±±¾©Èü³µpk10

The data is still trickling in from local councils, so few in the industry are prepared to hazard a guess at the likely rate of opt-out from the Electoral Roll. Figures bandied about range from 10-25 per cent nationally, though at local level there is likely to be more variation.

Demographic factors aside, there were big differences in the way the opt-out was presented to the public.

But even if some councils are found to have put the case for opting out a little too forcefully, the industry is having to come to terms with what effect the opt-out will have.

At data supplier EuroDirect, managing director John Dobson believes it may be less than expected. "If you assume a 10 per cent opt out, that means the edited version of the files we buy will be 90 per cent of the market," he says. "I can't think of anybody who wants to mail 90 per cent of UK market. And in fact, the last thing you want to do is mail the 10 per cent of people who do not want to receive direct mail. So the opt-out will ultimately improve response."

Dobson says that EuroDirect will use the edited version of the Roll for mailing purposes, using its 2001 Electoral Roll data to backfill the edited version, not for mailing, but for verification.

For the past five years, EuroDirect has been selling and developing Data Exchange, a data-sharing club with 70 different (and confidential) data feeds and 22 million records. Since its launch, a number of lifestyle variables have been added to the non-Electoral Roll-sourced names and addresses.

But Dobson insists that Data Exchange is not an alternative to the Electoral Roll. "You don't need an ER alternative," he says. "If you want ER data, 80-90 per cent will still be more than enough for everyone." For 100 per cent coverage, Dobson says, clients should eschew mail for door-to-door distribution.

Alternative lifestyles

At Equifax, the alternative to the Electoral Roll for validation and pre-screening is its Insight Verification database. CACI meanwhile has teamed up with lifestyle data company Claritas, which collects consumer data via warranty cards, to create the Consumer Register, which it claims offers "the first comprehensive database of individuals to provide complete and up-to-date coverage of every UK household".

The Consumer Register uses CACI's 2000 Electoral Roll data, cleansed to identify the 10-15 per cent of people who are likely to have moved since the data was collected, supplemented by several other databases.

These are Claritas's BehaviourBank and Lifestyle Selector Databases (more than 24 million individuals in 16 million households); Dataworks' High Net Worth file (14.7 million UK individuals, based on the share registers of the UK's FTSE 350 companies); and The Bounty File (the largest UK young family database, comprising six million records).

"The databases are complementary, rather than overlapping," says CACI marketing solutions division managing director Paul Winters. "The share register is predominantly male and wealthy, the Bounty database cuts across social groups with a female bias, and Claritas covers all social groups."

Experian has developed National Canvasse, made up of compliant register records, combined with name and address level data from Experian's Canvasse Lifestyle database, plus data contributions from partner organisations.

Experian senior product manager Patrice Bendon says the aim has been to try to create a seamless transition for clients from the Electoral Roll to the next stage.

"The Electoral Roll 2000 runs till this month and then it will be National Canvasse," says Bendon. "We have tried to replace like for like. The volume is pretty similar and the data variables are the same, so for a client it hardly looks any different."

The higher the opt-out rate, the more attractive these alternatives will look. But what if the opt-out rate is low, in single figures perhaps?

Won't all the alternatives become obsolete? Not so, says CACI's Paul Winters.

"We have sunk some costs into sourcing data, writing technical systems to handle the data feeds and the merge logic," he says. "If the opt out rate was five per cent, clients would probably not be prepared to pay the processing costs to find that 5 per cent." But Winters points out that the opt-out could be higher next year, so it was worth developing an alternative.

On the client side, Chris Wright, head of customer development at BUPA, which has used the Roll in the past for acquiring new customers, cautions against making snap decisions before the actual level of opt-out is known.

"We will not put specific plans in place until we see what percentage do opt out and how that varies by certain key parameters," he says.

And Wright agrees that the opt-out could have a positive impact, both in removing those with an "instinctive rejection" of direct mail, and in making direct marketers reassess their options.

"There are lots of other routes to market," says Wright. "DRTV and radio for example. This is a chance for companies to look again at how they reach consumers and reconsider all potential routes to market. That's a great opportunity."

CASE STUDY: BRIGHTHOUSE

BrightHouse (formerly Crazy George's) is a credit agreement retailer, specialising in furniture and electrical goods to the lower end of the market, who are often excluded due to low credit rating.

In 2001 BrightHouse teamed up with EuroDirect to create a prospect database.

The existing BrightHouse customer base was analysed for specific characteristics, so that prospects could be selected by being a 'lookalike' of existing customers.

Where new store locations were planned, combinations of EuroDirect's geodemographic CAMEO Classifications were overlaid on to the selection, to pinpoint BrightHouse's target audience. To ensure that the prospects targeted live in close proximity to a BH store, drive-time analysis was used based on previous analysis on how far customers typically travelled to a BrightHouse store.

Once the prospects had been selected, and the database built, Electoral Roll occupier updates were repeatedly carried out. Household and individual communication histories were also updated with campaign details, which then triggered response analysis.

Revenues attributable to five mailings run in 2001 totalled £762,500.

A similar recruitment campaign was carried out in 2002.