My metrics depend on the source. On my own website, I could theoretically see precisely what’s happening. I’m too lazy for that, so I just run a DB query to see how many carts have things in them without ordering vs. number of orders today

For Doordash/Uber Eats/Toast/Etc. carts, I generally get basic stats like % of carts abandoned (someone added food items, but never ordered from anywhere), and % of carts ordered elsewhere (they added items from multiple restaurants, then didn’t order from mine)

They give me a baseline to compare against, but I think it’s a national average, which isn’t useful. When I signed up, they gave me a few hundred dollars in credits, so I applied them towards offering free delivery for the first week or so, which is the main metric I compare it to

Broadly though, I agree with you. My business partner always looks as those as all potential lost customers, and I have to remind him plenty of people are just comparing multiple restaurants. I do also wonder how many are just bots as well