It’s the age-old debate about the feasibility of merchant data and how to keep it updated, really.
I do think that automated crowdsourcing is probably the answer (as Emma do for merchant logos) but this only goes so far as you can’t use it to correct maps without manual intervention, or to tidy up merchant names when they are formatted completely incorrectly.
Edit to add further thoughts:
Also, that doesn’t even account for all the issues caused by franchisees using different trading names on their terminals, businesses that may be running inconsistent software versions (and reporting different merchant names as a result) or having to link various “duplicate” merchants together to get an accurate picture of spending.
J Sainsbury, Sainsburys and Sainsbury’s should all report as the same shop, for example, as should John Lewis and John Lewis & Partners. Amazon is also sometimes Amazon, sometimes Amazon Marketplace, and sometimes something like Amazon EU Sarl. Those are just the examples I could think of immediately, but there are hundreds of cases of things like this.
That said, I would assume that work to improve merchant data is going to be needed anyway, as others have said in this thread, for Trends to work properly. So if you already need to clean up the data for that, it’s a shame not to also do the classic Year In Monzo. It’s a great marketing opportunity for social media, but it is also a good way to build customer loyalty as customers enjoy seeing the data - and since you’ve done in multiple times before, it’s become expected, like Spotify Wrapped, so it is going to disappoint some by not being there this year.
Somewhere, deep in Monzo Towers, and while a single tear falls down their face, a special projects team is slowly ripping up the contract they were poised to sign with Pret as anchor member of relaunched Monzo Points.
Why does steak rhyme with stake and bake, and not with streak and teak?
English must be bloody awful to learn as a second language. I’m currently learning Welsh, and although every word is seemingly 24 letters long with no vowels, once you understand the phonetics, it’s actually pretty rigid.