FEATURE PROSPECT DATA
Collecting data online shows now sign of waning
Demand for survey data completed online remains high and is still growing, it seems. This sort of data is gathered in a number of different ways. While outfits like Transactis and Abacus receive their records directly from contributors to their co-op databases, other data owners either gather records as part of their own acquisition process (online or paper-based lifestyle surveys, guarantee registrations and the like) or simply pay for records collected online by lead generation specialists.
“Our data collection is predominantly phone- based,” says John Pooley, Managing Director at The Data Partnership. “It’s a little more expensive to generate an online lead because you have to buy the traffic through affiliate networks. It costs about £1 to £1.20 per person that fills in a survey.”
According to Pooley (pictured right),
competition between data providers means that the cost of a survey can be as much as £1.50. “We’re up
against Acxiom, DLG and the
others, we are all competing to get these leads,” he says. “You also tend to pick up a younger demographic online, whereas paper and phone collection has a strong bias to females and the over-50s.”
REPUTATION MATTERS With some pure email lists still gathered unscrupulously and non-compliantly using techniques like scraping, reputation is also a critically important factor when buying data that includes email contacts.
“The quality of email and other data collected online can vary greatly,” says Pooley. “We work with GB Group to validate survey data in real time as it’s collected.”
With fraudulent survey entries common on the web, Acxiom does not incentivise its surveys and uses its own offline database to check new records.
“Online needs to be validated properly, no question about it, and all our emails must connect to a valid individual at a valid address,” says Product Manager Stuart Goldsmith. “We simply won’t accept it if it doesn’t have that. You need to have the data in different channels to accurately be able to link between them.”
26 May 2012
own Core data pool, including all CallCredit’s Cameo consumer segmentations, risk data and models. Claimed to “contain more individuals with more contact touchpoints” than any other UK file, the new database sports some impressive numbers. “We’ve loaded over 164 million records in the last 12 months from data that we’ve sourced from 80 contributing partners across 10 verticals,” says Chris McDonald, Managing Director for Data Solutions at Callcredit Marketing Solutions. “We believe it is the strongest data set on the market.” An attitudinal survey gathered online adds an
interesting extra twist. Instead of written questions and answers, the AIM (Attitudinal, Inspirational, Motivational) survey offers alternative images to respondents to classify them into one of 900 types. McDonald says they have gathered just under 1m surveys so far. Over 60% of the data is verified by a minimum of three sources, and 84% has at least one contact detail gathered less than six months ago, with all landline numbers screened every 30 days. Automation plays a big part in the update and refresh process. For example, there are real-time email feedback loops to instantly replace hard bounces where an alternative email is present, feeds of online purchase data, and CallCredit will also load results from client email campaigns onto the database to aid future email list selection. “That will help fine tune our database information,” says McDonald. “We can run tight selections on our email file as we know a great deal about them. It’s real data from real people doing real things.”
As well as multichannel contacts, the file includes what McDonald calls “holistic addressing”. Individual records can include Facebook details, Twitter accounts, indicators of email response and even device preferences: do they use an iPad, a smartphone or a PC – or all three?
That means that, besides list selection and tagging, one of the database’s key applications will be as an operational reference file to underpin the company’s custom online targeting services, and CallCredit has made a significant investment in “agile systems” to support this. As well as handling the automated refreshes mentioned above, the company’s high-speed database platform and decision engine is able to work in real time with web content management tools to handle ad serving or offer management. “Everything’s moving online but clients don’t always capture enough information to help make offers,” says McDonald. “We can help recognise them in real time and recommend the best personalisation of that session. There’s no-one out there in the agency world that has that ability.”
www.dmarket.co.uk
“We’ve loaded 164 million records in the last 12 months from data from 80
contributing partners across 10 verticals.” Chris McDonald,
Managing Director Data Solutions, Callcredit
The combination of Acxiom’s ILU and Personicx segmentations also has great potential in identifying and classifying website visitors. However, any ambitions to track customers online using persistent cookies and classify them by the sites they visit are severely tempered by public sentiment and pending privacy legislation. “Recognition will be key so that they can turn
these browsers into real people,” says Goldsmith. “But it will have to be targeting at segment rather than individual level. It will have to be accurate, but without identifying people individually.”
STARTING POINT
Every consumer marketer has to make the most of the data that his customers supply, day in and day out, and online behaviour and purchase information is no different. But when that data is insufficient, incomplete or simply missing, that’s when external data providers can step in. Whether it’s a list of cold email addresses or informing sub-second decisions, working with the right consumer databases can sharpen any marketer’s competitive edge. n
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32