Browsing customers mechanic

"The research on this page is old and might be out of date. Please do consider updating it, expanding it or adding your knowledge to it. WikiPim"

The browsing customers in the shop have always been somewhat of an annoyance for me. The times when you click on them and they say "leave me alone." makes me want to sic my thug on them. However, I always assumed they did have some effect on the game beyond irritation. No more. Now I want to know.

The current idea of browsing customers is that they come into the shop, look at something they're interested in (maybe, or just the wall), and come back the next day, possibly the day after and ask for an item you actually have.

This is a sort of nebulous function, as there's no real way to see if a customer asking for an item is the exact one that browsed the previous day. But if we can find some correlation, that is at least a step in the right direction.

I actually tried doing this a few days ago, though not all that rigorously. I recorded the browsers that came in, tried to identify what they looked at, and if they left angry. I didn't record every customer asking to buy or for an adventure item, only looked for customers who might have been browsers. Still, I didn't find any correlation over 4 in game days.

I plan to do more thorough research soon, below is the method I'll be using. It will be rather tedious and time consuming.

How you can help
In order to do research into this mechanic, I propose the following test. For 5 in game days, record all browsers and customers (ones that ask to buy items and for adventures) in your shop. For browsers, note their class, what they looked at (specific improvement if you can tell, if not location), and if they walked out cursing/frowning. For regular customers or adventurers, note their class, what they were asking for, and if it was an item you had in stock. Record any data you collect below, in any way that makes sense to you.

Data
I recorded everyone that came into my shop for 13 days in game. After reviewing the data I can see no pattern to suggest that browsing customers come back to buy or sell items in the area they were browsing. My current theory is they affect some sort of background stat measuring store popularity rather than directly predicting what you should make next.