The Harvard Business Review just published an article entitled Why You’re Not Getting Value from Your Data Science.
The article describes how data scientists are overwhelmed by the complexity and quantity of data, and how business experts, for their part, are underwhelmed by the tangible output of those data scientists.
Why this failure to get business value?
The data scientists were obsessed with fine tuning machine learning models rather than answering fresh questions, losing sight of the main purpose of their work: generating business value. In fact, when the author asked a room filled with 150 data scientists which of them had ever generated proven business value, no one raised their hand.
Me too. I would have expected about five hands to go up.
Although today’s technology has given data scientists great tools, along with an abundance of low-hanging fruit, the problem remains that a recent surge in unfocused data science programs, staffed with newly minted analysts, has created a high potential for program failure.
From my experience leading and evaluating data science programs over the years, I would call out three key problems commonly seen within data science and big data analytics programs.
- A failure to bridge the gap between business goals and analytic efforts
Math-y people focus on math-y work, typically producing a succession of faster, more accurate math-y methods. The ‘good but irrelevant’ data science tools are carefully crafted into ‘even better but still irrelevant’ tools. By human nature, we focus on the things that we are good at, but effective data scientists need to cross-focus on business goals. Agile principles such as short feedback loops with the business stakeholders are also difficult with projects that typically involve long periods of isolated R&D. They are, however, critical to long term business success of analytic initiatives.
- A failure to work in a lean manner
Data science projects typically are highly exploratory, with uncertain outcomes. Focusing on a few high-risk efforts means a high likelihood of failure. That’s why the principle of Minimum Viable Product (MVP) is so crucial for data science programs. Unfortunately, the MVP concept gets you failed out of most university analytics programs, so we’re looking at quite a paradigm shift for data scientists once they hit the working world.
- Lack of discipline in measuring results
I’d bet that most of the 150 data scientists surveyed in this study did produce measurable business value, despite not knowing it. I would bet that no one had the discipline to measure the results. The process of evaluating the impact of new initiatives is a part of being a data-driven company. You should really already have a measurement success metric in place before beginning the analytic initiative.
Four tips to start getting business value from your data science efforts:
Here is my advice for businesses who are uncertain about the value they are getting from their data scientists
- Live and breathe business intuition. There are certainly people in the company who have extensive experience with the customer, the product, the market. Start by talking with them. Harvest their intuition. Go back to them every few days to show your data and initial results. They will tell you if you’re doing something blatantly wrong. Oh, and make a habit of using your company’s website / product / services yourself (this is known as ‘eating your own dog food’).
- Measure Results. Don’t start a data science project unless you know why you’re doing it and what it looks like when it succeeds. Are you looking to increase conversion rates? Marketing ROI? Market Share? Customer Lifetime Value? Measure where you are now, where you think you could be, and how much revenue that translates into. As they say, ‘if you’re not keeping score, you’re just practicing’
- Work in an agile manner. Agile project management is key to making an analytics program effective, and it ties in closely with the first point above. Business intuition will fuel the feedback loops and provide input for specifying many of the project deliverables. Work in terms of minimum viable products and short delivery cycles.
- Appoint a leader with understanding of both business goals and analytic possibilities. Ideally, one person can match business value with analytic potential and lead a team of data scientists in attaining high value results. Finding such a person can be difficult (see my article Recruiting a Chief Data Scientist). In the absence of having such a person, set up a structured framework of continuous interaction between business and analytic experts.
As with most things that are worth doing, making a data science program effective can take substantial effort and will require several iterations before the program is structured in an effective way. Don’t give up if the initial program seems to be failing. As Mckinsey said in their recent report, The age of analytics: Competing in a data-driven world, leading firms which have already developed strong analytics programs are not only winning in their own fields, but are actively looking for ways to disrupt adjacent industries. Don’t get disrupted by them.
I’d love to hear peoples’ thoughts and experiences in the comments below.