Data pooling gets mixed reviews. Although there is an obvious advantage in targeting a wider audience of prospective customers who have a track record of responding to direct marketing, companies have serious reservations about putting what they view as their ‘lifeblood’ into an open market.
This is particularly true in the UK business-to-business market, where data pooling is still relatively new. It is more popular in the US, where it is well established in both B2C and B2B.
Data pooling provider Abacus is a case in point. The company was launched in the US in 1990 and opened in the UK in 1998, providing a B2C service. In November 2004 it launched a B2B offering.
Nigel Bennett, business development director of B2B marketing for Experian, is cautious, as are his clients. ìIf we are not able to fulfil a particular data requirement for a client, we may partner with another provider for a specific purpose. But we would not pool customer information into a single data concept into which other members of the pool can dip.
ìThere is a belief among customers and potential customers that their data is their crown jewels and that in a pool, it will be used in a way they do not want,î he says. Demand for pooling reflects this. ìI cannot think of a single time a customer has enquired after data pools,î says Bennett.
He does, however, concede some benefits: ìCustomer records where respondents or organisations are known to have purchased a product or service means they are likely to be more responsive.
Pooling in B2C & B2B
There is a general feeling that data pooling works better in B2C than B2B, not least because it is proven ñ mail order catalogues lend themselves particularly well to this approach. But the jury is out as to whether the same principles can be applied to the business market.
Simon Lawrence, joint managing director of Information Arts, says, ìData pooling has been available for some time in the UK in the B2C market and in theory it is extendable. It is a well established principle in B2C ñ if someone responds to offers through direct mail they are likely to respond to other direct marketing offers. And as a lead, they attract a premium if they are known to be mail order responsive.
ìBut in B2B, people do not do retail therapy,î he says. ìAnd the decision-making process in business is complex and difficult to understand. Unless companies are selling the same product, they cannot extrapolate the data. I am interested in data pooling from the perspective of developing more insight into how businesses operate, how they purchase. But it is not as effective as in B2C, where everyone shares data anonymously and shares customer behaviour through understanding the transactions undertaken by consumers: there is direct relevance in extrapolating behaviour in purchasing.
Green-eyed companies
However, Simon Lawrence is unequivocal about one thing: ìIf it is going to work in B2B, two companies in the same market ñ such as Vodafone and Orange ñ could not be in the same pool. The jealousy with which companies guard their data is not to be underestimated.
ìIf companies cannot see a clear business advantage, it is not going to get off the table and Abacus has not advertised any resounding successes, apart from participants such as Staples, which sells office supplies,î says Lawrence. ìThey could do a better job by pooling data on a bespoke basis, to develop highly granular propensity models for specific clients.î
Abacus agrees that the strongest areas are office products ñ stationery and furniture; plus promotional items such as mouse mats, mugs and pens; training and seminars, which are driven by active mailers to decision-makers in those areas. And the company claims to have exceeded its objectives. Andy McDermott, managing director for Abacus, says, ìWe had a target of 50 companies by the end of last year and we have got 60, and it will be possible to do the same again this year.î
McDermott sees analogies between the B2C and B2B markets. ìOn the consumer side, with a new catalogue you see match-rates of around 90 per cent from those already buying from someone else; with fashion and clothing, that is typically 95 per cent,î he says. ìWe have not seen those levels matched in B2B yet but we certainly see 50 to 60 per cent and that is increasing all the time.î
Abacus asks catalogue companies to contribute their entire customer database ñ active, lapsed, enquiries, suppressions ñ with a list of transactions that have taken place since the person or company became a customer. McDermott says, ìWe do not want product level data, we need the value and dates of transactions. We are looking for the best and most responsive customers.î
Tit for tat
The Abacus pool is run on the basis of mutual gain, so only those who contribute data can take data out. ìBecause our data is transaction-based and we are loading new data every weekend, it is very dynamic,î says McDermott.
He highlights the much documented problem with B2B data ñ high churn, often estimated at some 30 per cent. ìOver three years, you could see an entire contact list change,î he says. ìAbacus automatically updates those changes. We are contact-driven and are getting budget holders, which larger companies need.î
McDermott also sees telephony as a fruitful sector for data pooling, potential also seen by Saul Parry, B2B senior list broker for Orca Media. ìThe main players tend to concentrate on keeping customers ñ rather than finding new ones ñ but if they can find companies that purchase laptops for a mobile sales force, for example, those companies have a high intrinsic need for mobile phones. But it would be necessary to see competitors’ propensity data,î he says.
ìWe find it difficult to get people to hand over transactional data. But we could take that to build a better model on how often companies spend and use that to target other organisations that purchase in a similar way,î says Parry. ìPeople can exclude competitors from using it, but if they do that they will not be able to use their competitors’ data.î
In Parry’s experience, models in the business-to-business market have become more scientific and more robust in the past year.
Forgotten predecessor
Abacus is not the first company to offer data pooling in this country but its predecessor was aimed at a very specific sector.
Launched in 1990 as Book Buyers’ Master File and run by Shane Redding, it held data from scientific and academic publishers such as Cambridge University Press, the FT and Blackwell’s. Purchasers were scientists, academics, librarians and business book buyers. But it was too expensive to run and it closed.
Redding, who is managing director of Think Direct and consults in direct marketing, says, ìThe processing cost of taking monthly downloads and putting them into a central pool did not stack up against the revenues. Things have changed dramatically and where it used to cost £20 per 1000, now it costs less than one pound per 1000 to extract data.î
Overcoming scepticism
Redding makes the points that ìTraditionally, closed pools ñ where only those who put data in could take something out ñ were viewed as an acquisition tool. But Abacus is more about allowing participants a much greater insight into their customers, especially looking at lapsed customers who have contracted with someone else ñ they may be back in the buying cycle.î
Vic Godding, managing director of B2B data specialists Developing Data is one of the few to give data pooling unequivocally good reviews. ìWe get extremely good service from Absolute Intuistic, who supply data pooling,î he comments. ìWe get regular reports, active payment and after several years of involvement with them, and they still ask permission to use the data on each exercise.î
The importance of suppression
Godding emphasises the need for an outstanding merge-purge option to reduce duplication and extraneous data. Developing Data runs The Business Suppression File, a rival service also offered by REaD Group under the same name ñ talks are under way regarding pooling the two.
Says Godding, ìThe hardest thing is to persuade data suppliers to use suppression files because they reduce the size of the list. Realistically, the responsibility should lie with the list owner.î
Godding sees an opportunity for more data pooling in technology. ìThere are so many brands that do the same thing, so you could go to three or four companies in the PC market and put together a programme to provide them with potential leads.î
Data pooling has enormous potential, not only in mass market areas such as computers but in carefully targeted sectors, where detailed propensity models can be drawn up for specific clients. But the first task is to overcome the reservations of potential participants so that their mailings can meet their full potential.
