Economists / anti-trust lawyers (at least in the US) look at a butt-ton of variables, and there is significant literature on the topic (source: I have a graduat

Economists / anti-trust lawyers (at least in the US) look at a butt-ton of variables, and there is significant literature on the topic (source: I have a graduate degree in Economics - and love to talk about this type of stuff,) but just some of the top variables included: 1) Market definition (to your point - in the tech world, this can get problematic very fast.) 2) Market share 3) Market share rank (i.e. 4% among a million competitors who all have <1% would be pretty significant) 4) Ease of entry into a market (i.e. Google gets a lot of scrutiny in the search market because of the consensus that to build a good search engine competitor, you need a lot of data for machine learning models, which a startup can't just get $ to buy and/or cleverly build with IP.) 5) Ease of customer switching 6) Pricing behavior ...among many others. Historically, a lot of analysis & literature to demonstrate market dominance has been on elasticity of demand: e.g. if I as dominant company X adjust prices, and customer demand doesn't change (probably because there are no substitutes and/or the switching cost is too high,) then anti-trust regulators would probably be all on my ass. But in the tech world, new models are still being developed I think, as that concept is not 100% applicable.