In the days before big data, people used to run their businesses based on ‘gut feel’. Many still do, writes Minnaar Fourie, commercial director at King Price Insurance.
Then along came a tsunami of data, and suddenly everyone thought all their decisions were taken care of. But here’s the problem: Right now, we’ve got tons of data, but we’re still not making smarter business decisions.
Executives have smart dashboards, and more reports than they can use. And yet, they still feel left in the lurch with all this data. Is it trustworthy? How do they prioritise it? How do they relate disparate data sets before they can deduce some intelligence to improve their decision-making? The nett result is that they still need help making decisions.
The first step to solving this problem? Let’s stop talking about data as the ‘next big thing’ and instead focus on decision modelling. In other words, let’s focus on the process we use to make our decisions, rather than relying solely on data. Our businesses are drowning in data, but many have no way of doing anything with it, because very few mine intelligence from it. We must help stakeholders to make specific decisions by presenting useful data insights to key questions.
At King Price, we use something we call ‘decision science’ to turn the analysis process on its head. It doesn’t look at the data first: Rather, it takes a 360-degree look at the business problem and then looks to the data for answers.
This is key. To execute good decisions, we must start by looking at our company strategy and business priorities. If data isn’t contextualised for our specific needs, we still won’t be able to make good, fast decisions. We must sense-check whether the questions we’re asking of the data are the right ones to start with.
We also still need to allow for data exploration. There will always be some surprises. We need to look for and spot trends, and see what machine learning can unearth in terms of data that we had never anticipated.
It’s also critical to review our decisions after the fact to see where we could have done better. We hardly ever critique our decision-making. We don’t stop and question how we make decisions. We don’t interrogate the impact of the process on the final decision. We use data based on the intelligence we have at a specific time, but are we missing something? Are we overvaluing a certain category? What can we do next time in the decision-making process to get a better result?
This way, our learning loop becomes faster and richer, and we develop a mindset that will create a culture of learning and result in better decisions.
One example of how we at King Price focus on the decision-making process is when we’re deciding on ex-gratia payments during claims. Here, we distil the decision process into five questions.
The first question needs a binary answer before the decision-maker can progress to the next question. If all five questions pass the decision ‘gates’ then the payment can be made.
Another example is portfolio management, where it’s relatively easy to look back six to 12 months after a project has been completed to see whether the original anticipated benefits have been realised.
Decision-makers can then build the learnings from the benefits management process into project selection earlier in the project lifecycle, thus having better foresight to decide which projects to terminate and which projects should receive additional resources.
The bottom line? To make the most of our data, we need to start focusing on the decision-making process. Decision science is a real thing. And it’s through this multi-disciplinary approach to applying quantitative methods that we can make better, more informed decisions.