How People Use Google Search (New User Behavior Study)

How People Use Google Search
(New User Behavior Study)

How people use Google search – Hero
Brian Dean

by Brian Dean · Updated Aug. 20, 2020

We analyzed 1,801 Google user behavior sessions to better understand how people interact with modern Google SERPs.

Specifically, we investigated:

  • How many people click on ads vs organic results
  • Percentage of clicks that go to local, video and Google Shopping
  • Average search session length
  • Number of users that make it to the bottom of the first page
  • Lots more

And now it’s time to share what we discovered.

Here Are Some of Our Key Findings:

1. Google searchers use one of Google’s autocomplete suggestions 23% of the time. People that search for informational and local searches tended to click on an autocomplete suggestion more often than those searching with commercial queries.

2. 50% of Google users click on their result within 9 seconds of searching. And the average amount of time it takes a Google searcher to click on something is 14.6 seconds.

3. Only 9% of Google searchers make it to the bottom of the first page of the search results.

4. 15% of users modify their initial search term. This suggests that Google tends to serve up highly-relevant results. Or that Google users are adept at choosing the right keyword on their first attempt.

5. Only 17% of users bounced back to the search results after clicking on a result. Only 5% of users bounce more than once for the same query.

6. The majority (59%) of Google users visit a single page during their search session. Only 6% need to visit four or more pages in order to get an answer to their query.

7. 65% of searchers click on a traditional “10 blue link” search result during their search session.

8. Although this varied widely depending on the query, 19% of searchers click on a Google Ad during their search.

9. For local queries, 42% of searchers click on results inside of the Google Maps Pack.

10. 19% of users searching for a product click on a Google Shopping result.

11. On average, only 3% of searchers interact with a “People Also Ask” box. Although this behavior varies widely depending on the query. For example, 13.6% of people searching for supplements clicked on a PAA box.

12. Only .44% of searchers go to the second page of Google’s search results.

13. The average search session takes 76 seconds to complete. And half of all search sessions are finished within 53 seconds.

Methodology: How We Did This Study

For this research we asked 454 US-based human users to perform a set of tasks using Google search.

In order to get a broad set of different queries, we asked our subjects to execute the following search tasks:

Commercial (Physical Product): Find a car phone holder for a Ford F-150 truck under $20.

Commercial (Service): Find an airline credit card that offers at least 20,000 miles to new card holders in the first year.

Commercial #2 (Service): Find a small business credit card with no annual fee and an interest rate of less than 25%.

Local: Find a car accident lawyer in your local area.

Informational: Find the best video to show you how to cut your own hair.

Informational: Find the best way to buy groceries online in your local area without any added delivery fee.

Transactional: Find supplements that may help with lower back pain.

We then recorded each subject’s screen during their session.

Here’s a sample of one of those recordings.

We then reviewed each video individually to make sure that the survey instructions were followed. In total, we ended up with 259 users (1,801 search sessions). With an overall video recording length of 2,226 minutes (38 hours).

Finally, we annotated the video data. Entered each data point into a spreadsheet. And analyzed the results.

With that, let’s break down the results.

Searchers Use Google Autocomplete Suggestions 23% of the Time

Google’s autocomplete suggests a number of keywords during your search based on search history, geography and what you’ve typed in so far.

According to our research, a fair number of Google searchers use this feature. Specifically, we found that about a quarter of all searches resulted in a user choosing one of Google’s autocomplete suggestions.

However, the number of times someone used Google autocomplete somewhat depended on the query.

Specifically, people searching for commercial and transactional keywords tended to use autocomplete suggestions most often.

Key Takeaway: It appears that Google autocomplete has a significant impact on search behavior. Especially for transactional and commercial queries.

Half of Google Users Click Within 9 Second of Their Search

Next, we wanted to investigate “time to first click”. In other words, how much time a user spent scanning the results before deciding to click on something.

We hypothesized that, while many people instinctively click on the first result, that user behavior may be changing due to the high number of SERP features that Google now uses, including video carousels, people also ask boxes, and top stories.

Not to mention that SERP layouts can vary dramatically between different searches.

This combination makes it harder for users to quickly choose the first organic result on the page. Depending on the query, the first result may be above the fold. Or well down the page. And sometimes sandwiched between Google Shopping and Google Ad results.

We discovered that 50% of people that search for something in Google click within 9 seconds. We also found that 25% of users click within 5 seconds.

On average, the “time to first click” is 14.6 seconds.

It’s also worth noting that time to first click varied somewhat depending on the query type.

Key Takeaway: Google searchers spend significant amounts of time sizing up the results before making their first click. The mean “time to first click” for a search is 14.6 seconds.

9% of Users Scroll to the Bottom of the First Page

Next, we wanted to see how many Google searchers actually scrolled down to the bottom of the first page results.

We discovered that only 9% of people ended up at the very bottom of the results.

However, this user behavior was dependent on the type of query. People searching for our transactional “find supplements for back pain” query saw significantly more of the SERPs compared to those searching for the commercial “find a car phone holder”.

This is likely due to differences in SERP features.

Commercial keywords tend to bring up Google Shopping results. Which almost always appear above the fold.

On the other hand, transactional and informational terms typically present a more traditional “10 blue links” layout. Which can lead to more scrolling and clicking further down the page.

Key Takeaway: Relatively few (9%) of Google users make it to the bottom of the SERPs. Many either find an answer to their query towards the top of the page. Or, if they don’t find what they need above the fold, prefer to modify their query instead.

Only 15% of Google Users Modify Their Initial Search Term

We discovered that 85% of Google users find an answer to their query with their initial search term.

This may be due to the fact that Google (through updates like Hummingbird and RankBrain) has improved their ability to understand the true intent of a query.

Or that Google’s search results are simply better than in the past. Which leads to someone finding a relevant result towards the top of the page.

It could also be that Google users better understand how to search. In other words, they’ve been trained over thousands of searches to phrase their queries the right way.

Either way, Google searchers generally find what they need with their first query. Which is why very few ultimately change their query after their initial search.

Key Takeaway: Only 15% of searchers ultimately modify the keyword that they initially searched with.

17% Google Searchers Bounce Back to the SERPs

Although controversial, there is evidence that Google’s algorithm looks at “behavioral signals” to figure out whether uses are satisfied with the results.

Regardless, we wanted to see how many people bounced (or more precisely “pogo sticked”) after clicking on a Google result.

And we found that 83% of users didn’t bounce from the result they chose to click on. And that only 5% of users bounced multiple times during a single search session.

Although, again, this varied depending on the query.

Our research also found that the majority (59%) of searchers clicked on a single result. And that a small percentage of users (6%) visited 4+ pages in order to find an answer to their query.

Again, this shows that Google generally does a good job of satisfying a user’s query.

Combined with the fact that, as discussed above, Google searchers spend a fair amount of time deciding on the most relevant result to click on.

Key Takeaway: 17% of people that click on a Google result bounce back to the SERPs in order to choose a different result.

65% of Google Searchers Click on The Organic Results

Google has rolled out a number of SERP features over the last few years, including Featured Snippets, Knowledge Panels, Twitter cards and more.

Despite that, we found that 65% of searchers clicked on at least one “10 blue link” organic result during their search session.

That’s not to say that these users didn’t also click on an ad. We found that many users do click on both organic results and ads during their search session.

However, it is noteworthy that the vast majority of searchers end up clicking on the organic results at least once during their session. And a fair number (10%) click on multiple organic listings.

Key Takeaway: 65% of Google users click on an organic result during their search session.

19% of Users Click on a Google Ad During Their Search

Google Ad listings take up more SERP real estate than ever before.

Today’s Google Ad placements are also more prominent and now largely resemble organic results.

We found that 19% of Google users click on a Google ad. And that only 4% click on multiple ads during the same search session.

This number may be a bit higher than expected due to the fact that many of the queries in this study were commercial or transactional in nature.

In fact, we found that this behavior was highly dependent on the type of search someone was conducting.

Specifically, commercial and transactional searches led to significantly more Google Ad clicks compared to informational queries.

This is likely due to the fact that those searches generate more ad placements. And that the search intent is more geared towards product and service pages. Which may lead users to find ads highly relevant for their search.

Key Takeaway: 19% of Google users click on a Google Ad. However, this number varies widely depending on the type of search that someone is performing.

42% of Local Searchers Click on the Google Map Pack

For local queries (“sushi New York”) and even queries that Google can consider local depending on context, they present users with a “Map Pack”.

A map pack is a set of three results pulled from Google Maps that’s mixed in with the “normal” organic results.

We found that 42% of people searching for local terms ultimately click on a result in the Google Map Pack.

Considering its prominence in most local SERPs, it’s no surprise that a large number of users choose to find a local business using the Map Pack.

Key Takeaway: 42% of people searching for a local business click on a Google Map Pack result.

19% of Product Searches Result in a Google Shopping Click

Google Shopping results appear above the fold for many product-specific search queries.

So it should come as no surprise to find that 19% of people that search for a product end up clicking on a Google Shopping result.

Key Takeaway: 19% of people searching for a product click on the Google Shopping results at least once during their search session.

Only 3% of Google Users Interact With a People Also Ask Box

According to data from SEMrush, 43.21% of all Google search results have a People Also Ask Box.

Which makes them one of the most prominent features across all SERPs.

We wanted to know how many users actually interacted with PAA boxes.

And we found that very few (only 3%) of searches in Google result in a click on a People Also Ask box.

However, we did find that certain transactional and navigational searches result in much higher PAA box interaction (13.6%).

Key Takeaway: While People Also Ask boxes appear in a high number of SERPs, they’re not interacted with that often for most queries. In fact, only 3% of searches in our sample resulted in someone interacting with a PAA box.

0.44% of Users Visit Google’s Second Page Results

Not surprisingly, we discovered that very few users (.44%) make their way to Google’s second page.

This is in-line with organic CTR research that we conducted last year using data from Google Search Console.

In that study we found that .78% of users visited the second page of the results.

Key Takeaway: Only 0.44% of Google users visit the second page to find an answer to their query.

The Average Google Search Session Lasts 76 Seconds

When someone searches for something in Google, how long does it take them to find what they’re looking for?

According to our data, the average person ends their search session after 76 seconds.

We also found that 25% of search sessions were super fast (<31 seconds).

Like many of our findings, the timeframe here varied depending on the type of search our subject was conducing.

Again, this shows that Google does a relatively good job of packing the first page with relevant, useful results. And providing SERP features that help direct people to relevant content faster.

Key Takeaway: Although somewhat dependent on the query, our data shows that most Google searches are completed successfully in a relatively short time frame (76 seconds on average).


I hope you enjoyed this research.

The Google SERPs are rapidly changing. More ads. More SERP features. And even entirely new search experiences, like Google Discover.

Which is why we wanted to get a glimpse into how users interact with modern day search results.

If you’re interested, here are our methods and full set of results. And a link to a GitHub repository with the raw data.

Let me know if you have any questions in the comments section below.


  1. This is a genius idea Brian – nice work! Some super-useful data as well, thank you for putting it together

    1. Brian Dean Avatar Brian Deansays:

      Hey Nathan, thank you. This study was definitely one of our hardest to date. But in the end, it was cool to get an insider look at how people interact with the modern SERPs and different SERP features.

  2. This is an amazing and in-depth study. I’d say a lot of the results are things I suspected and things that I see myself doing often.

    I wonder how different the results might be if the searcher was needing to purchase a more expensive product like a new roof for their home or a new car. It’d be so interesting to be able to see how a fully organic user would react to the serps.

    1. Brian Dean Avatar Brian Deansays:

      Hey Cole, thank you. Yes, that would be intersting. Especially becasue that search would probably play out over the course of days and weeks.

  3. Hey Brian, Thanks for sharing this some really interesting findings. Not sure I would have guessed on many of these values. ‘19% of product searches result in a Google shopping click’ seems surprisingly low. As does ‘19% of Users Click on a Google Ad During Their Search’. Obviously not doubting your figures!!
    How different do you think the figures would be for mobile users?

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, Rich. That’s funny because I thought fewer people would click on Google Shopping results. It takes up a ton of SERP real estate. And it super relevant for shopping keywords. But there’s still basically an entire set of search results below the shopping results.

  4. Jeannette Avatar Jeannettesays:

    Awesome information! Thank you for all of your work on this : ) The demographics of your 454 users would also influence the results. I know many, many people, who do not really understand how the search engines work and have what an experienced user would consider awkward steps to find something. I’m sure your users were varied. Thanks again!

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, Jeannette. Absolutely: that would be interesting to look at. This research was pretty time intensive. So we didn’t really have a chance to analyze user behavior based on demographics.

  5. As always, great content! Just a heads up that there’s a typo on key takeaway #12. Should be 0.44% instead of 44%.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Eric. I fixed that typo.

  6. Victor Avatar Victorsays:

    Thank you for that very useful information.

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, Victor.

  7. Thanks, makes interesting reading. Are you able to say whether paid Ads are as effective (for the user getting what they are looking for) as organic. Also whether you have any insights into the profile of the users, most of the more tech savvy people I know (who are younger) say they very rarely click on paid but obviously many people are just wondered whether older=paid and younger=organic in a significant way? Thanks. Nick

    1. Brian Dean Avatar Brian Deansays:

      Thanks Nick. At least based on our data, it looks like people do get value from ads on certain queries. In terms of the profile, this was likely more of a tech savvy group than the average public. But we really focused on geography (US) and didn’t get into specific demographic info this time around.

  8. Great article, it shows 2 two very important things for e-commerce websites: A. Google Shopping is essential nowadays; B. The best place to hide a dead body is in the second page of Google. 🙂

    1. Brian Dean Avatar Brian Deansays:

      Well said, Antonio. And for sure: the second page basically doesn’t exist anymore.

  9. John Stevenson Avatar John Stevensonsays:

    Nice! Curious though, what was the study’s statistical significance? Under 2,000 sessions seems like a small sample size to draw some pretty significant conclusions given the individual sessions varied quite a bit.

    Regardless, appreciate the time the team puts into these as always.

    1. Brian Dean Avatar Brian Deansays:

      Thanks John. The reality of user experience research like this is that sample sizes need to be on the smaller side. For example, this earlier study by Nielean Norman Group looked at 471 queries. We looked at 1800. So this is actually a relatively large sample size for this kind of research.

  10. Great information Brian, thank you!!

    Sounds like your tested users executed a large number of searches. Do you think the results will match real world users who would likely conduct these searches spread out much more over time? Seems like when you have a long list of searches to do you’d move through more quickly and react differently.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Fred. There’s probably a few differences for sure. Thing is: when you search don’t you want to get an answer as fast as possible? And don’t you generally search for a few things in a batch? In that way the behavior isn’t that different than someone conducting a “real” search. Know what I mean?

  11. 9% of users actually reach the bottom of the first page is much lower than I thought. Really puts the emphasis on being on the top of that first page.

  12. Brian always presents incredibly relevant information, giving 120% of effort to his articles. I always learn new things that help my online business. I’m very grateful, Brian.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Jodi. Glad you enjoyed the report.

  13. Thank you so much Brian, I don’t know how you can manage to do this much stuff.
    I get bored writing and thinking after a certain period of time.
    Hats off man 👍

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome.

  14. Badal Sarkar Avatar Badal Sarkarsays:

    Thanks Brian for such a helpful article. That will help us to do seo works with more advance knowledge.

    1. Brian Dean Avatar Brian Deansays:

      Yup, any insight into how users actually interact with Google can help SEOs like us do better work.

  15. Leon Avatar Leonsays:

    Hi Brian,

    I have the following question about your article that I didn’t understand:

    “ Only 9% of Google searchers make it to the bottom of the first page of the search results”


    “Only 44% of searchers go to the second page of Google’s search results”

    The numbers go against each other, what am I missing?

    I am very grateful for all your articles and I look forward to everything new that will come from you!



    1. Brian Dean Avatar Brian Deansays:

      Thanks Leon. That’s actually a typo. The actual figure is “.44%”.

  16. Hi Brian,
    Excellent job … because I see a lot of people have some stats but I never know how they doing them…

    Anyway I was surprise about the : 42% of Local Searchers Click
    also the PAA only 3%…
    But the serp keep moving …


    1. Brian Dean Avatar Brian Deansays:

      Thanks Claude. I was surprised about the 3% PAA figure as well. But it makes sense if you think about why PAA exists: it’s kind of a fail safe for when users don’t find what they’re looking for. And most of the time, they do.

  17. Ed Avatar Edsays:

    “76% of Google Searchers Click on The Organic Results” – The graph suggests ~35% are 0-clicks. Was that stat meant to be 67%?

    1. Brian Dean Avatar Brian Deansays:

      Thanks Ed. You’re right. That was the wrong title on that graph. Fixed now.

  18. Miles Pfefferle Avatar Miles Pfefferlesays:

    How can this be true:
    “3. Only 9% of Google searchers make it to the bottom of the first page of the search results.”

    If this is also true?:
    “12. Only 44% of searchers go to the second page of Google’s search results.”

    1. Brian Dean Avatar Brian Deansays:

      The 44% number was a typo. It’s actually .44%.

  19. Jan Ranostaj Avatar Jan Ranostajsays:

    Great study and effort!!

    In point 12 you have a typo in number “Only 44% of searchers go to the second page of Google’s search results.” – shouldn’t be 0.44%? As showing in the graph?

    1. Brian Dean Avatar Brian Deansays:

      Thanks Jan. Good catch there. That was a typo. As you said, it’s .44%.

  20. Hey Brian

    Really cool facts about Google user behavior.

    Thanks for sharing

    Amit Garg

    1. Brian Dean Avatar Brian Deansays:

      No worries, Amit.

  21. SF Avatar SFsays:

    I’m not sure that your interpretation in #4 – “Only 15% of users modify their initial search term. This suggests that Google tends to serve up highly-relevant results. Or that Google users are adept at choosing the right keyword on their first attempt.” -is necessarily completely correct.

    I’m an academic librarian who’s done usability testing on searches in library catalogs and in article databases, and what I’ve observed (and have read studies where other people observe) is that that searchers usually don’t change their original search terms, even when it’s clear (to me, though not to them) that they’re not getting the results they actually need.

    One of the key things we tell patrons when doing reference education is to try synonyms, and you’d probably be surprised at the number of non-sarcastic “Oh… That’s a good idea!” reactions we get.

    This is probably less of a problem if you’re searching Google for something like “where can i get my car painted” than if you’re searching for “teenage smoking” in an academic database (where you need to search for “adolescent tobacco use” to get the results you need), but I’d still be wary of concluding that a search engine is doing a great job serving up results or that users are good at identifying search terms.

    1. Brian Dean Avatar Brian Deansays:

      Thanks SF. Yup, that finding is open to interpretation for sure. Interesting insight there. As you said, part of it may be that people just aren’t inclined to change their original search term -regardless of whether or not the results seem like a good match.

  22. Great studies thank you the team

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, Robert. Glad you learned something new.

  23. Guillermo Avatar Guillermosays:

    This is great. As a business consultant this shows me exactly where a business must show up to increase the likelihood of them getting the click.

  24. Fantastic! How are you able to pull this off. Takes so much effort and patience.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Rashid. This one definitely wasn’t easy. Lots of work to figure out the best tasks that will result in a wide set of queries. How to record the data, analyze it, etc. etc. But overall, I think we learned some interesting things.

  25. There is a mistake in the beginning where it’s 44% visit second page but later the right number is 0,44%, which makes a lot more sense.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Martin. Yup, that was a typo. It’s fixed now.

  26. Incredible data. this is the most useful survey I have seen lately about SERP.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Moshon.

  27. Super interesting stuff Brian! Appreciate all the work you put into these studies 🙂

    1. Brian Dean Avatar Brian Deansays:

      Thanks Davis.

  28. Excellent work as usual Brian!
    I drew the inference that depending on level of pain searcher has,
    determines, to an extent, Time on page, bouncing to multiple results etc.
    VERY surprised “people also asked” section was so low at 3%

    1. Brian Dean Avatar Brian Deansays:

      Thanks Larry. I was surprised at that too. But it does make sense considering that the PAA boxes are there as sort of a second option for when people didn’t find what they needed from the traditional search results.

  29. Very good article, and datas… (like usual)
    nothing really surprises me… except 0.44% of people go to page 2 of google… (i am one of them).
    58% click only on one link…

    1. Brian Dean Avatar Brian Deansays:

      Thanks Olivier. The .44% number didn’t surprise me because we found something very similar last year. But yeah, very few people visit Google’s second page for sure.

  30. Awesome post brian. I want to ask you that how do you come up with such amazing unique topics? Do you have any video or post that i can see to learn your way of keyword research?

    1. Brian Dean Avatar Brian Deansays:

      Thanks. I actually have a post coming out on that exact topic in about 2 weeks.

  31. excellent content brian, lots of insights while reading. thanks for sharing it!

    1. Brian Dean Avatar Brian Deansays:

      No worries, Vinicius. Glad you enjoyed it.

  32. Another killer post Brian, you always come with something new and interesting. Tons of really useful numbers and data. I was little surprised by almost 20% of users clicking on Google Ad, that’s really a lot, but I guess they blend in so well now that many people do not even see the difference…

    1. Brian Dean Avatar Brian Deansays:

      Thanks Tom. For sure: the ads really blend in. Plus, for lots of commercial queries they can actually be solid results.

  33. By far, this is the best study on the Google results page I’ve read.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Luci. We all put a lot of work into this so that’s nice to hear.

  34. N5 amazed me. Is that means that the results are super relevant to the user’s search?

    1. Brian Dean Avatar Brian Deansays:

      Or that people are good at choosing keywords. Probably both.

  35. ankit thakur Avatar ankit thakursays:

    Hey Brian Great work.

    But if people didn’t scroll down for the last search Result on first page of Google then there must be chances of getting less click if you are ranking after 5 postion?

    1. Brian Dean Avatar Brian Deansays:

      That’s right. Most of the clicks are at the top of the page. It falls off big time after position #3.

  36. Thanks for such an amazing post Brian. Very informative!

    1. Brian Dean Avatar Brian Deansays:

      No worries, Darshana. Glad you liked it.

  37. Great to get numbers on these! Thanks for sharing.

    1. Brian Dean Avatar Brian Deansays:

      Hey Simon, for sure. I’d like to do an expanded version with more keywords and participants. That way, we could get more numbers on things like Featured Snippet clicks.

  38. Thanks for some great analysis Brian – as SEOs this is something we often think about and act on, but if we’re honest a lot of our information comes from opinions and gut feel – so it’s great to see some fact-based analysis with which to inform our opinions and decisions.

    1. Brian Dean Avatar Brian Deansays:

      Hey James, you’re welcome. That’s true: we kind of have to rely on gut and feel because Google doesn’t share this sort of stuff. That’s actually one of the things that motivated us to do this research: we wanted to see how users interacted with modern SERP features.

  39. Hi Brian,

    Thank you for the effort in putting this together, wondersful stuffs. I am running a digital agency from Singapore and although the data is from the states, it provides a good benchmark for us to share with our clients. Some stats are surprising though, especially the PAA and the % of people click on multiple ads within a single session.. although it differs based on search terms.

    Good insights, enjoyed reading.

    1. Brian Dean Avatar Brian Deansays:

      Hi Ian, you’re welcome. As you said, there are probably some differences between countries. But I’m sure user behavior is somewhat universal too.

  40. Mike Carabott Avatar Mike Carabottsays:

    Great study Brian, so many insights.

    1. Brian Dean Avatar Brian Deansays:

      Hi Mike, thank you. For sure: I learned a lot from the findings. I was surprised by some things. And other findings were in line with what I expected. That’s why it’s good to do this sort of data-driven research.

  41. Marilou Borlagdan Avatar Marilou Borlagdansays:

    Always great stuff from you, Brian.

    1. Brian Dean Avatar Brian Deansays:

      No worries, Mailou. Glad to hear you learned something new from our research.

  42. A fabulous compendium of information as always Brian, many thanks.

    1. Brian Dean Avatar Brian Deansays:

      Hi Craig, you’re welcome.

  43. Great research and article! Thanks for the info!

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, Eddie.

  44. I thought the local search had 30% CTR, but it’s over 40 – that’s amazing! So I need to optimize my google business 😳.

    1. Brian Dean Avatar Brian Deansays:

      I mean, many local businesses focus more on local SEO than “normal SEO”. And according to our data, that approach makes a lot of sense.

  45. Really nice research – we are about to complete our blog article with the new knowledge by autocomplete suggestions.

    Thanks Brian!

    1. Brian Dean Avatar Brian Deansays:

      Hi Kim, you’re welcome. Sounds good. Let me know how it goes.

  46. Great statistics Brian. I think people use autocomplete feature more on mobile devices than on Desktops.

    1. Brian Dean Avatar Brian Deansays:

      Thanks Pawan. That could be for sure.

  47. Hey Brian,
    Thanks for the time and effort you put into for this kind of data-driven posts. This looks like a lot of work!

    This is kind of off-topic, but I wanted to ask how people interact with emails vs. social media/messenger messages in your personal experience.

    I mean, which has a better response + conversion rate? i know there is a lot of studies on that, but I wanted to know your personal opinion on this.

    Looking forward. Thanks.

    1. Brian Dean Avatar Brian Deansays:

      Hi Michael, thanks. I mean: social messengers definitely have higher engagement rates. The issue is how to reach people through those platforms. Which is why I focus 100% on email.

  48. Cosmin Avatar Cosminsays:

    Great article – doing a forensic audit right now for a car accident lawyer firm – this brings in so much insight regarding the importance of Ads, Titles & Metas as well as website landing page. Thank you! 🙂

    1. Brian Dean Avatar Brian Deansays:

      No worries, Cosmin. Happy to help.

  49. The top reason why it’s important to be at the top of the search results. Great research Brian, thanks for sharing 🙂

    1. Brian Dean Avatar Brian Deansays:

      For sure, Luka. It’s all about the top 3.

  50. Hi Brian,
    All your blog articles are fresh actual SEO insights. Thank you for that. 42% click on Google Maps. I have a Taxi company as client, and i believe it’s for them even more than 42%.

    In the Google Map – GMB. I see sometimes results in the top 3, without a website. Is there a tool to analyze Google maps?

    1. Brian Dean Avatar Brian Deansays:

      Hi Luc, for sure: I bet for certain local queries it’s higher than 40%. 40% was just the average. To answer your question, I list out some local SEO tools for that in this guide.

  51. Great article, Brian. If we assume the SERPS would be better for the user without ads, could you ever see a world where a new search engine without ads starts outperforming Google, which would then lead to Google removing (some of) their ads?

    Kind regards,


    1. Brian Dean Avatar Brian Deansays:

      Hey Jacob, interesting you say that. According to this small study, users find ads relevant for certain keywords. So I don’t think ads are necessarily a bad thing in some cases.

  52. As always, very in-depth and great work, Brian. Thanks for the share!

    1. Brian Dean Avatar Brian Deansays:

      Hey Justin, no worries. Glad you enjoyed it.

  53. Thanks for taking the time to do this research. I bet it was fascinating to watch users interact. I know as an ecom consultant and marketer, user interaction data is very valuable!

    I wonder what impact using a mobile device would have on your results. I’m betting the would be significantly different! Any plans to repeat using a mobile device to search?

    1. Brian Dean Avatar Brian Deansays:

      Hi Rebecca, you’re welcome. That’s true: I imagine the results would be different on mobile. The tricky part is recording people’s phone screens. It’s doable. But tough to scale to 1800 participants like we did.

  54. 1/ Great research as always Brian. Thank you.
    2/ What was the percentage demographic allocation in your research group b/w gen Z, milennials, gen X and Baby Boomers?
    3/ At the top of the blog, under “Here are some of key findings” did any demographic vary more than 15 percentage points from the average describe in items 1, 2, and 6?

    Have an awesome day!

    1. Brian Dean Avatar Brian Deansays:

      Hi Christopher, you’re welcome. We actually didn’t look at those types of demographic data points this time around. Although I’m sure we would have found some differences there.

  55. Thanks for writing this article. I forwarded this to our marketing team.

    I was most surprised by how many people use a suggested query from Google. 🤯

    1. Brian Dean Avatar Brian Deansays:

      Thanks Reuben. 23% is pretty high. On the other hand, Google is really good at suggesting relevant queries.

  56. Sim Campbell Avatar Sim Campbellsays:

    Ah-Mazing research, Brian! Hats off to you and the team.

    I can only imagine how a competitor in a niche can absolutely dominate a SERP and related queries by taking up not only organic real estate but Ad and map pack placements. All those combined could make that particular business a pretty penny.

    1. Brian Dean Avatar Brian Deansays:

      Hey Sim, that’s true. It’s not just about ranking #1 anymore. You also want to take up as much SERP real estate as possible.

  57. That’s the main reason to stay above search results. Great research Brian, for sharing

    1. Brian Dean Avatar Brian Deansays:

      No worries, Voski.

  58. Stan Avatar Stansays:

    Exactly. I’d say, that when a user doesn’t exactly know what kind of question they want to ask, they click on PAA boxes.

    I personally click on them also when I try to find similar things to what I originally asked.

  59. Really interesting stuff, but I wonder if the low clicks for People Also Ask might have something do with the tasks users were given?

    PAA is something I use a lot when researching a topic generally, whereas the tasks you’ve given are very specific.

    Do you think that if a task had been more open, along the lines of “Find out about FODMAP” there would have been a lot more use of the PAA section?

    1. Brian Dean Avatar Brian Deansays:

      Hi Hannah, interesting thought there. That could have definitely influenced it. As you said, we gave people specific tasks vs general research. Which could lead to people ignoring PAA.

  60. Fiona Avatar Fionasays:

    Great post Brian.

    Thank you for sharing it with us!

    What software do you use to create your graphs?

    1. Brian Dean Avatar Brian Deansays:

      Thanks Fiona. We actually work with a graphic designer who creates those for us.

  61. Gotta say, I dig how you start these off with a list of primary learnings. I find myself coming back to them often.

    If I may, though, what’s the primary reason you allow comments (and links) on your posts?

    1. Brian Dean Avatar Brian Deansays:

      Thanks Pete. It’s important to have a community on my blog.

  62. so in conclusion if after all hard work on your seo project you can’t get to top first page or even top 3 search result it’s kind of useless to continue, change strategy radically to improve or just abandon it and move on to other ventures

  63. Wow, Brian.

    You really know how to create an interesting study that’s interesting to your target audience and which helps you acquire tons of new backlinks to your site.


    1. Brian Dean Avatar Brian Deansays:

      Thanks Martin.

  64. Brian: this study is extremely helpful. I’ve been sharing tidbits from it in my Twitter, LinkedIn and Facebook networks all week. I wanted to let you know that this specific piece of data really correlates closely with my findings in Google My Business data for my clients: “42% of Local Searchers Click on the Google Map Pack.” Google My Business most likely has more mobile users than desktop users and that would explain why most of my clients have around 52% of their search traffic coming from maps vs. SERPs (depending on if my clients have physical locations or are service-based). Thank you, thank you for producing this Google User Behavior study.

    1. Brian Dean Avatar Brian Deansays:

      Hi Lori, Thanks for sharing that. Interesting to see that your field data pretty much aligns with our experimental data from this study.

  65. Do you think a change in the search intent would have an effect on your results? In the search terms you used, they appear to have a high buyers intent – however, if the users aren’t actually going to buy the buyers intent drops significantly and changes the search intent to more of a proof of existence. It’s the difference between asking a group of people how they would spend their money (which is theoretical, with no consequence), vs watching how they actually spend their money (which is practical, with time/cost consequences).

    My thought is that the searches with higher buyers intent would slow the process down significantly. For example, If i’m searching for something as small as small as a $20 phone holder – I may check a few sites and even go to page 2 because it would take me a few sites to find out what’s even available, and then having to decide which one I actually want. However, if I’m searching with lower buyers intent, I may not leave the top 3 results as long as I find something that “checks the box”.

    That being said, I realize that what I’m suggesting would be difficult to test. Even if you have the research group money to make purchases, it might not increase their buyers intent much because it’s not their money, and that reduces the inherent consequences of a “wrong” choice… Either way, would love to see what you think.

    1. Brian Dean Avatar Brian Deansays:

      Hi Kyle, good point there. You’re right: any sort of user testing like this only really simulates actual behavior. How people actually interact when looking for a specific product that they want might be different than what we found here.

      Also, we did mix in some informational searches so it wasn’t 100% buyer intent (or in this case, simulating buyer intent).

  66. Dimitar Margaritov Avatar Dimitar Margaritovsays:

    I was a bit worried if I am too late to the SEO business but apparently there is still a good future as only 20% of people click on ads according to your research.

  67. Sebastian Avatar Sebastiansays:

    Yes, local seo is different from “normal SEO”, but it’s still the same thing, only done slightly differently.

  68. Hi Brian, how many searches are done without any clicks at all?

    According to Rand Fishkin, zero clicks constitutes a large portion of overall google clicks. Was there any evidence to that in your research?

    1. Brian Dean Avatar Brian Deansays:

      Hi Amit, zero were done without any clicks. I think Rand’s data is valid. But in our case, we focused on more in-depth tasks that require a click. So it doesn’t include things like “time in NYC” etc. that make up a large chunk of searches.

  69. I just want to thank you for such great insights. I have learned so much about SEO from your blog.

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, Ivan.

  70. Aaron Parnes Avatar Aaron Parnessays:

    Great study and excellent findings, but not much in the way of surprises though.
    Any meaningful difference between mobile and desktop users in your study?

    1. Brian Dean Avatar Brian Deansays:

      Thanks Aaron. This study was done with desktop users. Otherwise, it’s really hard to do screen recordings.

  71. The fact that only 9% of Google users even make it to the bottom of the first page of results should scare the bejesus out of every website owner on the planet.

  72. Tsuyoshi Avatar Tsuyoshisays:

    No click search is increasing. Do you have any idea how much conversion is countable or uncountable? Countable means clicked in SERP and uncountable means no clicked, but Phone call, check direction, and so on.

    1. Brian Dean Avatar Brian Deansays:

      We didn’t look at that in this analysis. But I’d say that no click conversions are a growing thing. Which makes conversion tracking harder.

  73. Interesting conclusions, Brian, thanks. I often find myself clicking on the first ad, more and more lately as they are doing a great job to make it as similar as possible to the other results.
    I also agree that the amount of ads is increasing, hope we won’t get to the “How to get on top of Page 2” guides, due to first page being all ads 😉

  74. James Avatar Jamessays:

    How do you avoid bots adding lots of comments to your page? We get hundreds every day.

    1. Brian Dean Avatar Brian Deansays:

      Lots of moderation.

  75. Very in-depth study with some really interesting insights. 9% of Users Scroll to the Bottom of the First Page and 0.44% of Users Visit Google’s Second Page Results.

    So it’s true that Google Second page is the best place to hide a dead body. 😉

    Thanks for the post Brain.

  76. Great blog post with a lot of useful SEO information, I am currently in the stages of trying to increase traffic to my website via links. In your expert opinion, what are some great tools besides Google for SEO? I heard Moz is good but is it worth the price if you are just starting up? I would love to hear back and get some tips from this board!

  77. Thanks for the insightful article. It was really surprising how little users use the “people also ask” box!

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, George. That definitely stuck out to me too.

  78. Spock Avatar Spocksays:

    Hey, one question. Is there any way I could see the bounce rate per certain search query (in google analytics)?

    1. Brian Dean Avatar Brian Deansays:

      You used to be able to see that data. But not since Google changed the keyword data in GA to “not provided”.

  79. I read this article several times because it got so much inside of it. The complete report on user behavior. The article, that is helpful to make your future strategies. It actually shows the effort that you put in. And personally saying, for SEO you have to put all your efforts, to get the desired results.

    1. Brian Dean Avatar Brian Deansays:

      Thanks David.

  80. Congrats, great work.
    The research was conducted on the US-based population. But, I live in Poland.
    Do you think that the conclusions also reflect other countries and languages (European mainly)?

  81. This is great — glad you included the data about clicks on maps results — I am curious, although you identified that 42% of users click the maps results, do you have an idea what the remaining 58% are doing? Are they preferring ads or are they going to the organic listings below the maps?

    1. Brian Dean Avatar Brian Deansays:

      Thanks Brian. It was a mix. It would be interesting to see exactly where those local clicks went.

  82. To resume your study if we are not ranked in the first page of google or better in the five first positions, it is very difficult to have traffic coming from google

  83. That is one hell of an article! Thanks….started reading but don’t want to skim so now in my Google ‘Keep’ to read properly tomorrow, thanks in advance!!
    Cheers! – Nigel

    1. Brian Dean Avatar Brian Deansays:

      You’re welcome, Nigel.

  84. Brian,
    That’s some great work done by you and your team. Its interesting to know that people still prefer Organic results over Paid Ads despite of the fact that Ads appear prior to the Organic results.

Leave a Comment

Your email address will not be published. Required fields are marked *