People are increasingly asking assistants the questions they used to type into a search box, and in fitness those questions are unusually specific. Which gyms in a particular neighbourhood have proper strength equipment, what does a given class format involve for a beginner, roughly what does a membership cost in this city, is there a studio near a particular office with evening classes. The answers name a handful of businesses, and which ones get named depends on data the business often does not control directly: profile accuracy, review recency, how the facility is described on third party listings, and whether the site actually answers the question in plain language. That is the surface we work on, alongside the classic local search work it overlaps with.
Local recommendation answers behave differently from product ones, and fitness sits squarely in the local category. When an assistant is asked for a gym in a specific area, it assembles from map and profile data, review text, aggregator and directory listings and any local roundup content it can find, rather than from a brand website alone. That has two consequences. First, the off-site layer matters more than the on-site layer, which inverts the usual instinct to fix the website first. Second, being absent from aggregator and directory ecosystems can leave a facility invisible in these answers even when it ranks acceptably in conventional search. The query shapes are distinctive too. Proximity questions carry conditions such as parking, timings, women-only sessions or a specific piece of equipment, so those facts have to exist in retrievable form somewhere. Format questions are answered from explanatory content rather than listings, which is where a facility with real coach-led guides has an advantage over one with only a booking page. Pricing questions are answered vaguely because so few operators publish any structure. Assistants also hedge on injury, medical and outcome questions, so content respecting that boundary is treated as more reliable.
