Last year, we published the results of user tests which found that 41% of users were unaware of the distinction between paid ads and organic listings. 

Well, thanks to UX firm Bunnyfoot, we have an updated version of the test, which finds similar results.

This time, 36% of people tested still do not realise that Google Adwords are ads.

Furthermore, about a quarter of people don’t know that Google had any advertising at all. And this despite the yellow text box proclaiming 'ads'. 

Why this research?

The original research came from Bunnyfoot's work for a car insurance client who were investigating the effectiveness of Google Adwords.

During the research, the team found that 81% of users clicked on PPC ads rather than organic results. 

Further investigation found that 41 of the 100 individuals tested did not know that Adwords were paid-for adverts, instead believing them to be the most authoritative links.

Since that last research Google, as is its prerogative, has changed the way its presents ad listings, possibly as a result of EU anti-trust measures. 

This, in theory, should have made it easier for users to tell the difference between ads and organic results, though the issue is muddied by the fact that Google has removed the grey shading behind the ads. 

That said, the word 'ad' with a bright yellow background is a bit of a clue...

Current PPC ad format:

Old ad format:

So how has this change affected users' perception of search results pages?  

Research methodology

The user testing was carried out in multiple locations across London. 103 participants with a range of internet abilities were tested, all of whom used Google as their primary search engine.

The participants covered a wide range of demographics and were aged between 18 – 65 years.

An eyetracker was used throughout the sessions in order to record where the participants were looking, and this generated aggregated heatmaps (Bunnyfoot looked at other factors of search result understanding and only report a subset here).

All participants were then asked a series of post-test interview questions to gather further insight as to their understanding of Google’s results pages. 

The results:

  • When asked, 36% of users did not realise Google adwords were ads (a small change from 40% in 2012)

  • When asked, 27% of users did not realise that Google had any advertising. 

Note for the stats gurus amongst you: we of course realise that we have used a relatively small sample size (albeit a large one for user testing and eyetracking studies) and whilst the figures above have considerable sampling error it does not detract from their impact and surprise.

Have the changes to Google adwords made any difference?

The research suggests that the changes have made little difference to users' ability to distinguish between paid and organic results. 

Despite what I would assume was a clearer labelling of ads in the new formats, a significant portion of users still aren't seeing the difference. 

If I was a Google sceptic, I would suggest that the big G itself may have carried out similar tests to find the format that would satisfy the EU, yet still attract the most clicks. If clarity was the main factor, why remove the shading? 

It also highlights the propensity for web users to miss what might seem obvious to those designing and working on websites. 

The fact that 27% of those in the study when questioned did not realise that Google were doing any form of advertising in their results, further supports this, as well as being jaw-droppingly surprising in its own right.

What are the implications of this?

Pay-per-click is a very effective way of advertising your brand and reaching your target audience. It looks like about a third of people unknowingly click on ads and assume that ‘this is the best match’.

For the rest of those ‘in the know’ then the ads and the brands that pay for them still receive prominence, but the user can make an informed choice about whether to click a promoted link or not, depending on the context of the search.

Also, web designers and others 'in the know' should be wary of assuming ‘common’ knowledge on behalf of our customers.

According to Bunnyfoot CEO and co-founder Jon Dodd:

As our hundreds of user tests over the last decade have shown, it is very difficult to predict what customers’ knowledge or understanding is. When you do the tests, you are often humbled and surprised with how far off your assumptions are.

We conducted a follow up survey on this subject, asking 2,000 people whether they were aware of ads on Google

Graham Charlton

Published 8 May, 2014 by Graham Charlton

Graham Charlton is editor in chief at SaleCycle, and former editor at Econsultancy. Follow him on Twitter or connect via Linkedin.

2566 more posts from this author

You might be interested in

Comments (17)

Save or Cancel
dan barker

dan barker, E-Business Consultant at Dan Barker

Thanks for this, Graham. I'd love to see the actual wording of the questions if BunnyFoot are willing to share.


over 4 years ago

Graham Charlton

Graham Charlton, Editor in Chief at SaleCycle

Hi Dan - I'll ask Jon.

over 4 years ago


Jon Dodd, Managing director at Bunnyfoot Ltd

Hi Dan,

Its important that the research was performed in the context of task based testing - so we were not just asking people to look at a page and tell us what things were - but rather we gave them a relevant task, had them perform it, then probed for their understanding once they have performed it.

A summary of the protocol is provided below - from which we derived the results reported in this article. I hope this answers your query.

Task summary:
CONTEXT: Imagine that you have recently purchased this car (give details) and you’d like to get a quote for car insurance. Use Google as a start point to obtain a quote
START POINT: Google (directed to type in ‘Car insurance’ or ‘Car insurance quotes’)
DELIVER: Google results page
ACTION: If required prompt to select an appropriate result

Post-test Questions:
Open recall: Once they are in their selected site…
* why did you select that link?
* Please advise of which brands/companies you recall seeing from the Google search results?
* How do you think the results are sorted on google?
* Why are some listings higher, lower on the right side?
* What influences you to click on result?
* What discourages you to click on result?
* Recall any adverts?
* How did you know?

We may return to the results screen to further probe or confirm any understanding following the initial open recall questions.

over 4 years ago


Aaron Scott

Cracking article, funny enough we were talking about this earlier today and this gives me some definitive stats.

over 4 years ago


Kari Rippetoe

That last comment in the article from Jon Dodd relates right back to a post that was published just today on our blog that basically says "don't trust your gut when it comes to digital marketing decisions." We may think we know our audience, and then something like a landing page test throws us for a loop. The truth is in the data.

over 4 years ago

Pete Austin

Pete Austin, Founder and Author at Fresh Relevance

@Jon: Thanks for publishing the summary of the research protocol - this is extremely useful. Do I understand correctly that the only question about adverts was, "Recall any adverts"?

This question could be understood in two ways:
(1) Do you recall seeing any adverts?
(2) Do you recall the content of any adverts?

I assume the first meaning was the intended one and so you are treating "no" as meaning people didn't recognize the adverts - leading to commentary such as "about a quarter of people don’t know that Google had any advertising at all".

However, I would personally have assumed the second meaning. If other people did this too, "no" might mean they saw and recognized the adverts just fine, but ignored them and hence didn't recall their content.

Is the summary is accurate about the questions? If so, you might consider running a follow up survey.

over 4 years ago

dan barker

dan barker, E-Business Consultant at Dan Barker

As per Pete - thanks a lot for publishing & sharing the figures, and thanks for the kind follow up too.

And - as per Pete - I'd mentioned on Twitter I would also have assumed "Recall any ads?" to mean something different - as in "Do you recall which ads were present?" rather than "Do you recall whether any ads were present?". I suppose it could have just been shorthand in your summary though.

I had a chat with Graham & set a follow-up survey running. The answer "I've never seen an Ad on Google" is currently running at 11.5% after 348 responses. (of course, caveats apply on the wording, audience etc too).

Thanks again for sharing!


over 4 years ago


Jon Dodd, Managing director at Bunnyfoot Ltd

@dan and @Pete thanks for your responses and questions. Just to reiterate that this was face to face task based testing using a skilled moderator. The summary of the protocol shows the range of questions and general approach - but it is important to note that it isn't and question and answer, or fill in this form type post test interview. It is more of a conversational depth interview covering those topics. Hence information is probed and uncovered using appropriate questioning by the experienced moderator....

...on purpose it does not ask about adverts until if absolutely necessary at the end - because this leads the witness. We probe understanding in a more sophisticated way which is the key benefit and reason for doing face to face, one on one qualitative research.

@dan Your survey is interesting and I think the 11.5% is probably in line with our findings given your likely sampling bias (how did you recruit respondents btw?) and the leading nature of the wording. In fact given those two I am surprised it is actually that high.

over 4 years ago


Oliver Worthington

@Graham thanks for the info and @Jon the questions asked shown in the comments. In Australia we've seen Google display a multitude of "Ads" with the ad symbol and without, with shading variations and without. There is no need to satisfy the EU regulations here, so that would suggest thorough live audience testing.

over 4 years ago

Pete Austin

Pete Austin, Founder and Author at Fresh Relevance

@Jon: Thanks! The protocol was described more fully in the report from last year, which I've now read at:

I think the 36% figure [and last year's 40% figure] are explained by the composition of your sample, "With a view to exploring a representative of the general UK population rather than an online only population, test participants were recruited from an in-street intercept."

If that's correct, the result presumably means, "36% of [the general UK population] still don't realise that Google Adwords are ads".

This would seem to include people who are unlikely to use Google - a bit like including results from non-drivers in a survey of motorway sign recognition.

If you only include people who are say they would normally use Google for your task, the percentage who do not recognize ads is presumably much lower. Any chance of releasing this figure?

over 4 years ago


Jon Dodd, Managing director at Bunnyfoot Ltd

@pete - thanks for your interest and your comments

For the research reported in this post (as reported in the research methodology section above) ALL 103 participants used Google as their primary search engine. for the previous report the vast majority of participants also used Google as their primary search engine.

What might be interesting would be to repeat the experiment with people who had never used Google (as you would expect these people are relatively difficult to find and are not representative of the wider population at all) to see if the Ad signal was getting through at all - I would expect that the %'s in this case would be much higher

over 4 years ago


Joey Moore

Jaw-dropping is the right phrase to use here!

@jon, thanks for sharing.

This must be particulary annoying for any advertisers looking to reduce bidding on their own brand terms if they are already coming tops of the organic results.

I'd love to know how much of Googles revenue is derived from competative brand bidding and own brand bidding to combat it.

over 4 years ago

Ivan Burmistrov

Ivan Burmistrov, Usability Expert at interUX Usability Engineering Studio OÜ

How do you explain the difference between 10% and 36-41% in the previous studies?

Do you believe that a yellow box [Ad] is solely responsible for this discrepancy?

about 4 years ago

Pete Austin

Pete Austin, Founder and Author at Fresh Relevance

@Jon Dodd: Just seen your comment at 11:28AM on 12th May 2014.

It's quite possible that we're both right: "ALL 103 participants used Google as their primary search engine" AND the surveys "include people who are unlikely to use Google", given that you don't say how many of the participants had little Internet experience.

I could use Google infrequently, but it would still be my primary search engine if I never used e.g. Bing.

A trivial example is that e.g. if I just click the current Google Doodle, or search for Hello World, I don't see any adverts.

about 4 years ago

Pete Austin

Pete Austin, Founder and Author at Fresh Relevance

OK. Seems (a) we have people reporting they don't see Google adverts and (b) we know - from the suggestions in my previous comment - that Google doesn't necessarily show adverts alongside search resultss.

So what is there a lot of, on the Internet, that people might search for and that Google might not want to advertise against?

... Have a guess ...

You are probably right. It's porn. I just tried searching for a couple of porn stars and the results came back with no adverts alongside them.

So my my theory is that these people why don't see Google adverts are searching for porn and other NSFW categories where there are no Google ads to see.

Feel free to check for yourself, but I prefer this theory to the alternative that theye are unable to recognize adverts.

about 4 years ago

Doug Kessler

Doug Kessler, Director at VelocitySmall Business Multi-user

Feels to me like the confusion is entirely intentional on Google's part.

The old yellow tint-box was more obvious to me than the little orange 'Ad' signal -- which disappears in the chrome.

Another example of the blurring of the line between ads and editorial.

Caveat Lector.

almost 4 years ago

stefan martinez

stefan martinez, Analyst at Flow Simple

Hi Graham,

Great article! Thanks for all this useful information.

Happy to hear that even with the changes made on Google search over the years there are still so many people clicking on ads. I look forward to reaching more of your articles.


Stefan T. Martinez

almost 3 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.