Why Sitelinks Are A PPC Worst Practice

Share with:

A few months ago, one of my coworkers asked me a thought-provoking question that I’ve been ruminating on ever since. She asked, “Does Google’s increased push on the use of sitelinks contradict their best practice to make ad groups as specific as possible and to drive users to the most relevant page? Say I’m advertising blue widgets. Long-standing best practices would be to have a very specific ad group pertaining only blue widgets and using my blue widgets page as the destination URL. Now instead of just being able to send them to my blue widgets page, I’m being pushed to include less relevant pages to keep my ad at the top of the page – Widgets, Widget History, Widget FAQs, etc. If your campaigns and ad groups are properly organized, sitelinks are only useful in limited circumstances.”

I thought this was an interesting perspective – one that I agree with. With our B2B clients, I usually don’t use sitelinks, for this very reason. The client has specific goals for each product or service, and we structure our campaigns and ad groups accordingly. The client doesn’t want us sending traffic to other pages within their website – these pages may not be optimized for conversion, or they distract the visitor from taking the action that the client really wants them to take.

Also, there are times when 6-10 or even 1-2 relevant links besides the landing page just don’t exist. Again, the client has a specific product or service they want us to promote. Maybe they even have a budget dedicated to that product or service. They not only don’t have other pages for us to send traffic to, but they don’t want us using their budget for that traffic!

This problem is more common for B2B advertisers, to be sure. I discussed it with Jeremy Brown in a post back in 2012.

This isn’t the first time I’ve covered the pitfalls of sitelinks. Back in 2011, I wrote a post for Search Engine Watch about the not-so-great aspects of sitelinks. While 3 years is an eternity in search, and Google has fixed most of the issues mentioned in that post, there are still shortfalls. Conversion tracking is still a challenge.

And Google doesn’t make it easy to see how individual sitelinks are performing. Take a look at this example:

sitelink data

At first glance, it looks as though the Contact Us sitelink has driven 8 conversions. Not so fast:

this vs other

In reality, no one is clicking on “Contact Us” – they’re all clicking on the ad itself.

This isn’t unusual, but to new PPC manager, or to clients looking at their own data, it’s misleading and confusing to say the least.

But I digress. I’m not the only one who thinks sitelinks might just be a worst practice. Andrew Goodman, in his famous rant “Why I Hate Sitelinks,” lists 11 reasons why he believes sitelinks are problematic. #1 on the list really resonates with me: “Where is the testing? Where are the key performance indicators (KPIs)? It’s impractical and/or irrelevant to test them; you can’t get actionable feedback.” Indeed.

I’m not totally anti-sitelinks. Sitelinks, and ad extensions in general, are a great way to take up more screen real estate. For advertisers with a robust catalog of related products and pages, sitelinks make a lot of sense. But they’re not for everyone, especially those with tightly-themed ad groups or those with only 1-2 relevant landing pages.

What do you think? Are sitelinks the love of your PPC life, or are they a worst practice? Share in the comments!

Related Posts:


  1. I believe that in a crowded landscape where competition for P1 to P3 to high well formulated site links are essential, reporting from adwords on the other hand is, I agree, rubbish and can cause problems. In my opinion they serve a purpose but they are also situational, not all clients need to have then and not for all ads, products or services.
    I also believe in testing site link, using them to supply offer messaging and using them to promote related content…

  2. Where I hate sitelinks is when I have to use them just to take up more real estate for the CTR boost (when the extra clicks don’t even come on the sitelinks – as you point out!)

    I love them where the client site is split into different customer segments that aren’t always reflected in search queries. Men/Women is a good example where site have a men’s section and a women’s section but queries rarely mention a gender.

    As you say, this is much more common in B2C but this is mainly because such segments are obvious. Similar things exist in B2B in most cases, but figuring out what they are is hard.

    • Melissa Mackey says

      Great points as always, Richard!

    • Good point Richard, there are some ways that you’re unable to segment keywords better as it’s not reflected in the search query. I had not thought about that.

      As I think back to sitelinks I’ve created the first thought is 1) Google likes these and then 2) how will this help searchers find what they need better? Mostly falls under #1 though since we create a limited number of landing pages, as Mel mentioned.

      This makes me think about sitelinks differently.

  3. I just want to say for the record, that the “This Extension” vs. “Other” thing drives me absolutely insane. 😀

    • Melissa Mackey says

      Couldn’t agree more, Susan! Why bother having it when the “This Extension” data is almost always tiny or zero? Crazy.

  4. The other thing using them can do, especially for lead generation, is dilute your conversion data. It isn’t practical to create a unique version of your lead gen form and thank you page to correspond to every campaign. If you want to use AdWords conversion tracking, you need a unique page to point to. Same if you want to track those user’s behaviors in Analytics. Can you try to extrapolate using segments and other data points, sure, but it is not nearly as clear if there is only one place a visitor can go that equals a conversion for that campaign.

    All the landing page science screams about the importance of having only one, super clear CTA. Sitelinks break that concept before you even get to the landing page in most cases. Although Richard Fergie’s example about men’s and women’s products is an example of a positive situation where you could push to a well designed landing page about a more specific product type.

    Has anyone seen data about the success (meaning actual conversions) for sitelinks that was not produced by Google? I know they generate clicks, but how do those clicks’ conversion rates stack up on average? Curious.

  5. “Now instead of just being able to send them to my blue widgets page, I’m being pushed to include less relevant pages to keep my ad at the top of the page – Widgets, Widget History, Widget FAQs, etc.”

    Not sure I agree here – if the user has decided to be one of the few who actually click on a sitelink, doesn’t that suggest the linked page is more relevant to them?

  6. Martyn Wright says

    You’re right in saying no-one clicks on the sitelinks. I rarely see more than a handful of clicks on them, but that isn’t to say they aren’t useful. The extra shelf space and messaging will boost CTR on the main ad even if they don’t click on the sitelink itself. I’ve managed to increase CTR (on brand terms, admittedly) from 15% to 30% by adding sitelinks, without any change in position. Less effect on non-brand campaigns, but still helps to a certain degree.

    • Melissa Mackey says

      They do boost CTR. Totally agree with that. My point is, what do you use for sitelinks when your client only has one relevant landing page? Tough challenge.

      • Get your client to build more relevant pages. Think about supporting features/benefits/offers etc and have a page for each with more detail and strong CTAs. The PPC account changes on a daily basis and the site should also be changing regularly to improve performance. Neither of these should ever be set in stone.

  7. Hmmmm… I’m not sure I agree. I can see some cases where sitelinks are not as useful as other accounts, but if you are in top positions for any account, you need to think about them and have really good reasons not to use them. Use them even if only for the CTR bump. If no one is clicking on them anyways, use them as extra razzle dazzle for your ad. (I know some people hate this, but advertising for better or worse is what we do…) Take up the real estate, promote trust indicators, engage users in your social media efforts if you don’t have great “other” conversion oriented landing pages to use.

    Reporting and optimizing is ridiculously limited, and third party mgmt platforms are doing a better job at presenting optimization data than Google (which is rare).

    Let’s face it. Search users are pretty lazy. They want Google and its advertisers to do the work for them. I would love if most of my traffic came in on highly specific kws and I have highly relevant ad copy and landing pages that speaks to exactly what they are searching for. Users often aren’t clear in their intent when typing in a query box. Sitelinks can also help qualify the click on your ad. “We do this, this and this – but not this other thing you are looking for where the terminology is close to our product/business.”

    I dunno, if I was a client and saw my ad on a page with other beefed up ads, I might question the strategy not to use them. You just need to be ready to have these conversations.

    • Melissa Mackey says

      All true, Lisa. Usually my conversation goes like this: “We need to optimize the rest of your website and/or create additional pages that are optimized for conversion to be able to use sitelinks. Not only do the links have to be relevant, they have to go to different pages, and we don’t have any different pages.” It’s a real dilemma for B2B in my experience.

  8. Another great use of sitelinks is to expand on the “why” with your ads. You’re only given 95 characters to hook someone and in many cases, why someone should be a customer is MUCH more important than what the product/service is. Sitelinks give you potentially four times as much room to explain why.

  9. Great article, Melissa. I think that what’s important for everyone to notice is that you’re not against sitelinks in general but in Google considering Sitelinks to always be a best practice.
    They can absolutely be useful in some cases but in some cases if your ad group is tightly themed they shouldn’t be necessary.

  10. Proper landing page optimisation and even an actual appraisal for our client’s goals have yet to be properly implemented at the agency I’m at, I don’t think I’m the right one to take it further but I’m glad to know now about some of the implications of sitelinks.


  1. […] Why Sitelinks Are a PPC Worst Practice: The title says it all. And I’ve got news for you….it’s right. Check out the post to get the full story and join the conversation. […]

  2. […] A very interesting article about Sitelinks and if they are good or bad practice […]

Speak Your Mind


This site uses Akismet to reduce spam. Learn how your comment data is processed.