Geodemographics: the birth of big data

x Sponsored content on Research Live and in Impact magazine is editorially independent.
Find out more about advertising and sponsorship.

I have attended most MRS conferences since the late 1970s, but a presentation that still sticks in my mind is the one given by Ken Baker, John Bermingham and Colin McDonald in 1979, introducing the first commercial application of geodemographics*. 

The British Market Research Bureau (BMRB) team had been inspired by a lecture given by Richard Webber at the Centre for Environmental Studies in 1977, describing his pioneering development of a classification of residential neighbourhoods (CRN). This was shortly after BMRB had completely redesigned its main sampling frame using data from the 1971 census – what it believed to be the first nationwide computer-automated sampling frame in the UK market research sector. 

This was important because it made it relatively straightforward to add the CRN 36 cluster solution to the sampling frame, and the 1978 TGI dataset was also back-coded with the clusters. It had cost BMRB the princely sum of £160 to obtain a classification! 

Webber moved to CACI, and Acorn – the first full, commercial geodemographics process – was launched by the end of 1979, adding the codes to the full list of census variables. 

Another important development at this time was the Royal Mail’s introduction of a national postcode system for UK addresses, and its financial inducements to encourage database owners to add postcodes to their address records. 

I believe this also led to the birth of ‘big data’, because – by using the postcode – geodemographic codes could also be added to each record. This eventually provided links to other data, especially from surveys such as NOP’s Financial Research Survey. Such systems would become standard in market research in the years ahead and a whole new sector of marketing analytics was born.

Returning to 1979, the findings presented in BMRB’s paper opened our eyes to a new world for marketers. As Baker described in his introduction to a reprint of the paper in two special issues of JMRS that celebrated the Market Research Society’s 50th birthday, they enabled the ‘where should I?’ question to be answered. 

The indices presented by the authors provided a new view of consumer consumption patterns by residential area – whether it was for wine purchasing, credit card ownership, book buying or exposure to media. They also demonstrated its value to social research.

Since then, geodemographic coding – and the different systems – have become ubiquitous within market research and marketing. CACI and Experian (Webber developed its Mosaic system) remain two major players in this field, but there are several other companies providing such products – updated after the 2011 census – plus an open data ONS/UCL system. The concept has become international – I spoke at the CCN (now Experian) launch conference of Australia Mosaic, in Sydney, in 1992.

There is also the very influential MRS-affiliated Census and Geodemographics Group, founded by Barry Leventhal – who devised the Pinpoint system and FRUITS classification – which celebrated its 25th anniversary in 2014.

Some years ago, Webber was awarded an Honorary Fellowship of the MRS for his work in this field.

Peter Mouncey is a Fellow of MRS and editor-in-chief of  the International Journal of Market Research 

Reference:

*The utility to market research of the classification of residential neighbourhoods, by Ken Baker, John Bermingham and Colin McDonald (BMRB), JMRS Vol 39 No 1, January 1997 – Proceedings of the MRS Conference, March 1979, Brighton.