It’s not uncommon to talk to potential clients who consider themselves to already be very much data-driven in the way that they operate. However it’s very rare to find a potential client that truly is exploiting the full potential of the data that they hold. That’s because companies often confuse business intelligence with predictive analytics, or think that once they’re using their data for business intelligence that they’re doing all they can to get value from it. Neither of these things is true. Predictive analytics is not the same thing as business intelligence, and if you’re just using your data for business intelligence applications then you’re almost certainly not getting as much value from it as you could be. But how exactly does predictive analysis differ from business intelligence? What should companies with established BI practices be doing next? In this blog post I’ll explore these questions and make some practical suggestions regarding next steps for those who want to move beyond simple BI applications for their data.
What is the difference between business intelligence and predictive analytics?
Business intelligence is about using the data you hold within your company to report on historical trends and current business performance. The best BI applications enable business users to get easy access to their data in order to quickly gain insights about the current performance of the business or to identify trends and patterns in past performance. Business intelligence is a backward glance over the shoulder to see what happened yesterday, last week, last month, last year. Good business intelligence applications give you this kind of information at your fingertips, making them very useful and popular with business leaders. Business intelligence is vital and good use of the insights gained from BI is a great starting point if you want to be data driven. However the one key limitation of BI is that it’s backwards looking. It can give you a huge amount of information about what’s already happened, but what it can’t do is tell you anything about what’s going to happen next. This is where predictive analytics comes in.
Why is now the time for predictive analytics?
Predictive analytics goes beyond these backward-facing views and uses the data you already hold in your business to look forwards and tell you what’s going to happen in the future. Not only that. Predictive models and algorithms allow you to not only predict the next most likely outcome but can also tell you what’s the next best thing that could happen. And good predictive analytics tools will automate this process for you, so that your business decision making becomes fact-based and truly data-driven rather than based on subjective judgements and hunches.
Your business intelligence tool can tell you which of your products is currently selling best, and show you trends in your product sales over time up to this point. But what if you want to know how well a particular product is going to sell in the future? Perhaps you’re planning an advertising campaign. What effect will this campaign have on future product sales? Which of your customers are most likely to respond to the campaign? This is what predictive analytics can tell you. Perhaps, using your BI tool, you have identified that your customer churn rates having been rising. What should you do about this? Without the additional insight of predictive analytics it’s hard to be sure. Using the same data that you already have to build a predictive model you can find out which of your current customers are most likely to be thinking of leaving you in the next year. Not only that – a good predictive tool can tell you which of the various actions you could take to keep them is likely to work the best for each customer.
With the vast variety and volume of data now available this move from information availability into insight and insight with impact is more important than ever before. The most effective organisations today have honed their ability to be data-driven: they can quickly mine and model all of this data to find the most meaningful patterns or combinations of data to predict the next best actions or outcomes. And use this insight to make a difference in their decision making.
What are the first steps you should take if you’re serious about moving into predictive analytics?
I’ve helped many, many companies to make their first moves into predictive analytics and the advice I always give is to start small and roll it out slowly. Don’t try and build an entire analytics or data science practice in one go. It’s much better to work incrementally. Start by engaging your business users to find a business problem where you believe you can have a measurable and meaningful impact. Start with one that’s easy, and where small percentage gains can make a significant difference to the bottom line. Maybe a 1% increase in cross selling would result in a noticeable revenue impact. How could that 1% best be achieved? Perhaps a 2% increase in repeat purchasing would mean a significant bump in sales figures. Which customers should be targeted with this offer?
Define the key metrics that you’re interested in, identify the data that you have to work with and then get into the analytics quickly to find new patterns. Work with your business stakeholders and keep them involved throughout the process, so the analytics aren’t a mystery to them. Really focus on building a simple, useful model that can be deployed quickly. Throughout this first project keep in mind that there is no such thing as a perfect model so don’t bother trying to build one. Instead concentrate on finding a model that’s good enough to make a difference. And measure that difference. Once you have the evidence that predictive analytics works on a small scale it will be much easier to roll it out more widely into other areas of your organisation. This considered, start small approach will increase the odds of success and adoption and lays the ground well to build a culture of data analytics driven decision-making.