SPA's Broad Industrial Groups (BIGs)

Whilst indicating the magnitude of the workforce, knowing just the total workforce will not help greatly in determining what the mix is like, for retailers and leisure groups, eg: white collar office or blue collar manufacturing (important dimensions in estimating lunchtime trade potential). SPA's Broad Industrial Groups or BIGs summary of industry mix is designed to help here.

The Standard Industrial Classification (SIC) breaks down the workforce data into very specific categories: too detailed for real-world applications. Collapsing the 4 and 5 digit SIC codes into broader 2 digit SICs helps only a little

We really wanted to get down to 6 or 8 coherent industrial categories formed from these detailed SIC codes. However, we also wanted to group these together not only on industry type but including some dimension of the social class of the workforce.

Two different approaches were tried;
· Use of Government statistics on the Social Class of employees classified by SIC
· Grouping together industries still further (based on their similarity)

The official statistics seemed highly relevant to Workplace Segmentation, giving Social Class breakdowns for employees in each 2-digit SIC code. Figures referred to SIC 1980 (not SIC 1992 as used to analyse Market Location's PrimeFile Plus) so a quick translation was built. The result was a crude capability to take any workforce profile (numbers of employees by SIC) and to estimate its social class mix. If it worked, this would be invaluable in indicating the kind of potential customers available at lunchtime.

Unfortunately, the Standard Industrial Classification classifies administrative Head Offices with the industry concerned (the HQ of a mining company with Mining, etc). So, although the London HQ of a big chemicals company probably has far more white-collar staff than the company's smelliest chemicals plant, the statistics combine the two. Any estimate of the social class mix of a local workforce might therefore be subject to significant error. Market Location's PrimeFile Plus includes Head Office flags, but without separate class data for Head Offices and Branches this doesn't help. An impasse had been reached.

Grouping SIC codes together offered a different approach. If we could get down to 6 or 8 coherent industrial categories, then differences in workforce mix should also be clear-cut. In fact, grouping together many SIC codes proved easy and uncontentious. Getting down as low as just 6 or 8 industrial categories, however, was another matter. The problem was how to decide what should be grouped together.

At this point, the Government's Social Class statistics were revisited as a source of guidance for the grouping process. Ultimately, simplified Broad Industrial Groups (BIGs) were defined based on similarity of activity and social class mix. These were as follows:

BIG: Agriculture - Agriculture, Forestry, Fishing etc.
BIG: Minerals - Mining, Quarrying, Other Mineral Extraction, Metals & Minerals Manuf
BIG: Industry - Manufacturing, Engineering, Construction, Energy, Water & Recycling.
BIG: Distribution - Retail & Wholesale Trade, Catering, Transport & Communications.
BIG: Business & Finance - Finance, Real Estate, Computers and Business Services.
BIG: Public Service - Public Administration, Defence, Public Services etc.
These broad groupings have proved very useful as a component in SPA's Workplace Segmentz, which focus on the industry and broad status mix of employment in different areas.

It is not claimed that this is vastly better than estimating social class from the SICs and using the estimated class profile in segmentation. Either way, the administrative HQ of a chemicals giant is still treated as Chemicals. Any estimate of the social class mix of a very local workforce might therefore be subject to some error at individual company level. However, HQs of these sorts of organisations are usually in city centres where white-collar employees dominate the areas around these "anomalies". The mix, therefore, is still predominantly white-collar.

Certainly in segmentation terms, our work suggests that, when combined with other data and used in Cluster Analysis, the Broad Industrial Groupings work better. Using BIGs, retains the fact that there was some employment classed as Chemicals etc. in the service-oriented environment of some big city centres, and these are classified accordingly. By contrast, estimates of the Social Class mix lose the Chemicals angle, and the areas affected are lumped in with other city centres with a similar Class profile. The advantage may be marginal, but it seems to be real.

To illustrate how BIGs can "pull apart" different workplace environments, consider SPA's 9 Workplace Segmentz Groups:

Code Group Name
W1 Major Magnets
W2 Large Centres
W3 Industrial Zones
W4 Distribution Zones
W5 Service Zones
W6 Housing & Industry
W7 Homes & Distribution
W8 Dormitory & Service
W9 Agricultural & Mixed

If we now examine the mix of BIGs across each of these Workplace Groups, clear differences emerge:

Profiles of the Workplace Groups - Postcode Type & Broad Industry

Notes:
All figures are indexed. 100 = the GB mean of the "retail-focused" sample Postcodes.

back to top>

SPA Marketing Systems Ltd
Leamington Spa, Warks, CV32 6PT
Tel 01926 334978
Email:info@spamarketing.co.uk or Contact us

Please Note: If you wish to PRINT this page out. Click on the page first then select print