Career Switch: What a data analyst actually does

An interview with Selina Jones – Data analyst

What does your job involve?

At the moment I work for a higher-education establishment, and we have a big customer-relationship-management (CRM) system full of details about alumni and donors. Part of my job is to delve into that data, to analyse where gifts are coming from and work out who’s most likely to make gifts in the future.

I run a lot of other queries, too: I might need to find people that meet a particular criteria, for a mailing or perhaps a telethon, or I might need to create a report on something we’ve done lately. So I pull the data out and ask: what did the picture look like before? What is it now? What’s changed as a result of our actions? There’s an element of predictive modelling too: I’ll use statistical software to identify people with particular characteristics that we’re interested in, then model other people who look similar.

What technical tools do you use?

Our database has an built-in query tool, which I can use to pull the data off and put it into Excel, or for statistical modelling I’ll use IBM’s SPSS software. In previous jobs I did more SQL coding, using plugins such as Oracle and Crystal Reports to extract data – that was when I worked for a local authority, looking at health- and social-care data. I also have a background in Access – I used it to pull out data from various databases, and sometimes people would come to me and say “my database doesn’t work any more”, so I’d have to go into the Visual Basic code and work out why. I wasn’t trained in that, but I was able to pick it up, stepping through and debugging bits and working out why things weren’t working. I’m now in a big organisation with more hierarchy, so I don’t get to go into the back-end these days. There’s no more writing SQL, which can be frustrating because you’re limited to what’s in that custom CRM. But I expect that will change at some point, because what people want is becoming more complex.

Did you need a technical background to get into this line of work?

Actually, I originally studied classics at university. It was interesting, but not particularly useful. However, I did do a postgraduate certificate in social-science research methods, in which there was a statistical element. Really, though, my career started with a basic data-entry assistant job, which was my first introduction to databases. It was here that I completely fell in love with data, and all my jobs since then have revolved around data analysis – becoming more in depth as my technical knowledge has increased.

The move from a local authority into the fundraising profession might seem a big change, but I’m still using the skills I had before. People are recognising that these skills are transferable, so it’s perfectly possible to move between fields.

How might someone get started in this career?

There are plenty of opportunities out there: every organisation seems to want someone who can get to grips with data. You can actually get started with basic Excel skills. It sounds silly, but many people don’t understand things such as pivot tables and conditional formulae, so at the lower end of the salary spectrum, you’ll definitely find jobs where companies may need someone who can use Excel.

When you want to move onto statistical analysis, you may be able to get training in a package such as SPSS. Otherwise, you can download R for free, and work your way through the online documentation and support.

The thing about this career is that you have to be someone who enjoys playing with things. You have to put your own time into exploring datasets, looking for patterns and seeing what kind of things you can do. Often at work you’ll be given a specific task, and you won’t necessarily have time to do much exploration, so you need to know how to look at the big picture and find what you’re looking for.

What’s the worst part of the job?

Perhaps the most frustrating part is when you’ve spent hours and hours pulling together a report, and then somebody’s first reaction is “wouldn’t it be great if we could see such and such as well?” People don’t always recognise the time and effort that goes into this type of work, or they may not immediately grasp why you’ve focused on one thing rather than another. But as long as you’re clear about exactly what you’re giving people, they’re normally satisfied.

What’s the pay like?

When I started working with data I was on around £20,000, but data (and especially “Big Data”) is becoming enormous, and as more organisations see the value in data, salaries are only going to improve. In the job I’m in now I’m actually treated as the senior person in my team – that was a real surprise at first, because normally in other teams I’ve been quite junior. But that’s the way things are going.

Interested? Here’s some further reading…

1. Moneyball: The Art of Winning an Unfair Game by Michael Lewis (or watch the film with Brad Pitt) (www.amazon.co.uk)

2. CoolData (cooldata.wordpress.com)

3. Stats With Cats (statswithcats.wordpress.com)