08/utilizing-google-adwords-sitelinks-my-way/
Thanks for the sitelinks insight!
-Matt
Thanks for your comment, Matt. The example was from a B2B company who sell oscilloscopes (equipment used by engineers to test electronic designs). “Oscilloscope” is quite a broad category (there are many different kinds with many different price groups), similar to your “Men’s Sweatshirts” adgroup.
The product sitelinks were:
– Basic
– Bench
– Mixed Signal
– Performance
The intent sitelinks were:
– View Tutorials
– Arrange a Demo
– Get a Quote
– Download Manuals
The product group pages showed the different models related to that sitelink and included a combination of learn/buy/use call to actions. The intent landing pages were more specific to the intent. As an example, the tutorials page was very “Learn” focused, offering a selection of popular whitepapers and primers.
Also, thanks for sharing your work on the campaign vs. adgroup conundrum. I agree that this is definitely the way to go to improve relevance.
Imelda
I’m not sold, I’ve found customers wander around and don’t convert also feel that the site links take customers “off message” and distract from the intent.
Thanks for your feedback, Jenna. It’s interesting to hear that you have not seen good results from sitelinks. I think this is why it’s so good to test, as results can definitely vary from business to business.