It's not conceptually that difficult a task, but the devil is in the details. I'd encourage you to find a local ham radio club or maybe a makerspace that has some RF-oriented folks, and see if some of the members there are willing to help.
But loosely speaking:
1. You need a spectrum analyzer. It doesn't have to be a real lab quality, bench spectrum analyzer, but you need something that can measure power vs frequency.
2. You could possibly get by with an RTL-SDR dongle and the right software.
3. Given something that can do the measurement in question, look up the frequency band(s) used for cell transmissions. Go outside somewhere in your area, where cell phones work fine, and measure the power spectrum around that frequency band. Figure out what the normal range(s) are.
4. Go to the store in question and repeat the process.
5. Note any frequencies where there is a large delta between what you measured outside and what you measure inside. If such a thing is present, there's a good chance that that is the source of the problem. It's not a guarantee, and there are other possible explanations, but finding a big spike of energy in the same range that phones operate on would be highly suggestive.
6. IF (5) turns up something, finding the source involves something called "radio direction finding" which is another big can of worms. Again, a ham club or something similar would VERY likely have members who are experienced doing this sort of thing and who would probably be willing to help. But if you can't find any help, search the HN archives for "radio direction finding"[1] and similar terms. This has come up before and there have been some good discussion with lots of pointers to useful resources.
There's a bit of interesting stuff in this post:
https://news.ycombinator.com/item?id=28054017
and some more here:
https://news.ycombinator.com/item?id=27857290
[1]: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...