We use cookies. You have options. Cookies help us keep the site running smoothly and inform some of our advertising, but if you’d like to make adjustments, you can visit our Cookie Notice page for more information.
We’d like to use cookies on your device. Cookies help us keep the site running smoothly and inform some of our advertising, but how we use them is entirely up to you. Accept our recommended settings or customise them to your wishes.

Will the real data scientist please stand up?

Confessions of a Data Scientist

Data science is an emerging sector, and it’s a bit like the Wild West where rules are still being made up. Perhaps that’s why impostor syndrome – the fear of being found out to be a fraud or feeling inadequate – is so prevalent. The more experienced I become as a data scientist, the more convinced I am that we need to tame this beast once and forever.

Confessions of a data scientist

Most of us haven’t got to be working here at Merkle Aquila through taking a traditional route.

A lot of us have come from non-mathematical or non-technical education. I have a marketing degree and became an analyst eight years ago, so whatever I’ve learnt has been on the job. Some of the most talented analysts I’ve worked with came in to the profession straight from school, quite by accident, and by learning and developing their own style of analysis. Because get this, there is no prescribed career path or a professional qualification recognised across the industry.

Feeling like everyone’s more qualified?

Previously people entered analytical roles from statistical backgrounds – and they are always going to feel underqualified compared to technical people when it comes to computational challenges. And if you come from technical, you have the opposite problem; of being able to implement things well but not having the statistical reasoning to understand everything that’s happening in the background.

Made worse by comparing ourselves?

It’s likely that everyone’s coming to data science with a perceived weakness. Made worse by comparing the worst we know of ourselves with the best we see in others.

You go to trade shows, and you read articles, and everyone’s putting their best work forward. Day-to-day though, it’s rare that deep learning and AI solutions are required, and 95% of what you do in analytical roles isn’t very showy but is still very useful.

What can you do?

You can’t be great at everything. Rather than trying to be, pick a few things you are interested in and develop a niche.

Focus on strengths, not weaknesses. Identify what you are good at and develop that skill.

Coming from a marketing background, I chose to focus more on CRM. That led me to segmentation and propensity modelling, and I understand those. Just don’t ask me to develop a convolutional neural network anytime soon!

I have also been involved in recruiting and designing job specs, and we list complex requirements in the hope of someone having some of them. Realistically none of the applicants are going to have all those skills because if they did, they would want a more senior role. So, don’t hesitate to apply for certain jobs if you see a checklist of requirements and can’t tick off 100%.

Join the team