Joining Monzo as a Data Scientist

Read the article

Hi 👋 I’m Izak and I joined Monzo as a data scientist in Customer Operations at the end of February 2022.

Working in Monzo Data has been absolute ✨ magic ✨ and it’s definitely been the best step in my data science career so far. In this post I’ll explain why, and hope you’ll consider applying to Monzo Data too. We’re hiring Data Analysts, Data Scientists, and Data Directors!

🚴 My journey to Monzo

Like many other Monzonauts, my journey to joining is quite unique. I’ve worked in a few other data roles, including in management consulting and actuarial risk, but never at a fintech company. I didn’t know much about Data Science at Monzo but I loved the product, so when a friend (and now fellow Monzonaut) told me about Monzo’s approach to data, and the challenges they work on, I knew I had to apply. 

The interview process was fun yet challenging. I had an in-depth take-home test on metric formulation (which included extensive SQL and data visualisation), an interview with a data case study, and a collaboration interview. These all resembled real-world problems and at times it felt more like co-working sessions than interviews, showing me Monzo’s working culture which solidified my ambition to join. Luckily enough, I got the offer! 

So what makes being a Data Scientist at Monzo great?

🚀 Monzo is built on data

In previous roles, I often felt ineffective at my job. My teams were stunted by lacking fundamentals; things like a data-friendly tech architecture, the right analytics platform, and the ability to generate the data we needed. I’ve also seen excellent data work get lost in rigid hierarchies where org structures were ill-designed to consume and action on it. Or even worse, where work simply evaporated into thousands of unread emails.

But at Monzo, things are different.

From its founding, data played a core role in shaping both Monzo’s technology and its organisational setup. Generation and consumption of data are intentional and critical parts of the company. Every single engineer contributes to data generation. And every single Monzonaut accesses data insights to measure their own work.

And this means, the power of data (and the teams behind it!) is scaled to its full potential. 

💾 The data stack works

I’ve never encountered a setup where analytics scales as intentionally and effectively as Monzo. Our data stack is filled with millions of data points on all our backend and front end analytics events. We use BigQuery and dbt to transform all of these events into analytics tables and metrics. 

This all happens in community-generated and reviewed codebases of version-controlled SQL (my dream!), with the appropriate unit and integration tests to ensure nothing breaks. This means that Data Scientists work with the same level of code collaboration as Software Engineers.

Analytics are published to everyone (yes, everyone) through Looker, allowing anyone to pull their own data (no more annoying ad hoc data requests in your inbox!) Everyone looks at the same metrics, which are defined centrally. There are no data silos and there is a single version of the truth. It’s the data dream.

Not everything is perfect and from time to time a pipeline fails. But this doesn’t ruin anything and critically doesn’t become the personal responsibility of data scientists. We have a dedicated data platform engineering team who are amazing at resolving such issues and charting the course for long-term improvements.

But having the technical data setup is just half the story.

🧑‍✈️ The data organisation works

Monzo has a very intentional and thoughtfully-designed data organisation. It alleviates data scientists from wondering how to be effective at their work, enabling the most optimal use of data skills. To my fellow data Monzonauts this point may seem self-evident, but in prior roles I’ve seen how difficult work can be without a robust organisation framework.

At Monzo, data people belong to a central data 'discipline', but are deployed to specific domains, which we call Collectives. I’m in the Operations Collective, where we define, measure and improve everything about our customer service and underlying systems. My day-to-day projects revolve around everything to do with operations data.

If I need to, I can dive into the joint brain power of the entire data discipline - all other data people in other Collectives (such as Personal Banking or Borrowing), to help me validate my approach and share knowledge. This is incredibly useful because more often than not, someone has solved a problem similar to yours.

And belonging to the data discipline also means having a community of like-minded people to connect with formally and informally. We have monthly All-Hands discussions, bi-weekly learning sessions, weekly cross-discipline chats, and frequent socials. We’re a fun bunch 😀. The screenshot below shows one example of keeping in touch with the Data Discipline.

A slack post announcing the next data team fireside chat

🧪 Data science at Monzo has a clear scope

In some companies data science is a vague mix of data analysis, experimentation, data engineering, machine learning, forecasting or business intelligence. As Monzo scaled, we’ve found one person can’t possibly be an expert at all of these simultaneously. This is why more recently, data science at Monzo has been defined to mean data analysis, experimentation and impact measurement. 

For example, in the Operations collective we want to reduce Handling Time of customer queries, while maintaining the highest quality service. 

My role as a data scientist is to work in a multi-skilled product squad to identify potential opportunities to reduce handling time. I’ll estimate the potential impact of such opportunities, and feed this into the proposal. We then design and run rigorous experiments to assess how much faster we’ve made things (I’m finally using my stats degree! 🧑‍🎓 )

The role requires a blend of knowledge from business understanding and process mapping to more rigorous inferential statistics and coding. Different data scientists bring expertise in different areas. One colleague is an expert in experimental design from completing a PhD. Another has worked widely in different Monzo teams and knows our processes and their reflections in our data inside-out, while yet another is excellent at improving code efficiency and internal tooling (and keeping our Pull Requests in check!). 

It really is this blend of skills which make it such a great place to work and learn.

✨ Monzo’s culture is magic

Every company says this, but at Monzo you can really feel it. I certainly can’t articulate everything so I’ll highlight selected examples:

  • We really trust each other. If you have a bad day, or your family member is ill, or if you have an emergency, there is no judgement in taking time to fix things. We are humans, and we respect each other’s humanity. No one is watching your hours worked. 

  • Monzo defaults everything to transparency, whether it’s the reports sent to our executive team in an open Slack channel or openness about recruitment process

  • Monzo is really inclusive: we’re encouraged to use inclusive language wherever we can. We celebrate our diversity. Our London office has only gender neutral bathrooms and our company mascot, Hot Chip, is non-binary.

  • And my favourite: we’re hard on problems, not people. If a mistake is made we attribute it to the problem being difficult, not the person being inadequate. This is the single biggest shift from all my previous roles.

It’s only been 3 months, but you can tell I’m delighted being at Monzo. If you’re interested in knowing more about working here, please reach out on my LinkedIn, or have a look at our open roles.

Thanks for reading! 😀