What did you want to be when you grew up?
I never had any particular aspirations at high school, and I was (still am) average at mathematics to some people’s surprise. It wasn’t until my undergrad that I liked finding things out and statistics seemed the way to do that. Even when I decided to pursue it academically, I didn’t have a job in mind.
What did you study at university, and did you find that it prepared you for your career?
I studied Liberal Arts and Sciences for my undergrad in The Netherlands, went to St Andrews to study my Master’s in Statistics and got my PhD in Actuarial Mathematics from Heriot Watt. My Masters degree was a decade ago when data science/data engineering really didn’t exist as they do today. My university time prepared me as well as it could. University is all about core competencies and learning critical thinking. I’ve been fortunate enough to be taught by absolute subject matter experts, passionate about their field of study; it’s infectious and I admire their patience.
Can you describe your career path so far?
It was only when I was about to graduate that it occurred to me to find a job. I’d never done internships or had any relevant work experience and realised how competitive the market for graduates can be. I decided to do econometrics at BrandScience (now owned by Annalect), where I had a fantastic time. It was a relatively small, specialised operation where juniors were given responsibility for modelling. I worked on pricing and marketing effectiveness for large accounts. It’s important to mention that my first months I spent formatting data in Excel, and nothing else.
After some time I felt I wasn’t done learning yet and decided to get my PhD. I made the most of my PhD; I did industry internships, freelance work, started a business, presented my academic work across the world and worked with experts in the field. The small business I started got picked up by Colliers and this is how I rolled into my role advancing its internal analytics offering, centred around the automated valuation engine I brought with me.
Can you tell us about your role at Forecast, and your responsibilities?
At Forecast we help companies with financial modelling and analytics. I joined Forecast 18 months ago before an Advanced Analytics group existed. In that short space of time we’ve grown from just me to a team of 12 consultants specialising in business intelligence, data science and engineering, 10 of which are based in Edinburgh. My responsibilities are split between talking to clients, both existing and new, checking in with the team to talk about their projects, and leading some projects. I’m also part of the management team.
Consultants at Forecast typically work in teams of up to three for several weeks or months. Every proposal for work or scoping involves me at some point, but also largely draws on the input of relevant team members. Beyond scoping, I’m involved loosely in all projects as I need to make sure everyone is getting on well and would typically only be asked to weigh in from experience.
What advice would you give someone who was thinking of studying data?
Data is either a tool to support a core business or the data is the core business itself. I like statistics because I could find things out and for me that mindset of curiosity has helped me a lot. The core skills required as a data analyst are relational databases where the data sits, SQL to manipulate and extract, then to be analysed using Python/R or specialised dashboarding software.
I would encourage students to consider these (generic) technical skills threshold skills; they’re unlikely for you to get ahead of the pack, but the other things you have to offer will.
What are the challenges of working with data?
Technical challenges are exciting, but I’d like to mention at least two factors that can really hurt the effectiveness of a data analytics team or individual. There are plenty of barriers to doing analytics outside your control, especially in larger institutions, where legacy infrastructure can dictate your daily life when it comes to, for instance, deployment of a solution and its adoption. Another challenge can be an organisation’s or managerial analysis paralysis, which can manifest itself in two ways. Some managers might be overwhelmed with the possibilities; they have access to data or can get it easily, but are uncertain what questions to ask. Others might know that data is collected within their organisation data, stored and even used to drive insight, but this manager feels disconnected. It may be a marketing team who is looking for access and capabilities but doesn’t have it; it feels too far away and paralysis kicks in.
What are the biggest trends in data, and what do you think the next big thing will be?
I like to think of the typical, infamous hype-cycle curves and figure out where we stand at the moment. There’s an important distinction to make between analytics, Machine Learning and Artificial Intelligence at the cutting-edge and in the real world. Advancing the cutting edge stuff is critical and people discuss another AI winter coming, attributed to (perceived) limitations of deep learning, but at the same time we see incredible applications of Reinforcement Learning, not even mentioning the impact of Quantum Computing on the field of analytics further down the timeline. I would separate that from the analytics that most professionals get to do and here I see extremely positive signs. As the hype goes, we see more and more pragmatic commercial applications (often cloud-powered) rather than analytics for the sake of having a data science team.