James Rubinstein
1 min readDec 2, 2020

--

Nice post, Daniel! Here are a couple of examples I like to use for broad and ambiguous queries that help me think about it.

"Spider Man" is broad because it can mean a bunch of different things: movies, comics, even pajamas. It's the sort of query I'd enter if I were shopping for my nephew who is a big Spider-Man fan. I don't know what I want, but I'll know when I see it.

"Stiletto" is an ambiguous query because *I* know what I mean by that, but the search engine does not. I might mean "knife" or "shoe", but it's *very* unlikely that either is going to satisfy my need equally.

"Mixer" is a pretty good example here, because I might mean "Kitchen aid" or I might mean "Tascam". It's very unlikely that I mean both equally. I mean one or the other, I just didn't tell you (the search engine).

"Shirts" then is a bit broad, but it's also a bit ambiguous. I probably mean "mens shirts" or "tshirts", but either could fit the bill. I almost certainly don't mean "women's shirts" though. So for shirts, several categories may fit the intent, and several definitely do not.

--

--

James Rubinstein
James Rubinstein

Written by James Rubinstein

Search nerd, data nerd, and all-around nerd-nerd. He has worked at eBay, Apple, and Pinterest, and currently leads the Product Analytics team at LexisNexis

Responses (1)