Predictive Analytics is STILL just Science Fiction
Danny Brown writes about predictive analytics and – as a marketer, reading content for marketers – I can just imagine the widening eyes as my fellow marketeers scroll through each line. Brown is serving up the stuff of marketing legend.
Instant connection. Instant relevance. High on context and [personal] memories and direct to [the consumer].
Brown claims the Excalibur of marketing can be fashioned from big data. In the not so distant future, Brown suggests, marketers would essentially become data scientists, studying endless reams of likes and links and shares, searching for connections. Upon finding a pattern, heat-seeking sales robots will be deployed, and sales will be made with staggering accuracy. It all sounds quite nice. Nevertheless, this future rests on the awesome power of predictive analytics – and today’s solutions just aren’t there yet.
I see the lure, my curiosity is piqued by the possibility, but there is a chilling doubt that a perfect marketing future will ever exist. The reason is that prediction – at the level Brown is talking about – just isn’t feasible – something he acknowledges in the same post. Brown quotes a data analyst from a conference he recently attended.
..it would take 1,000 data analysts working 24 hours a day, 7 days a week, more than 300 years to sift through everything currently available to us.
So then, Marketopia is within reach – if only we had faster computers, smarter analysts, more potent coffee, etc. Unfortunately, no. While in theory all of this works – it doesn’t work now and there is no sensible reason to invest in developing the technology to do it. In fact, by the time that such an investment does make sense, there will already be a cheaper more reliable alternative: Mblast calls it Contextual Intent.
We’re not about predicting who will buy what, at which times. We’re about being proactive in the search for new customers – this is the core of marketing from Ogilvy to Oblivion. Predictive analytics assumes that with enough data, people are powerless to resist an offer. Instructive Insights assumes that people are far too complex to pin down like that – its better to let the customer tell you what they need. We don’t need alien tech or warp speed CPUs to do it. We just need an intelligent search engine for social data. The difference between predictive analytics and Instructive Insights is that some percentage of the time predictive analytics will be wrong. Instructive Insights is never wrong because there is nothing to be wrong about. You tweet that your car broke down, @AAA tweets back. You post your wedding date, ads featuring all the wedding necessities feature in your news stream. Mblast connects customers with a need to brands with a product – its really that simple.
Interested? Drop us a line.