> This disproportionately rewards chains and already-central venues. Chains benefit from cross-location brand recognition. High-footfall areas generate reviews faster....

I think this is very likely false if you mean compared to the status quo ante. Before Maps, a well-loved but hard-to-find venue just wouldn't ever be seen by most people, and the absence of reviews made branding more important because it was all you had to go on. I'd be very doubtful if the proportion of independent cafes and restaurants decreases when Google Maps enters an area. (Couldn't find any causal research designs though....)

The more general point that the algorithm is not neutral (and probably never could be) must be right.

(I asked ChatGPT but it ended up with: "We have almost no clean exogenous variation in Maps rankings or feature rollouts at fine geographic scales that would let you estimate impacts on entry, survival, or market structure in a neat DiD/IV way.")

Before GMaps, we had the Zagat Guides, which were an important way for many restaurants to start pulling in traffic.

Who the hell cares what garbage chatgpt vomited based on your unspecified chain of prompts?