Custom services as ranking signal

[JS Girard]

I've seen whispers and mentions of ongoing test for almost a year on this topic, but no one seems to have actually published on this, so I'll share my conclusion, mostly in bullet point because I've not ran a long, systematic test with documentation, just tracked a couple edits here and there

1. Custom services appear to be a signal, but a weak one
   • Name signals in particular will usually drown custom service signals.
2. They are only really interesting as long as Google does not know what verticalis relevant to the keyword.
   • In Quebec, I can tell in some industries whether this is the case by whether the keyword triggers the LSA display (vertical is known) or the traditional finder (vertical is not known). This may not be functional everywhere
3. Once Google DOES associate the keyword to a vertical, the main category signal will compete with the services signal (but not necessarily drown it completely the way name signals do) and the GBP will lose ranks.

I've seen it happen "live" in spring 2023 for "nettoyage thermopompe" (yeah we do mostly French SEO, sorry about that) which was unassigned then became associated primarily with with HVAC companies. Our client is a duct cleaning business and felt from first to 4th position. For comparison "nettoyage climatiseur mural" is still unassigned and shows a greater variety of categories, with our client in 3d position.

In other words, the more unpredictable/varied the displayed categories are, the more valuable custom services become.

 

[JeffClevelandTN]

@JS Girard Thanks for taking the time to document this and post your findings. We've likewise experimented with this for roughly a year and experienced almost identical results as yours, but did not document it as thoroughly as you have. Thanks for sharing!