Small Data is where it's at - you heard it here first
So what are we to make of ‘Big Data’? I ask because it seems marketers cannot make it to the end of a typical day without encountering the phrase about 40 times in and around the workplace at the moment.
According to the New York Times we are living in the age of Big Data and it’s hard to disagree. It’s associated with retail analysis, market research, social media – pretty much every major sector of the marketing industry appears intertwined and intersected at every point by the concept.
Big brands such as IBM and Google are increasingly praised for their ability to “handle” Big Data and derive competitive advantages as a result. Consulting firms cite the exponential rise in data storage as key evidence that not only are we entering the age of Big Data but that mastering it will take us to the “next frontier for innovation, competition and productivity”.
A recent survey of more than 200 senior American marketers by Forbes Insights found that the more companies used Big Data to drive marketing, the more successful their campaigns were and the more sales they generated. If only the research had also asked these executives to actually explain what they meant by Big Data first – we might have been getting somewhere.
Because despite all the big headlines and even bigger predictions – the idea that Big Data represents something fundamentally different from what we have experienced before is clearly a big old load of bollocks. Yes, with the surfeit of new data methods, touchpoints and data collection tools, we are clearly in a position to experience more data than ever before.
But most marketers need more data like they need a third nipple. Most struggle with cross-tabs and making sense of a 40-minute focus group. I have actually seen a marketing manager watch eight consumers tell him over and over again that his product concept was “shit”, “ridiculous” and, my personal favourite, “clearly not serious” – only for the manager in question to conclude that the results were not conclusive and order more groups.
Ever since Lord Leverhulme admitted that half of his ad budget was wasted, he just didn’t know which half, it’s been clear that a great proportion of marketers aren’t actually able to work with even the tiniest sliver of empirical data to direct their marketing strategies.
So the idea that a sudden injection of infinitesimal amounts of new data is somehow going to improve the situation is like arguing that if you make your gerbil listen to Radio 4 he will stop eating his own shit for breakfast.
You can make the amount of data available to a marketer bigger and bigger but without the adequate approach to thinking about research and insights it will have no impact whatsoever. Yes, a few savants like Amazon and Tesco might be able to operate at a level where Big Data can and is useful. However, for the average marketer in the average business, the real challenge isn’t making their data any bigger but rather learning to collect, analyse and use the relatively tiny amount of insights they already have access to and have so far ignored.
So I am starting a new movement centred around Small Data based on three operating principles.
First, if anyone in your office uses the phrase ‘Big Data’ throw something at them and then say: “It’s about Small Data these days, haven’t you read the latest stuff in Marketing Week? You disgust me.”
Second, remember that there is no point having any data if you are not market-oriented in the first place. Unless you realise that everything you think about your product is potentially nonsense and that the consumer’s perspective is the only one that matters, the biggest data of all time won’t make any impact at all.
Third, forget about terabytes of data and go with the smallest data of all. Get out of the office and spend a few hours with a loyal customer who loves your brand and try something beyond most marketers – shutting up and listening to the people who pay for everything. A sample size of one always beats Big Data when you know what you are doing.