Researching operations at Monzo: Building a product our customers never see 🔍

Hey everyone - I’m Luke :wave:

I wanted to share a blog I wrote that some of you may find interesting.

User Researchers at Monzo help us understand what different people need from our products, so that we’re building the right features in the right way. We don’t just apply those practices to our external products though: I’m a researcher in our Operations team, and work on a tool called ‘BizOps’ which is used by our customer operations (COps) team.

In the blog I’ve tried to shed some light on this unseen part of Monzo, and how user research helps us build an internal tool that’s every bit as good as the customer app.

If there’s any questions about what I’ve written or what I do here at Monzo, feel free to post them here and I’ll do my best to answer :blush:


Really interesting read @LukeK, thanks for sharing.


Thanks for swinging by @LukeK :wave:


All hail BizOps!

Such a good system, unlike other banks having 6+ account systems, we use 1 for the most part. :raised_hands:t3:


The systems I have seen other banks use would shock you Monzo are paving the way.


Chase’s current server room is advanced, with built-in operational redundancy:

For clarity, I am a minor Chase customer. Gratefully minor too.


One thing I’m curious about is, how much of this stuff is shared with or also used by the US team? Obviously the front-facing app is the same, which is nice in a way, but aside from US-specific stuff and stuff the US team won’t need yet because they’re not at sufficient scale, I imagine there would be a lot of commonality in needs.

1 Like

As I’ve understood, they’ve built the UK system from scratch* not using any of their legacy US platform.

(A lot of it is assembled from platform providers, like 10X, but you get the point!)

I’m surprised the app is the same / similar - can you point us to any screenshots of the US version?

A thought I had recently: I wonder why JPM passed on (or didn’t investigate) Pi1, the platform that Dozens runs on. It would have been a cheap acquisition and seems to me to be more robust and feature rich than what they’ve ended up with.

This line caught my eye:

Some of these tasks are relatively simple, like a customer wanting a reminder of their PIN.

Given that you can retrieve your PIN in the app and that :monzo: is only (mainly?) catering to those who have self-selected an app-only bank, presumably looking for ways to self-serve - or nudge towards self-serve - to avoid COp contact at all continues to be a main thread

I am sure it is passing me by on this overly warm day, but would there be any discloseable reasons why a PIN reminder would need COp involvement in some cases?


Sorry, I meant Monzo, since I haven’t even seen the Chase UK app (also because Monzo’s the one where everyone, UK and US, use the same app). It’s because of the “everyone uses the same app” as opposed to separate apps for each country that made me wonder how much is shared behind the scenes.

The US Chase app looks nice but doesn’t really look like the UK app screenshots I’ve seen floating around. I can edit this with some screenshots of the US app (with details scrubbed) later.

1 Like

Hi Jason :wave:

At the moment we route all US customers to a dedicated team, which BizOps does automatically as it can spot when someone is contacting us about a US account. There are definite common threads in terms of what people need, but our response is often dictated by regulation and law - which is obviously very different in the US.

One of the research pieces I’m working on at the moment kind of relates to this: I’m exploring whether you can define in advance all of the things a customer could realistically contact their bank about, as if we can determine that, then it gets much easier to match an issue with the correct response (regardless of what market you’re in).


1 Like

Hi Michael! There are indeed reasons… but sadly not disclosable ones. :zipper_mouth_face: One interesting aside though is that quite a lot of things aren’t even disclosed internally, as our FinCrime team keep a very tight lid on how certain backstage processes work. Which obviously makes it slightly tricky to research them. :smile: As a researcher I can usually get access to anything I need to know about, but I’ve definitely learned to tread very carefully when sharing findings.

1 Like

Not 100% related, but has any research been done on the quality of merchant data vs number of fraudulent complaints (or similar)?

1 Like