The start of a data trail

I'm often asked how I got into 'data' given my background was in advertising and media.

I joined advertising at the tail end of the heady days of the 80's. At the time advertising was hailed as the new 'profession' alongside Doctors, Lawyers, Accountants...

Having done econometrics, I was particularly interested in the economics of advertising. How could we understand as much as possible about people we wanted to target, so we could identify the right media channels to find them? And then: How could we prove that our campaigns were having an effect on sales?

Often the communication choices we made were based on research with small sample sizes done months prior to campaign implementation. Campaign results were too difficult to extract from other factors... Incredibly frustrating. Clients were way ahead of us in their knowledge of what was going to work, so I spent a lot of time with them trying to acquire that learning.

It became abundantly clear after unsuccessfully trying to integrate a multivariate econometric approach into an agency, that most of the agencies weren't interested in knowing whether campaigns worked, they were interested in producing great creative. For me these aren't mutually exclusive. The best creative should engage the customer in a way which makes them want to desperately have the product.

I left Adland to start a business working with Venture Capital funds. We were valuing the future potential of a customer database or target audience in businesses they were seeking to acquire or invest in. We were also responsible for evaluating the performance of the Marketing Department (and suppliers) in those companies.

It struck me - still does - that many organisations had not linked Marketing to the customer database. Product, sales and marketing initiatives were not linked to the knowledge the company already had on their customers. As a result opportunities were being missed and marketing was anything less than optimal.

A friend introduced me to a business with lots of data. Data at the sharp-end of grocery retail loyalty. What people were buying linked to who they are and where they live.

I was asked to develop their media proposition. They had the seed of an idea that the data they had could somehow help advertisers and agencies with the decision on where to invest their cash. An amazing idea. I could understand customer actual (real) shopping behaviour (not self-reported) to do better targeting, but better still, if I put an ad on TV I could see who was buying it within a matter of days / hours.

Holy Grail stuff. It's now a multi-million pound business.