Back to blog

Moving fast and embracing mistakes: A chat with our CTO, David Goaté

Natalie
Content Specialist

10.1.2021

-

8 min read


At Marshmallow, we know that being an engineer is so much more than just fixing problems and shipping code. Our engineers are at the heart of what we do, and their ways of working - shipping fast, iterating regularly, working autonomously, innovating our product offering, and openly learning from mistakes - have actually shaped our company culture and values. 

We chatted to our CTO David Goaté about why our engineers' ways of working are so essential to achieving our company mission, how our engineers learn from mistakes and the opportunities unicorn status will bring...

1. We have a very strong set of values here at Marshmallow. Do our engineers have their own values or guiding principles? 

In engineering, we’re driven first and foremost by our company values. For example, our value “move fast” can be seen in the way that we iteratively work towards small incremental milestones; use feature flags; seek to eliminate friction in getting from engineer work-station to production; and so on.

If you’re a Marshmallow engineer, you’re iterating as you go. We want to ship real products to customers quickly so that we can discover how they behave through learnings, rather than through guessing. 

But the important thing at Marshmallow is that the speed comes with velocity; we’re moving fast in a very specific direction! We have the user research piece upfront, and our product managers and designers engage with customers beforehand while also analysing data to help drive decisions. 

2. Iterating as you go is all about making mistakes and adapting quickly. How do we make engineers feel comfortable in making mistakes?

It comes down to psychological safety. Our “be open and honest” value is about creating the space for people to share their mistakes knowing they have the support of people around them. We encourage our engineers to assume the best of each other and to remember that everyone is working with good intent. We can’t achieve speed if people are afraid to make mistakes, as this leads to risk aversion!

We also have rituals that try to cultivate an open and honest culture around making mistakes. We run pre-mortems, looking at what might go wrong with a certain project and how we can avoid those mistakes. Having a considered conversation and making decisions together means that if those mistakes happen, we can all acknowledge that they were expected and just part and parcel of trying to move quickly and learn!

Then there are our blameless post-mortems. In these sessions, we try to analyse why a mistake happened in a very accurate and factual way. The focus is on what we can learn, not who is to blame. Postmortems help create an environment where we frequently normalise things going off plan. As a result, people don't feel fearful about moving fast and taking risks, which is key for innovation.

3. Lots of engineers would blame technical debt on moving too fast. What are your thoughts on technical debt, and how does it work at Marshmallow? 

Technical debt is a very divisive topic for engineers. But our guiding principles at Marshmallow steer us towards optimising for speed and learning. 

One important distinction to make is the technical debt you consciously accrue vs the debt you unintentionally accrue. We see consciously incurring technical debt as an enabler. It's a positive thing that allows us to build a product and deliver it quickly. 

Some might argue that this is the wrong approach. But a 100% perfect solution is unachievable, and chasing it leads to diminished returns. There’s a lot to be said for architectural simplicity!

The crossover between technical debt and our product roadmaps is what really matters to us. If there are problems that are causing friction or inflexibility in areas that we’re actively developing then we’ll address them. But, a lot of the time, we're consciously leaving pieces of tech alone that we know could be improved because it’s not part of an active product roadmap, and isn’t causing any significant issues at the moment. There are so many more interesting and more pressing things to be focusing on, things that will deliver value to our customers and our developer experience. 

In any company there are infinitely more problems than can actually be solved. We believe in being very deliberate about where we focus our precious time to achieve results for our customers.


4. As CTO, you must have to make a lot of tough decisions on divisive topics like technical debt. Thinking back on your time at Marshmallow, what’s been the most important lesson you’ve learned?


I’d say there have been two really important lessons. One is the importance of culture, not just in engineering but as a company. Our company culture is an enabler of our future success. I think our culture resonates all the way through to every engineer we hire. So it’s vital that the way people think, approach their work, and interact with one another supports and grows our culture. 

The second thing is something we’ve already talked about - the importance of speed. In the very early days of Marshmallow, we were working in an environment of uncertainty. We’d spend a lot of time building things without necessarily knowing we were building the right things. I thought we needed a more mature product to get off the ground. But you’d be amazed at how far you can go with so little! All we needed was a product with fairly basic functionality that we could iterate on. Seeing that changed my thinking. Now I'm a big believer in how much speed matters. 

As a startup or a scaling company, speed is one of the main competitive edges you have against more established companies who can put more people and money towards solving a problem. The ability to listen to what customers want and react to that more quickly, or the ability to iterate on your products and provide new things that no one's seen before, are the things that set us apart. 

A quote I recently came across, which I subscribe to wholeheartedly is “Fast gets good quicker than good gets fast”.


5. How do we make sure that the products we are creating at speed are built to a good standard?


We aim to hire the best. We make sure we have technically strong people in our teams who have good judgement. People who know whether a decision they're making about incurring deliberate debt is a good one. 

At Marshmallow, we're aiming to constantly improve so that we can better avoid context switching or shifts in direction that are unnecessary. To do that, we have to make informed and technically sound decisions that’ll keep us on the right track. So, we listen to the user research and always look at the data. It’s also really important that we’re using technologies we understand and know we can rely on to make sure we're moving quickly with confidence. 

Lastly, our culture of honesty supports healthy debate about trade-offs and the impact of technical or product deficiencies. One of the things I value most about autonomous, cross-functional product teams working towards a clear purpose is that everyone is pulling in the same direction and ultimately wants the same thing. It’s so much easier in this environment to get the right results. We can avoid a mindset of product vs tech, and instead focus on the mission of the team.

6. Fast growth is very much the story of Marshmallow at the moment, especially with our recent Series B funding. How do you see the role of engineering evolving as Marshmallow continues to grow?

I think our evolution will be really interesting. We're getting to a place where the foundations of our motor product are working really well. But there’s still a lot more we want to do there in terms of innovation - particularly around the claims space and how we deliver more value to our existing customers by making them safer and providing them with the best possible service and value.

But, as well as continuing to improve our motor product, our Series B funding will allow us to bring new products to market, introducing lots more challenges! We’ll be working towards leveraging synergies between our product offerings, whether that’s in fraud prevention and detection, or in customer self-service and automation. All of those things will continue to evolve in terms of bringing innovative products to customers, allowing for much fairer pricing, and creating a more user-centric experience.  

International expansion is also not too far down the road for us which will present an exciting new engineering challenge as our product becomes more global.


As our chat with David proves, it’s an exciting time here at Marshmallow, especially for product-focused engineers! If you’re looking for your next big opportunity, take a look at our current openings
here.