I trained as a statistician in the late 90s but found that analytics in industry were disconnected from the real challenges that business faced. A passion for programming and problem solving led me to a consulting role at Insightful, where I met my Mango co-founder, Matt Aldridge. We shared a mutual belief in the potential of advanced analytics to drive value and believed in a “data-driven” future for business. We founded Mango Solutions in 2002 to provide pragmatic advice and consultancy to help organisations deliver value from “data science”.
My early work at Mango revolved heavily around the R Language. I was the first R Consortium president, co-authored the “R in 24 Hours” book, instigated R events such as the EARL conference and taught over 200 R training courses.
Today, Mango has 75 employees and uses R, Python and AI to help leading organisations deliver value from advanced analytics. Personally, my role has developed from a data science practitioner to someone who offers strategic advice to business leaders. My data science background, together with 20 years of commercial consulting experience, allows me to quickly help customers discover opportunities, build capability and plan for the realisation of data-driven value.
My proudest achievement to date is the creation and growth of Mango Solutions and the building of incredible data science and data engineering teams. I am continually amazed by the talent that we attract – that really is the secret of our success. Beyond that, I’m lucky enough to be in a position to help a range of organisations achieve their analytic aspirations; my proudest moments are when we help our customers realise value and are able to celebrate the wins together.
As a data scientist I’m inspired by those with “practitioner” backgrounds who are positively impacting traditional businesses with data. As such, beyond the members of the Mango team, I’m inspired by people like Harry Powell (JLR), Magda Piatkowska (BBC), Tom Smith (ONS) and Ryan den Rooijen (Chalhoub Group).
Professionally, very much so. Mango’s ambitions for 2019 were built around growth and investment, so last year was been about recruitment, opening new offices, investment in sales and marketing, product development and the development of a new analytic workflow. Outside of Mango, the uncertainty caused by Brexit loomed large over 2019, impacting many of our clients in industries from retail to the public sector.
The hype around data and analytics has brought us to the peak of inflated expectations and so 2020 feels like a critical year for the industry to deliver realisable and repeatable value. This will put more pressure on data and analytic leaders to demonstrate a return on investment, eg, educating the business and moving data science from an “artisan” activity to “business as usual”. In 2020, we’re expecting to work with more clients looking to embed the right culture to enable analytics to succeed, while continuing to help others in the development of high-performing data science capabilities.
I’ve always believed in the power of data and (advanced) analytics to drive better decisions – this can positively impact not just the world of business, but our ability to help the most vulnerable in society and to combat some of the biggest threats facing our future.
However, where I see opportunities, I also see threats. As an industry, we need to ensure qualified and experienced leaders are in place to ensure strong governance and ethical constraints. Without this, I fear we could see significant issues around the incorrect use of analytics, which could undermine trust in data and analytics.
When we speak about data, we use a range of “defensive” (govern, protect, manage) and “proactive” (leverage, share, monetise) words. I think the biggest tech challenge is that most data platforms are built around the “defensive” elements, yet we measure impact against the outputs from “proactive” activities.
Our clients’ key tech challenge comes from the need to generate value from platforms designed for “defence”. We need platforms that enable us to deploy pipelines that turn data into wisdom in a repeatable manner. The technology exists, but a lack of understanding of modern technical approaches can often constrain thinking.