Let’s talk about the time savings involved with high-performance analytics. I’ve blogged before about the time reductions we’ve seen when analyzing big data with our high-performance analytics solutions. We’re quoting time savings of up to 92 hours, but what does that mean? Or, as I like to ask, “So what?” Well, it’s not just the time savings that matter. It’s what you do with that time – and what you’re getting within that time frame that matters.
It comes down to this: do you want to be reactive or proactive in your decision making? Let’s take an example from retail. A lot of high-performance analytics products can give you a report in seconds about last week’s sales or last season’s sales for every item in every store. But that’s reactive. How can you become proactive in the same amount of time? You predict which products to stock in each store for each season, including how many of each size and each color for every store.
If you can get THAT kind of information in a matter of seconds instead of days, you really are making a difference. Before high-performance analytics, when it took days to find the answer to that question, you had one chance to build the predictive model, send the results up the chain, get a response on how to proceed and maybe apply one or two changes to the model before making your final prediction and planning your entire quarter based on the results.
Now, if you can get those results in a matter of seconds, you can improve the model multiple times before sending the results for review. You can select different variables, ask more what-if questions and even provide multiple options for decision makers to determine how they want to stock each store for the season. You even have time to go back and forth for a few days getting input from buyers and store managers to improve the model based on their feedback. Or you can make a change in a few stores and add the results you see there back into the model for next year or next quarter.
When we talk about changing your processes based on the time savings from high-performance analytics, these are the changes we’re talking about. It’s a matter of bringing more input and more ideas into the decision making process, and having the ability to ask more questions of your data in a shorter time frame – all with the end result of making better and more accurate decisions.
Now, use our retail scenario to answer that original, “so what?” question. So what if you can save 92 hours in your modeling process? What does that mean? The obvious answer is that you can build more models. But that doesn’t mean you make 100 more predictions. It means you can improve your original prediction in 100 different ways. Or you can bring 100 new ideas to bear on the original model. Or maybe you can shorten your seasonal planning timeframe by 100 days. The ultimate result, the ultimate “so what” is that you can improve revenues for every store in the retail chain. You plan better for each store and each season, and you have higher quarterly earnings and less overstock to clear out before the next season starts.
That’s a whole lot of “so what?” And it’s not all about the time savings. Time – thanks to the power of high-performance analytics – is just the factor that makes the rest of it possible.