A Heat Map Click Study For Google’s Local Carousel Results

Last week I published a post where we recorded 10 random people’s reaction to Google Local Carousel results. I really found some great takeaways by looking at qualitative instead of quantitative behavior of searchers. After the study though, I wondered what the Click Through Rates of Carousel might be. In our study of 10 we found only 2 people clicked on carousel results and 4 people clicked on Yelp’s organic listings.

As I was sharing my findings with a few friends Matthew Hunt from Small Business Online Coach said he would like to conduct a heatmap test on the same search phrase to see what we could find. He set up the following study:

He asked 83 searchers to “Click on the part of this google search results page that most interests you.”

Here are the results:

Original Search Image

chicago-restaurants-Google-Search

 

Overall Clicks

Screen Shot 2013-06-25 at 9.53.35 AM

 

Heat Map

Screen Shot 2013-06-25 at 9.53.57 AM

Some Takeaways

40 out of 83 or 48% of the total clicks were on carousel results.

12 out of 83 or 14.5% of the total click were on the map.

Interesting enough, the 3rd carousel result had 9 Clicks and the 8th Carousel result had 6 Clicks. These two results were also the same two results that were clicked on in my study of 10 searchers.

Screen Shot 2013-06-25 at 10.23.48 AM

 

In my study both searchers mentioned the amount of reviews as a determining factor for clicking. Both images also have very full pictures of well prepared food. Alinea Restaurant has by far the most reviews and the best overall Zagat score but as a few people mentioned in my study “I am not exactly sure with the score means”. Also Alinea has a very poor picture of a gray building exterior to compete against much better pictures elsewhere.

One challenge with images is that Google is not letting you choose what picture to display on the front page. They choose it for you from the images that are uploaded. They want to see pictures of the place. I am not sure if they means they would rather that than pictures of the food (clearly for restaurants this would be a better option).

Ethical SEO Consulting did a similar study on the search phrase Pizza Delivery in Denver. While we have very different outcomes on which carousel results were being clicked on it was interesting to see that their map had 32.4% of total clicks and the map on our search had 14% but the carousel search result on their study got 30% of the clicks and we had 48%.

I’m very interesting to see other takeaways from this in the comments. Anything standing out to you?

 

21 Responses to “A Heat Map Click Study For Google’s Local Carousel Results”

  1. Interesting results Mike. Again, confusion re the review scoring. However, folks are ‘trying’ to seek out review info based on clicking on images with a high number of reviews. I imagine if the gold (or red) stars representing review scores would be a lot better indicator of review quality for the average user, and CTR would likely go up on those with a higher star count. Google, are you on this?

  2. Mike – thanks for the mention. ;)

    I also think reputation management is going to be much more important too, b/c once your image is clicked on in the carousel, then anything google can find on your company’s name is displayed below in the SERPs.

    If a restaurant or business has weak reviews on any of the major 3rd party reviews sites then this could pose as a problem. Be sure to see what shows up when your business name is searched.

    I think photos will play a big role in CTR’s. SMB’s can easily split test this out with a good SEM’er assistance. Just take a screenshot like this one and have their designer swamp in and out different photos. Then use a tool like usabilityhub or some other click test service to know which photo will most likely work best.

    Now it looks like from what I can see so far it usually pulls the first photo you uploaded… Not 100% sure if that is true… and if Google Places functions like it used to, then you’ll need to delete all your photos and and re-upload them to get the one you want to display first. I wish they would allow you to choose a feature image (from any of your photos.. yours or users uploads) to represent your business in the carousel.

    It is interesting that the photos with food got clicked on more than the ones with the location buildings.

    Nice post Mike.

  3. Cool data guys. Looks like things might be headed down a conversion optimization exercise related to the photos people choose for the local listings.
    Interesting thought.

  4. Great post. I was wondering about the carousel and how the average user interacted with it. So can you say now with relative confidence that the primary factor of whether you show up on a certain keyword result is sheer number of reviews?

  5. GOogle doesn’t always use first photo uploaded. I thought that as well but was told otherwise. They choose something that represents the “Place”.

  6. Thanks, for posting the study results, Mike. Have you seen the raw data? Is there a correlation between # of reviews and CTR on the carousel?

    I would also be curious if the mean response time was similar between those clicking on carousel results and those clicking on other results types. Are people spending more or less time looking at the carousel results than the organic/map/adwords results?

    • The tool we used doesn’t go into time. For that you might just want to watch a few of the videos I posted on my study where we recorded 10 people. Watching them interact was very enlightening for sure.

  7. Nice work, Mike, and thanks for the mention of our click test.

    There are a couple of things that I believe may account for some of the differences in our results:

    1. The prompt; our prompt to clickers was:

    “You are searching for a nearby pizza place. Where do you click?”

    Yours was:

    “Click on the part of this Google search results page that most interests you.”

    We are asking fundamentally different things of the user, which could have an impact on their behavior.

    2. You give them more places to click by showing a larger image, which is more representative of a real search, and probably was a reason for the greater diversification in the click distribution.

    Between both studies, it is also clear that there are a few things that seem to attract substantial clicks:

    – Good images
    – The map on the right
    – The upper left carousel results

    Excited to see what others point out!

  8. It would be nice to see where the user click on the next screen after the click on the carousel. In will access to the website or on the map / photo in the right?

  9. Top of that organic search has a lot of action. Not really surprising though. The data just goes to show you how important an image can be when it comes to conversions and click-throughs. I’m surprised to see how many clicks the actual map got. Almost no activity on AdWords?

  10. Question – where are the paid search results? They seem to have been disappeared with the carousel.

  11. Is the carousel US only?

  12. Interesting post Mike. Great discussion in the comments also. High quality images of food drive CTR unless ambiance if the center piece of an establishment’s marketability such as for a night club. Invest in a few good shots that make an awesome first impression.

    Learning how to optimize the images, both based on selection for carousel and how that images are sized can make a huge difference as Google cropping is not always optimal. Did yours or Matthews testing in this area yielding any useful findings that you can share?

    For reviews, number and quality of reviews are important with 5-10 creating a tipping point in our experience. Thanks for sharing.

Leave a Reply