Geotagging Images via API

[Tim Kahlert]

Hi!

We all know that Google strips off Exif-data from photos uploaded to GMB which leads to no effect on rankings etc. I've also seen a post by @JoyHawkins where she tested the effect of low-competition keywords in file names with no results on rankings either.

However, some people in my local SEO community believe that geotagging photos (on posts) via the LocalViking API lead to expanded ranking areas. (an increase in the areas they're targeting through keywords and location data inside image files) Due to a lack of reliable/valid experiments and significant results, I'm having a hard time believing this.

That´s why I would like to test this as reliably as I can.

My plan is to post at least 4 times a week using two keywords and two locations (2 posts each). I'm using low-competition keywords in areas where my client´s ranking is average (#5-#20). It´s certainly easier to see improvements on the rank position spectrum #20-#5 than #5-#1.

What else would you consider critical on a test like this?

Thank you!

Tim

 

[JoyHawkins]

I would say it's critical that the businesses aren't doing anything else that could be influencing their ranking. If you're optimizing the website or building links, for example, you can't properly isolate what was responsible for the increase.

I'm actually in the process of testing this one myself

 

[Tim Kahlert]

I would say it's critical that the businesses aren't doing anything else that could be influencing their ranking. If you're optimizing the website or building links, for example, you can't properly isolate what was responsible for the increase.

I'm actually in the process of testing this one myself

Thanks for the quick reply, Joy! That´s a good point. This is exactly how a test needs to be - not influenced by any other factors. I think I will go ahead and track the rankings for a bunch of keywords for a while (maybe 2 weeks on a daily basis) before I actually start the test. Then I can choose the ones whose rankings have remained unchanged and have not fluctuated too much.

I'm looking forward to your test results as well.

 

[JoyHawkins]

Tim,

I'll be doing a big presentation on images at some point next year but the TLDR of the test that I just finished yesterday on geo-tagging images is that it has no impact on ranking.

 

[Brock1321M]

We have done some testing on Images, and what I see if Google has gotten much better at understanding photos and identifying what exactly is inside the photo. I think it's a great way to build product/service and local relevancy. I don't know if Joy or anyone else has tested this but would be cool to see others' opinions.

 

[JoyHawkins]

Yes, we've tested this as well and it's going to be a part of that presentation on images. For some types of businesses, images really matter and what is in the image. Based on my testing, I've found it's not uniform across every industry so it really depends on if Google is showing images in the search results.

I've also only seen it matter for organic (so images on your site) and I've never seen it for Google My Business photos.

 

[Tim Kahlert]

Thanks for sharing your insights, Joy! I will post the results of my testing in a few weeks here as well.

 

[Kaycie Smith]

Hi Tim,
Curious how your testing of geotagged images went.

 

[JeffClevelandTN]

Yes, we've tested this as well and it's going to be a part of that presentation on images. For some types of businesses, images really matter and what is in the image. Based on my testing, I've found it's not uniform across every industry so it really depends on if Google is showing images in the search results.

I've also only seen it matter for organic (so images on your site) and I've never seen it for Google My Business photos.

Hi Joy, will your presentation also include any organic impact made by use of captions and alt-text? Looking forward to it!

 

[JoyHawkins]

Hi Joy, will your presentation also include any organic impact made by use of captions and alt-text? Looking forward to it!

Yep! I'm going over the impact we saw in both local & organic rankings for several different strategies involving images. I learned a ton with these tests.

 

[Tim Kahlert]

Hi Tim,
Curious how your testing of geotagged images went.

Hey Kaycie!

Here's a link to the case study: The Truth About Geotagging Images On Google My Business | Tim Kahlert

 

[Kaycie Smith]

Thank you @Tim Kahlert

 

[JeffClevelandTN]

Hey Kaycie!

Here's a link to the case study: The Truth About Geotagging Images On Google My Business | Tim Kahlert

Thanks Tim for sharing your case study!

 

[JoyHawkins]

I finally published the test we did on geotagging and it's listed here: Does Geotagging Photos Influence Ranking? - Sterling Sky Inc

 

[Lluc]

Thank you Joy for sharing the results of your test on geotagging photos, a great resource to refer people to when they insist that “geotagging is the secret to increasing local rankings”.

Every once in a while, we have also tested the impact in rankings of geotagging photos, specifically for hotel listings. We experienced no noticeable change in rankings or we were unable to attribute changes specifically to adding geotags/EXIF data to the photos uploaded to GMB.

However, we did notice that certain photos maintained the description field. This text is then shown on Google Hotels (The “owner” photos come from Google My Business). This has no relation to adding the coordinates to the photo but wanted to share it.

Google Hotels / Desktop / Photo section

When you click on “see more” link, the full text appears and takes up a considerable amount of space.

When you click on the actual image, the text is displayed on the top of the image.

For this client in particular, we added the same metadata to 10 new photos but the description text only appeared in 3 of the photos that had metadata. We did the same test with several other clients and the results were similar. In many cases, no description text appeared once uploaded.

The text we used was unique content, but it was the same text on all the uploaded photos (per hotel, of course) with all the Exif data we could provide.

Examples of other hotels.

On Mobile

Nothing happens when you click on "Ver más/See more".


We think that this is a bug from Google but it is still interesting to see that some of the meta-data is kept and used, and not completely stripped off.

We tested this pre-Covid and have not tested any further since then. Our initial idea was to use keyword rich content, hence the long text, not knowing that the text would show up on Google Hotels/Photos. Since then, we have hypothesized about which type of text we should use, thinking mainly from a conversion or brand point of view. We have not tested anything yet but would love to hear any comments, ideas or suggestions!

 

[JoyHawkins]

Thanks for sharing Lluc!

 

[JeffClevelandTN]

Hi @Lluc, this is super interesting!

You mentioned the EXIF text used in the pictures was unique content and used on several pictures. I was curious if the content was unique to only the pictures, in other words was it ever used, for example, as content on any webpages or any Google specific property other than images? I did an exact google search for text that was showing on a couple of your screen shots and found matches on some web pages. You mentioned this was done pre-covid, so the content may have been re-used since then for other purposes.

The reason I was asking about the uniqueness of the picture description text, it would be very interesting/valuable to know if Google indexed that text by searching specifically for that unique string of text. Did you happen to test that type of scenario and did it show up in any organic search or image search results?

I'm likewise going to try replicating your results by adding the appropriate EXIF to several pictures. Hopefully I'll have similar results where the text is visible on a handful of images. I may try to come up with a test like @JoyHawkins did with her use of descriptive alt-text in her case study, specifically the Michelin tires example.

Thanks so much for sharing!