Zamano’s CTO Ross Conlon talks about striking the right balance between innovation and maintenance with IT, and reveals how a revised IT strategy shaped the development of its latest mobile marketing product, Message Hero.
Can you tell us about your own role and the responsibilities you have at Zamano?
I’m the executive-level owner of all IT aspects of Zamano, and that would include responsibility for the high-availability platforms, the development team, internal IT – basically everything. It would also include the data strategy. We’re a development company, so it includes the IT development strategy going forward.
What is that strategy?
We have 30bn rows in our database. What we’re doing now is, instead of trying to develop modelling on that, we’re using external tools as much as possible. We use a tool, for example, called Tableau which is a massive analytics engine. We get a lot of data in every day and Tableau is building behavioural models around that. It gives us a feedback loop in the business.
We get real-time feedback on how people are interacting with our platform. So you tweak things, or you can predict future behaviours from small amounts of data. It’s kind of like driving a car with your eyes closed – you can look behind you in the rear-view mirror and see if the road is turning. You’re getting in small bits of data every day, and you apply that to what you already know – which is your rear window. You’re using the data that you have.
What kind of an impact has that had on the business?
We wouldn’t have a business if we didn’t have this, because you’re always adjusting, based on the facts in the data. People probably wouldn’t really see it, but a lot of our margin is based on these adjustments. Things change so quickly that if you stay stagnant, you might as well be walking backwards.
How extensive are Zamano’s operations from an IT perspective: how many users are there, and how extensive is your office network?
It’s actually quite small. We outsource all the non-core stuff. We outsource our database [administration] function to a company in Dublin, DNM, but we keep our development teams – they’re employees of the company. We’re running a completely outsourced model. We have a small team that just manages external resources.
What more can you tell us about your approach to outsourcing?
Our platform, as of this year, is fully up on the cloud. It’s done with Digital Planet in Dublin. We view them as partners. We’re not just customers; they’re almost part of our team. There is a clear demarcation, but it’s not a case of we tell them what to do. They advise us, they look at our platform and make proactive suggestions on what we should so.
We try to do nearshore outsourcing just because for size of company we are, trust is a pretty big word, and it’s good to see our providers face-to-face if there’s any issues. But also, we want to be a big fish in a small pond. We don’t want to be in a queue and someone doesn’t know how important a problem this is for you.
Was there maybe an element of supporting local companies in your decision or were you dispassionate about choosing which service providers to use?
We were totally dispassionate about it. We wanted the nearshore model – there’s obviously a function of cost but it’s also a function of the benefits, as well – and it was all in and around same price. But it was the benefits that swung it from us. We aren’t experts in areas where we need these resources.
When you’re relying on partners to do a massive amount of work for you, you’re handing over a lot of responsibility, so you really need to trust them. We call them a lot. We have access to the CEOs [of those companies] if we want – it does give you that comfort.
There’s been a big drive to standardise and automate delivery of IT as much as possible – is this a trend you’ve embraced, and if so, how?
What we’ve tried to do is, we use a standard database. So there are lots of tools we can get online that help us to manage that. When we don’t have to develop something, we won’t. We use a lot of SaaS [software as a service] tools. Several years ago, we would have developed these things internally. It’s like a recipe for our company, with different ingredients that we put together to suit ourselves.
How much of your budget is consumed with just maintaining the systems you’ve got, and how much goes towards doing innovative things that will push the business forward?
It’s massively swayed towards innovating now. In terms of the maths, I’d say in or around 25-30pc is in terms of just keeping the whole thing running. We’ve got retainers on call, because we have a 24-hour support agreement, and we employ the resource who maintains relationships with our outsourced partners. So, it’s about 70pc on innovating and developing.
Do you think this ratio will change, or is that as good as you hope to get it?
To be honest, I’d say that’s what it’s going to end up. Obviously, cloud computing could end up getting cheaper, but we’re pretty happy with that.
How long did it take to get to that split?
Since maybe 2010, I’ve been working on dramatically changing our IT model. We used to own our own servers and now we basically rent resources – that might be servers or people – so it might even have been as high as 50pc before.
What can you tell us about Zamano’s IT strategy: what objectives have you set for the next year?
We’ve moved everything into the cloud this year, though we do have one or two office servers still to move – but that’s minor. We’re now in development mode, developing new products and services so that’s our objective now. We’ve sorted a lot of the IT issues. The way we’re trying to do that is using the Lean Startup methodology by Eric Ries, where you do a prototype of something and you get it out as fast as possible, and you innovate on that or you pivot based on the feedback. It would be considered an inexpensive way to innovate.
You want to create a situation where you’re getting feedback continuously from real people. You discover surprising things in the feedback – stuff you never would have thought of. You know you’re aligning to the market very quickly because you’re getting that feedback on a continuous basis.
The company is growing right now, and looking to expand – does that mean you get an increased IT budget to support that, or do have to work with the ‘do more with less’ approach?
I’d say we’re probably going to have same budget, but we’re using a lot of external open source projects and so the result for us is that what would have taken a few developers a few years ago to build, we can do now with one. We’re using a lot of libraries. It’s like a recipe or jigsaw. I see our developers more as connectors of APIs or tools.
When you’re mapping out the technology infrastructure, do you need to factor in the planned growth of the business, or is that issue off the table now that you’ve moved to cloud which can scale to whatever you need?
It is [off the table]. We factor it so we could scale horizontally. It was historically a massive concern, but it’s gone to nothing now.
You recently launched a new product, Message Hero. Given what you’ve told us, were you able to be faster to market, and use fewer resources than with the old way of developing?
It’s essentially text messaging for small business, we saw how people were using it and we decided to go to a full launch. A quick test, see that it works and get it out there. We did this under the Lean Startup methodology and we got a beta out within two or three months.
In terms of development, it might have taken us six months beforehand, and it was done to significantly higher quality because we were using libraries. It’s almost incomparable. You’re talking about a better product in half the time.
Is part of your role to understand technology trends and identify opportunities for new products and services that Zamano can provide?
It is ultimately led by what the customer wants, but you have to be aware of trends in the industry. I’d say technology underperforms over two years but massively over-performs over six to seven years, so you can’t ignore megatrends: they will ultimately happen.
But we’re looking at growth of mobile browsing, which is rocketing at the moment. People like us are using online tools a lot, rather than developing things ourselves. We use Gmail, Wonderlist, and Tableau for reports, where historically we would have developed these things internally.
The result is that we are spending more time focusing on our core. Our resources are very much aligned to where they need to be, whereas someone in the past could have been working on a project that wasn’t revenue-generating but was one we needed.
What’s been the biggest change that you’ve noticed about technology over time?
Ten years ago, something had to be just functional. Five years ago, it had to be easy to use. These days, it has to be functional, easy to use and people have to like it. Technology is getting more fun – more of an experience. It’s getting us to think about technology in a different way. You have to focus on the engagement with the customer, the interfaces they use, the experience and the behaviour that we have.
Most of the development on Message Hero was on the front end, not back end. Years ago it would have been the opposite.