Contextual analytics: the map to marketing success
If you visit San Francisco and you want to drive from the Zoo to Chinatown, chances are you’ll use online maps. You’ll see a few possible routes and for each one the distance and time to destination will be calculated and displayed. That’s all you really need to make an informed decision; either you care about how long it takes or how far you travel, or both. Distance is based on GPS location and drive time is based on a whole lot of data points like time of day, traffic, roadworks, car accidents and major events. Maps track your progress in real time as you drive towards your goal and alert you in real-time if you deviate or need to suddenly change direction. Genius! If only there was a similar tool for helping businesses reach their destination of choice…
Becoming the mapping tool for marketing optimization
When we use all the data available to us from a multitude of connected devices we know what content is likely to work in what situations — and precisely how well it will work. If a business wants to shift 600,000 winter jackets through stores across the US before the end of winter then it should be possible to make recommendations about the kind of content that needs to be created and how it needs to be targeted. Each different scenario suggested should contain the measurements that marketers care about: total units, average margin and projected profit. Decision-making will be simplified and outcomes will be improved through a process of automation. Select the most appropriate destination and let relevant content dictate the best route to the outcome you are aiming for.
Beyond describing behavior and predicting results
When it comes to analytics we can stick with describing what we see in the data, or we can apply more complicated analysis and algorithms and get a fairly concrete idea of what we should be doing next.
Descriptive analytics is what it sounds like; showing us the data we have so we know what’s happened already. It’s obviously useful — if you have a store you need to know how many visitors you’re receiving, how much they spend, how long they stay — and whether that’s trending up or down. But we can do more. Predictive analytics goes a step further, using more advanced algorithms and modelling to predict outcomes. If we raise prices, how much more profit will we make, and how will sales be effected?
Prescriptive analytics is less about ‘what will happen if we do this’ and more about ‘what can we do to get the result we’re looking for’. So if we’re looking to move winter coats, is lowering prices our best bet? Or would we be better off promoting coats to people who we know are interested; because they last bought a coat 3 years ago, because they’ve added coats to their online store wishlist, or because they’re spending an awfully large amount of time in the coat aisle?
Like it says on the label, prescriptive analytics prescribes the best way to move your product — or the best route to Chinatown, given your constraints — including time of day, fuel costs, traffic conditions and weather.
The case for contextual analytics
It’s this context that makes all the difference. Without knowing the contextual variables it’d be difficult for your map or GPS service of choice to deliver the best result. Obviously if you’re sent through an area with heavy congestion or surface flooding, your travel experience is going to be sub-optimal and your goal (arrive in Chinatown as quickly as possible) isn’t going to be met. Even though the platform has the data it needs to plan a route — it knows which roads run from A to B — without the context it can’t choose the best route, and it certainly can’t update or revise that course on the fly. Or the drive.
Contextual analytics takes into account not only the data that directly relates to the activity you’re interested in — driving or shopping — but it considers environmental and situational factors when prescribing the best course of action. A sudden unexpected snowstorm in New York will have a material impact on your ability to move coats in your New York locations, and the unexpected nature of the situation means you have to be able to act quickly to take advantage of it. A prescriptive, contextual analytics platform that can pull in that data and show you — in real time — how best to leverage it (send this promotion to these customers at this time) is going to be far more useful than a tool that simply tells you how many coats you sold after the fact.