Hailo: The secrets behind the Tech City startup's global success

Dave Gardner is an architect at Hailo, the taxi app that was born in London's buzzing Tech City but is now powering its way across the world. Having worked with startups throughout his career, ITProPortal was naturally keen to rack his brains on all things business, from original ideas to growing pains to what it really takes to build a successful company from scratch.

What do you do in your role as an architect at Hailo?

I'm responsible for leading our application design along a route that we think is good and sane. By 'application' I really mean the back-end systems. So as part of that role I try and persuade people to do things in what I consider to be a nice way. Like design, and how we should build high level systems, push best practices and have a hand in choosing technologies, for example Cassandra.

What is Cassandra and how do you use it to help Hailo?

Cassandra is a distributed database. At the moment in production we store customer records and some other stuff in Cassandra. In the future we're moving more and more of our data to Cassandra from relational databases like MySQL, and the reason we're doing that is because we want to leverage some of the features Cassandra has that make it attractive to us. Features like the fact that it runs on more than one machine, and is able to tolerate failure. So if one of the nodes stops working, the whole database doesn't stop working – you can still carry on. And it is also able to run on more than one continent. We have a cluster of machines in Europe, a cluster in the US and a cluster in Asia, and they send data back and forth between them to keep themselves replicated and up to date.

I know you've done a lot of work with startups before. What appealed to you most about Hailo?

It sounded like a real problem that real people might have, both from the drivers' side and the customers' side. Drivers spend a lot of their day driving round empty and - where we launched in London - they were having work taken away from them. People were moving on to more convenient services like Addison Lee, so we solved a problem for the taxi driver in that we could connect them to consumers who weren't necessarily in their line of sight, basically making their day more efficient and helping them earn their money quicker.

We also solve a problem for consumers, in that they can hail taxis that they can't necessarily see and they can have the convenience of being able to charge it against a credit card automatically without having to fish around in a bag at the end of a journey. In terms of what attracted me to Hailo, it was really just that the problem seemed real and exciting.

Hailo is arguably the most successful startup company to have ever graduated from London's Tech City, and recently picked up two awards at Appsters 2013. Is this because the original business idea was good or is there a lot more to it?

I think a lot of it comes down to that. It is clearly a good idea, and it was an idea in the right place and at the right time, in that we were the seventh Tech City app to launch in London. But now we're the biggest. So although it was a good idea, it wasn't necessarily an idea that nobody else had. It's just that we had a very good team that founded the company, with a lot of experience, both in terms of the taxi trade but also in terms of running startups. We have three guys who had successfully run very large companies and startups. Putting those two elements together, I think the execution was good, basically. The management team was able to execute well, and therefore we were able to take a commanding position in London and expand.

How has Hailo found the process of trying to expand globally?

I can only really answer from the technology perspective because I'm not really involved in the more business aspects. From a technology perspective, there are a number of challenges. We now operate in 15 cities on three continents and we want a seamless experience for consumers, so that they can hail a taxi in New York to the airport, get out in London and hail another taxi straight away in London. We needed locality. We wanted it so that if you're in New York, your phone would connect to a server in the US, and similarly in London.

There's some challenges in terms of operating computer systems that span three continents, and there's also the challenges of growth, in terms of scaling up systems and processes so that they can cope with lots of things, like the number of people trying to add features. If you're operating in Japan there's a lot of localisation work and a lot of requirements that flow in from Asia. I think we have about 60-70 developers now. We need to scale our processes in terms of allowing all these people to work effectively but also scale up the systems to allow for a greater volume of requests. When you've only got one city with a few hundred cabbies, some things are quite easy, like reporting. But when you've got a system with the number of things going on that we have going on, those sorts of things become a challenge.

What is the best piece of advice you would give to a budding startup?

When I did my startup, I wish someone had told me to focus on making something useful for other people and persuading them to part money for it. If that is a slick technology product, great, but it doesn't have to be. It doesn't really matter how good your technology is, if people aren't buying it. You have to get traction. You need people to want your thing. Sometimes that involves listening to people but you also have to know when to keep it simple. You can achieve a lot with a simple product.

I did a startup way before I started at Hailo, and I spent far too long building my product without actually trying to sell it. Without really going out and doing – as a technologist – the 'terrible' job of having to go and actually speak to human beings and try and persuade them to part with their money for my product.

If I was starting a startup tomorrow, I would be challenging myself to get my first customer. And then get my second customer. And almost do the minimum amount of work required to do that, rather than spending three years building something and then discovering that no one wants it.