The five minute CIO: Lars Holzman

13 Dec 2013

Lars Holzman, manager of TripAdvisor's Dublin-based engineering site

TripAdvisor’s Dublin-based engineering site manager Lars Holzman tells us about the key to hiring a strong technology team, and why commercial and technical challenges are two sides of the same coin.

Last September, TripAdvisor announced its first European engineering hub in Dublin. What are some of the practical challenges involved in setting up the team here?

Creating a brand new team in a new location is always going to be a big undertaking. For us, the key priority has been to try and make as many potential candidates as possible aware of our presence here and the type of work we are doing. It always takes time to find and recruit top engineering talent, so that is what has been taking up most of my time, as well as setting up our office working environment.

I have been having some amazing chats with people in Dublin, both potential recruits and people who are just eager to help out, so I’m very confident we will have a strong core team in place by Christmas with further recruits coming in beyond the New Year. In terms of the other logistical bits and pieces that go into getting an office up and running, with the support of the community here, and in particular the IDA, everything has gone quite smoothly so far.

How do you handle liaising with a team in the US that’s operating in a different time zone?

First of all, one of the great things about Dublin is that the majority of our global engineering team is in Boston, so the time difference there is not that bad. That being said, there are a few things we are doing to further reduce the disparity: one, wherever we can, we try to reschedule global meetings that require Dublin participation so that they take place during the normal working day here.

Two, we’re encouraging flexible working wherever possible. This means that we are happy for staff to shift their core hours to fit the day’s demands, such as coming in a bit later in the morning so that they can be around for a meeting that might be at, say, 6pm.

We are mindful that people have commutes, families, and so on, so we are ensuring that our processes are built around the fact that after 5pm Dublin time, things have to be scheduled in a way that respects that. This doesn’t mean we won’t have processes – such as site releases – that occasionally slip past 5pm Dublin time, but if they do we will have them move back in hour increments so that people can get out of the office and have time to get back online.

From an IT point of view, why did you choose Dublin as your engineering hub when skilled technology graduates might be in more plentiful supply elsewhere?

Any business will tell you that it is a challenge to find and hire highly skilled people, which is who we are looking for here. But I do think Dublin is very competitive when it comes to the pool of technically adept candidates available.

When we first explored setting up a European base, the fact that Dublin had one of the highest concentration of IT graduates of any of the locations we considered was one of the key factors in bringing us here.

Of course, we are not the only company in the market for great talent, so naturally where there is a strong pool of talent there is also strong competition among companies looking to tap into that. The advantage we have is that we’re offering really challenging engineering roles, working on projects that are key to the business globally, the results of which will be seen and used by millions of people every month – not many companies can promise that.

What do you look for in a good technology person?

For all of our engineering roles, we are looking for four key attributes that any candidate must possess. One, a strong computer science background. We need an engineer who doesn’t just build things but that understands what they are building and using the sorts of knowledge and tools that are gained only through the study of data structures and algorithms. Most people in this category would have at least a BSc or MSc in computer science, though we do meet people who get this understanding through practical experience and self-learning.

Second is the ability to problem-solve. We solve problems at huge scale in front of a worldwide audience, with daily website releases and essentially zero downtime. We need people who don’t just come up with a single solution for a problem but can come up with a few and determine which is right for the sort of business problem they are solving.

Third: good coding skills. We aren’t a research company, we build things. We need individuals who can turn their great ideas into functioning, high-quality software.

Finally – and this might be the most important – we need people who take pride and ownership in what they do. At TripAdvisor, we run small teams focused on specific business objectives. These teams are expected to pitch in the skills they have, or learn new skills, to not only design and build the software but also ensure that it is high quality and continues to run correctly.

For better or worse, some people see themselves as exclusively ‘technology’ guys rather than someone with a rounded career path that includes commercial experience. Do you think it’s possible to combine a tech and a business mindset in one person?

I don’t only think it is possible, I think it is crucial.

There are always multiple ways to build any given feature and the only way to move quickly, and achieve the overall business goals, is to have engineers who are interested and capable of helping the business progress towards those goals.

We are a company that runs light on formal specs but heavy on communication between product management and engineering. It is completely true that some technologists have no interest in taking on people management responsibilities and would prefer to build things for most or all of their career – and we offer great career opportunities for these sorts of individuals within TripAdvisor. That doesn’t mean they can’t or shouldn’t be deeply involved in understanding the actual problem they are trying to solve and that requires at least some basic knowledge of the business are they are working in.

One way we commonly see that in TripAdvisor engineers is that our engineers are often quite interested in the metrics we use to assess our site’s success daily: be they site performance, A/B test results, or otherwise.

In your own career, have you deliberately tried to combine the commercial and technical sides to the role – and given your computer science background, did you have to reskill yourself in order to get this balance right?

I am very lucky to be surrounded at TripAdvisor by a good number of technologists who have taken on business responsibilities. This runs all the way down from our CEO, Steve Kaufer, who also has a computer science background, and certainly is true in the engineering organisation where almost all of the management team have developed into leadership roles from technical roles.

In some sense, I see the commercial and technical challenges as two sides of the same coin. To be a good engineer, you need to understand the business problem you are trying to solve. To have a successful business in IT, you need to understand the system you are building to ensure you are finding the shortest path to deliver the results without introducing poor quality or excessive technical debt into the system.

At TripAdvisor, we start everyone in the engineering team with a small project and let them work their way up to bigger projects as they progress. This is how we see even junior engineers tackling huge problems within our organisation. As the projects get bigger, more understanding is needed of the business challenges involved, but this is where the great leadership we have in place is crucial in helping people such as myself gain that balance.

In your view, should an IT department exist to fulfil what the business wants, or should it take a more proactive role in showing the organisation what’s possible given the capabilities of the technology?

Certainly, looking at it from an engineering perspective, your IT function needs to be tightly partnered with the rest of the business to the point where it is fully involved in the discussions every business has to have about its future roadmap and vision for growth. That needs to happen from day one, in my opinion.

When this works well, the engineering group is then able to effectively inspire the roadmap with cool things that are possible, but also it will be in a better position to respond to business needs that are brand new and not represented in current technology.

Our goal at TripAdvisor is always to deliver new products quickly, iteratively, and in a way that makes a difference to the millions of people who use our site. Success is measured by key site metrics and not in terms of number of projects delivered, etc. In a sense, we are lucky to have the phenomenal traffic we do because that helps us more effectively measure progress very quickly against real business metrics from both the business and IT perspective.

You’ve led some major projects in your career: in your experience, what’s the secret to a successful outcome?

There are some things we do very well at TripAdvisor to ensure we get our big projects done right:

One, close partnership with the business – down to the individual engineers. It is important to make sure that you aren’t overly focused on the specifics of the project as originally conceived but instead focused on the business objectives of a given project.

Two, a large reliance on data and measurement. For any big project, it is important to find ways to ensure that what you are delivering meets the business need. It is also important to ensure that the goals/measurement of those goals is focused on the actual business need and not artefacts of that need, which is to say particular bits or pieces of technology that may not end up getting the desired results.

Three, people. There is no substitute for great people working on any given project. This isn’t an area that you can rush or cut corners on. If you want success, hire people who have a history of success and a deep understanding and passion for what they are doing.

Gordon Smith was a contributor to Silicon Republic