I’ve added some more thoughts on how this stuff applies to local SEO on SearchEngineLand.com.

UPDATED 05/29/24:
I went through all the data last night and found that 99% of this data is for map nerds. There is very little here that gives us any indication of local SEO. But that is not to say there is nothing here. Rather the limitations of using ChatGPT to summarize these documents. For example, https://hexdocs.pm/google_api_content_warehouse/0.4.0/GoogleApi.ContentWarehouse.V1.Model.GeostoreFeatureProto.html has a lot more information than shown in our summary, so check out the links in the GSheet below to dive deeper. Here are the few things I found interesting about our data:

1. 1. Meaning
      “The relative importance of this category for this feature according to the data provider, as one of the values ​​from the above list. Importance is a measure of how well the specified GConcept describes the feature. If the meaning of this GConcept is unknown, this field should not be set.” ChatGPT left out this critical part of the original document: “An example is a gas station with a supermarket and an ATM. All three GConcepts are very relevant, but the gas_station GConcept is the most important.”
      
      How can you use this?
      A “GConcept” is, as far as I know, something like a type of business or a business category. In this case, “prominence” is the relative importance of each category when a business has multiple categories. Using primary and secondary categories can have an impact. For example, if you are not ranking well for your primary category, Google may think the secondary category is more important.

1. 1. gobiSite
      “Whether it is a Gobi site, that is, a site from a Gobi domain that should be boosted for a category query using that Gobi domain. For example, amazon.com is a Gobi store domain for a category query [hdtv] but some sites (like askville.amazon.com) from amazon.com should not be boosted.”
      
      How can you use this?”
      “Gobi” is the word for “cauliflower” in Punjabi. It can also be used as a mild insult. Urban Dictionary says, “In English, it sums up quite well as ‘fat bastard.'” So I suspect that this refers to sites that try to cover as many topics as possible. This may be Google Maps’ version of the Site Reputation Abuse algorithm. If you’re a cauliflower, you may need to put that subdomain on a new domain.

1. to count
   “The number of web pages that contain the URL in their breadcrumbs”
   
   How can you use this?
   This is in a document about SiteLinks. It may underline that sitelinks need to be placed in breadcrumbs on as many pages as possible in order to appear in the SERPs.TLDR: These documents are interesting, but not very informative so far. Stay tuned!

ORIGINAL POST

As expected, SEOs are studying the Google API docs that leaked over the weekend. For more information, check out the posts by Rand Fishkin and Mike King.

I thought it would be interesting to summarize all the documents relevant to “Local SEO” that we can find.

Here is a summary of 271 Google API documents and 461 attributes that we identified as relevant to local SEO. We identified various terms that implied “local,” then crawled them with ScreamingFrog v. 20 connected to ChatGPT and had each document summarized.

Access the GSheet directly at https://docs.google.com/spreadsheets/d/1kFg8QJ6IVUcGRyrK5bHZehRp9Yl7vSpjSeuatsJBjfU/edit#gid=0

We’ll be analyzing these documents over the next day or two and summarizing the data. If you find anything interesting, feel free to DM me on LinkedIn and we’ll add it to the sheet.