By next spring, market researchers could at last have a new social
classification system to replace the antiquated and much criticised ABC1
social grading system.
This week, David Rose, Professor of Sociology at the University of
Essex, unveiled his long-awaited review of the system, four years after
being asked by the government to do so.
Professor Rose’s classification system is primarily for use by the
Office for National Statistics in time for the next national Census in
But the reason it is causing such intense interest is that he argues,
and the research community hopes, that it can be modified for use
outside the Census to replace ABC1 altogether.
For the next four months, the new system is being evaluated by a panel
headed by Erhard Meier, director of IPSOS-RSL, and Corrine Moy, director
of statistical services at NOP Research, who are ’road-testing’ it.
Results will be compared with the existing ABC1 system and current
government classification system to determine whether it provides better
tracking of the modern consumer and to check how well it
The results will be presented in a paper at the Market Research Society
conference in March, when the reaction of the commercial research
community to the new system can be truly judged.
The crux of the new system is that it offers a more modern, classless
view of the British consumer. Like the old system, it is
occupation-based, but it more accurately reflects people’s purchasing
power by their positions in the labour market.
For example, under the old system, policemen were regarded as skilled
manual workers in the C1 group. Now, they are seen as associate
professionals, which moves them into Rose’s second level of occupations
(see box, page 30). The new system also separates out self-employed and
small employers, better to reflect the trends of modern society.
Of his new technique, Rose says: ’It isn’t designed to try to divide
society up by consumption. Neither is it a refined segmentation by
markets. It is a similar guideline to ABC1 in the sense that it uses
Rose’s system interests marketers and market researchers because it is a
more accurate tool with which to draw distinctions between purchasing
Details like including self-employed and small employers is seen as a
’We see an opportunity there for an alternative to ABC1, but we need to
evaluate it,’ says Barry Levanthal in his role as chairman of the Census
Interest Group at the MRS. He is also vice-chairman of Berry
Better or worse?
’We need to know whether the new system will be better across the board
or worse,’ says Levanthal. ’Classifications are used for all kinds of
different purposes, from product usage to targeting, and we need to
establish the strengths and weaknesses of the new system.’
He adds: ’What I don’t know is whether the new system will help to get a
handle on the new consumer, as it is based on employment and
I’m not sure how it will work on media consumption. The old ABC1 system
is more complex in a sense as it includes an element of lifestyle, which
the new one doesn’t.’
Levanthal’s concerns will no doubt filter through at the MRS conference,
but ultimately it will fall to market researchers to advocate it, or
not, to their clients.
An MRS validation would certainly help a wholesale conversion. But such
a change to Rose’s system is already being received with caution
generally by the market research, data warehousing and direct marketing
community. More likely is some half-way house arrangement whereby the
two systems run in tandem.
This may cause chaos. Like a currency, unless everyone uses the new
system it will not work. One side issue is that many market researchers
may have to be re-trained. Another is that any wholesale replacement
will make a nonsense of trend data.
Even Professor Rose has gone on record as saying that even if the
research industry gives his new classification system the thumbs down,
the ABC1 system still needs an overhaul. Any change will be a difficult
process, particularly if the new system entirely replaces ABC1. ’The old
system is ingrained among media buyers and advertisers’ says
The arguments for and against adopting the new system are well
The ’anti’ lobby is arguably taking an ostrich approach. It’s almost a
question of ’better the devil you know’. ’Familiarity brings quite a lot
of benefits,’ asserts Corrine Moy of NOP Research.
’Yes, it is outdated and, yes, it is widely used,’ says Bill Portlock,
planning director with direct marketing agency WWAV Rapp Collins.
’The ABC1 system is a good broad brush, but very clumsy; it’s never been
a totally sharp tool. But it’s part of our lingo now; it’s common
currency,’ he adds.
’How we segment a database, in terms of who wants the clients’ products,
and then prepare a precise model, is a sophisticated exercise, but we
still go back to ABC1. We speak ABC1, but understand what’s going on
Users call the shots
’I wouldn’t resist a new system,’ says Bill Blyth, director of research
at Taylor Nelson. ’But these things tend to be user-driven and everyone
is using ABC1.’
Roger Titford, strategic research director of marketing services company
Fast Marketing, has been using ABC1 for 20 years and is naturally rather
loyal to the system.
’It does depend on what you’re using it for,’ he concedes. ’If you’re
talking about consumers generally, it’s fine - and what I find strong
about ABC1 is that it suggests an outlook on life. But for certain
product fields you might need Acorn Type 32, Younger Empty Nesters or
High Income Families.’
And this is where the ABC1 system falls down. It was originally
developed out of government classifications that were current half a
century ago, and has been much derided over the past ten-15 years.
The system is essentially a definition of wealth rather than attitude
and, although it is perfectly adequate for broad consumer definitions,
as people have become better educated, move jobs more frequently,
confuse gender roles and have more disposable income, the lines have
become very fuzzy indeed. The idea of there being only six ways of
classifying the population is outmoded in the 90s.
The ’E’ classification today mixes state pensioners and unemployed in a
wide age spectrum, while ’A’ singles out top managing directors and
company directors. The real change has happened in between these two
classifications - to the Bs, Cs and Ds.
Consumers today have more facets to their lives that are more important
to them and these are difficult to grasp in a single classification. We
also identify more family classifications today, such as working women
and single households.
’More traditional socio-demographic criteria are no longer useful and
need to be updated,’ says Melanie Howard, director of the Future
Foundation, a consumer research, social analysis and commercial think
’Class, income and gender are no longer the accurate prediction of
consumer behaviour. People are not going to stay in their class
classifications through all their lives. Women are no longer classified
by the job of their husband, for example. People are getting married
later, are divorcing more frequently - the changes have been legion.
Values and attitudes are changing too.’
The jury is out as to whether the very complexity of the modern consumer
can be easily boxed into a single, easy-to-use classification method.
The vocal dissenters of Professor Rose’s report suggest that it has
shortcomings and would be difficult to implement.
’If a new system were to be adopted by government and influential
clients, then there’s no reason why it shouldn’t be influential,’ says
Simon Atkinson, Associate Director of MORI. But he adds: ’It would need
a strong consensus.’
Amsterdam-based ESOMAR issued a paper on new socio-economic
classifications but there has been little enthusiasm for it. Will the
research industry in this country suffer from the same inertia?
The Future Foundation’s Howard doesn’t believe there is a system out
there to accurately track the consumer’s complex nature, including
’It will help government analysis, but marketers will have to find new
ways of segmenting their own markets,’ she says.
’It’s a question of how marketing can catch up with reality. We need to
understand the differences. We’ve moved away from grouping people
together to explain their similarities. We are all becoming
individualised and sophisticated.
Companies are going to have to do more sophisticated marketing, which
has already started with the direct marketing industry through vehicles
such as loyalty programmes.’
One method being used by the Future Foundation to capture changing
values and systems is a technique called ’fuzzy clustering’, based on
It allows consumers to be recognised and defined in different ways
according to the time of day. Consumers can also inhabit different
categories in different parts of their lives, summed up by the example
of a Sun newspaper reader who is studying for a degree.
Our secret lives
’Everyone has a secret life,’ says Stokes Jones, editor of the Henley
Centre’s annual ’Planning for Consumer Change’ study, released this
’We try to warn marketers that their markets may be hiding. I may buy my
work shirts from Jermyn Street, my casual shirts at Muji and listen to
Sibelius and I’m all the same person,’ he adds. ’We’re more complicated
as consumers because we’re more complicated as people.’
A cornerstone of this year’s study is a classification of the complex
consumer called polyglotting, and this may well provide a part answer
for the research industry.
’Instead of looking at consumer identity we look at modes of acting and
behaving at different times,’ says Jones. It is effectively another form
of fuzzy clustering.
’We segment people according to occasions instead of what you think
their identity is all the time. The question is not what are their
backgrounds, but when will they be in this mode? When will they want to
chill out? When will they want to splash out?’
As well as polyglotting, the report identifies 20 new trends affecting
the modern British consumer. These include the need to deal with
increased stress, the blurring of age gaps, increased interest in New
Age philosophies, the need for ethical values in business and the way
consumers see brands as part of their identity.
Jones knows it will be a long haul to persuade media owners to move away
from social grades, however. ’It’s backed by them and they all know what
it isn’t telling them, but it’s the best thing they have.’ The basis of
his thinking is that consumers are changing at a rate faster than the
ability of companies to understand them, which in turn requires new
tools to track them.
’Everyone tries to fill in the cracks in different ways and tailor
research to their specific needs,’ he adds.
’We’ve developed a more intimate way of looking at consumers’ lives and
I hope they will incorporate polyglotting intelligently to evolve the
Already, research organisations such as the Henley Centre, the Future
Foundation and Research International are looking at ways of studying
consumer behaviour at close hand, either by living with them, filming
them or getting them to fill in diaries as they go through the day. All
of this shows how a replacement for the ABC1 classification will never
be enough on its own; marketers need more.
So, at the very least, a new system to run alongside, or augment the
ABC1 system, seems the most popular option.
This is already happening in the sense that most companies already have
in place their own, internal ways of segmenting their consumers as they
relate to their business. The problem is they can’t take this language
outside the walls of their HQ. Acorn, Fruits and Mosaic have also helped
One suggestion, from Taylor Nelson’ s Blythe, is to add new
classifications to ABC1 - say, job levels and educational levels - in
order to update the system.
Whatever the outcome, the rapid pace of growth experienced by
relationship marketing should force change of one kind or another. But
whether the research industry ends up with more confusion rather than a
more effective new classification system is a moot point. The new
choices are certainly testing everybody’s attitudes.
TOWARDS A NEW ABC1?
In his newly revised Social Class System, Professor Rose proposes that
consumers be divided into 17 narrow classifications by occupation,
taking into consideration employment relationships between managers and
the managed. These 17 classifications can be grouped into eight broad
1 Professionals (currently ABC1s): doctors, vets, teachers, lawyers,
librarians and social workers, plus business executives and successful
2 Junior managers in smaller organisations (currently C1s): journalists,
police sergeants and constables, plus hospital nurses, research
executives and market researchers.
3 Intermediate occupations (currently C2s): clerical workers,
secretaries, those formerly classified as skilled manual, plus dental
nurses, telephone engineers and computer technicians.
4 Small employers with fewer than 25 employees; the self-employed.
5 Supervisors of those in semi-routine and routine occupations, ie
working class supervisors.
6 Semi-routine occupations (currently Ds): sales assistants, supermarket
cashiers, all drivers and assembly line or factory workers.
7 Routine occupations: domestics, couriers, porters, refuse collectors
and all labourers.
8 Never worked or long-term unemployed.