Why we don’t have a Business Intelligence team at TransferWise
At TransferWise we don’t have a Business Intelligence team, but that doesn’t mean we don’t have business intelligence. In fact, our work is incredibly data-driven. The way our analysts operate is pretty unique — it works. Our aim at TransferWise is to ensure that analysis translates into meaningful impact. Fast.
The analyst’s greatest fear isn’t what you’d expect
Like all analysts, we worry about getting the numbers right. But our greatest fear isn’t messing up a formula — it’s a much worse scenario: working tirelessly to produce cutting-edge insights, only for them to be ignored by the rest of the company.
When hard work and interesting conclusions don’t translate into meaningful impact, things can get pretty frustrating, pretty quickly. This scenario usually starts with the best intentions. An analyst will crack a burning problem, and send round a flashy new dashboard or mind-blowing presentation — but for some reason, after reading these insights and clapping the analyst on the back, no-one does anything differently. Did no-one understand the information? Did they simply not act on it? Or perhaps the wrong questions were asked in the first place?
With this in mind, we organise our analysts differently
At TransferWise we don’t silo our analysts into a separate ‘Business Intelligence’ department, as if thinking can somehow be separated from doing. Instead, all analysts work within truly operational teams — sitting side-by-side with product managers and engineers. Whether it’s within our operations, currencies or our performance marketing team, the analyst’s goal is to help their team maximise impact. (At TransferWise we work in autonomous teams — read more about that here).
Since a team’s impact is measured by its chosen KPI, an analyst’s goal is the same as everyone else’s in the team: do whatever is needed to have a positive and sustainable impact on this KPI. This guiding star helps analysts prioritise their work — the clearer the end-goal, the easier it is to tell whether a piece of work is having an impact, and the quicker, less impactful work can be de-prioritised.
The offence is where we have an impact
Analysts are free to work on whatever projects they believe will ‘move the needle’ on their team’s KPI. One piece of guidance we give at TransferWise, is to avoid getting stuck on ‘defence’. We see this as setting up the data infrastructure, preparing and running reports, fixing bugs, or adding endless new slices and metrics to dashboards. While it is essential to spend some time on defence, we find that often this is not where our analysts and data scientists have the most impact.
We prefer our analysts to be on the ‘offence’. For us, this means using data to find novel opportunities to make TransferWise better for our customers. The main challenge here is ensuring that any findings translate into action. But since our analysts are deeply embedded within their teams, often the whole team help to prioritise these ‘offence’ analyses. This means they’re waiting with anticipation to use the result — meaning very little work is ever wasted.
Staying networked is important
Despite working in separate teams, our data scientists are still networked. We learn from each other, share insights, help each other with the most difficult problems and rotate around teams. One way we share knowledge is with regular ‘analyst hackathon’ days. We gather in one place and collaborate on ‘offence’ topics that cut across many teams. Hackathon success is measured by the positive impact any findings have for our customers.
Helping teams help themselves
An important part of not having a Business Intelligence department, is ensuring that analysts do not become ‘bottlenecks’ for data. At TransferWise, our aim is that all individuals are able to use data to answer questions — with or without the help of our analysts. We make our data freely available to everyone that wants it, and train them in how to use it effectively. In this way, our analysts act as mentors, and use their expertise to take on only the toughest questions.
In fact, everyone needs to be data fluent at TransferWise. Our Product Managers, developers, marketeers, Operations and Customer Support agents all need to be able to work with their own data. As a result, no team waits around for someone else to ‘run the numbers’. A happy byproduct of this is that our analysts spend almost no time performing tedious data pulls or updating reports for others. This enables our network to be extremely lean (currently 8 of 500 employees) and focussed on high-impact work every day.
So, does it work?
Yes, it does.
To add some colour, here’s a recent example of impactful analysis from our Experience team, who focus on making the experience of moving money convenient and hassle-free.
We do our best to deliver a seamless online experience for customers, but sometimes they need to get in touch with us. Understandably, when a customer has sent us hundreds or thousands of pounds, euros or dollars, they expect their queries to be dealt with quickly, simply and transparently. So Marko, an analyst within the Experience team, decided to figure out how his team could improve the experience of those customers that we need to speak to.
The Experience team tracked how many of our transfers require a manual intervention on our side, but Marko realised that this wasn’t really what customers cared about. What matters to our customers, is whether we get their transfer around the world quickly and reliably, ideally without requiring any extra work from them.
So Marko did some investigating, and realised that only a small % of transfers require our Customer Support team to reach out to customers, but a few tricky transactions take up the majority of the time spent calling customers. Marko was able to list the major reasons we needed to contact our customers, and took these to his team.
Marko’s team were awaiting his findings, and together they worked through the list and identified the “quick wins” for our customers. One of the main reasons we need to call our customers, is that sometimes they transfer us a different amount to the one we were expecting (a ‘mismatch’). The Experience team quickly built the option to resolve these mismatches through ‘self service’ online, removing the need for a lengthy phone call.
The adoption of this self-service facility by customers was rapid (chart below). Marko and the Experience team can clearly see the positive impact they’ve had for customers, and move on to tackling the other items on Marko’s list. This type of great result happens every day in teams that are organised for impact.
Organising for impact
‘Business Intelligence’ is only valuable if something happens as a result. If nothing changes for your customers, then there’s simply no point.
At TransferWise, each team builds business intelligence into everything they do. This allows analysts and data scientists to focus on the toughest questions: those that have the greatest impact for their teams and our customers.
If you’re interested in joining our lean, embedded, impact-focused analytics team, please check out our jobs page.
For more from the TransferWise team, follow our Medium publication TransferWise Ideas.
You can also follow TransferWise on Twitter here.
