Accurate data is a critical component in any business marketing activity, so it’s amazing how little importance businesses still place on it.
As Nigel Bennett, business development director of B2B marketing at Experian, emphasises: Without accurate information, companies waste time and money, lose opportunities, damage their brand and irritate customers.
Terry Hiles, managing director at Capscan agrees: An effective marketing database must have three fundamental elements; accuracy, completeness and good management.
Quality data enables businesses to measure campaign effectiveness, customer loyalty and profitability, comply with legislation and create accurately-targeted marketing campaigns. So what can businesses do to ensure that their data is kept accurate, clean and relevant?
This article analyses the four main approaches and access their relative advantages and disadvantages.
1. Cleansing data using telemarketing
How it works: Teams of people at a call centre work down a list of data records and call each business to check whether the information held about them is correct. According to John Price, managing director at Price Direct, even firms that have signed up to the Corporate Telephone Preference Service (CTPS) can be contacted by phone for the purposes of checking data quality and ensuring that information is recorded correctly.
Pros: As callers actually contact the business, factual information, such as phone numbers and addresses, can be accurately ascertained. Phoning also allows the caller to enquire about ‘soft’ information such as the best contact for particular functions.
Cons: Telephoning is time-consuming and therefore expensive. Subjective information is less accurate as employees may not know specifics such as contact details, may give agents the wrong information, or may refuse because company policy forbids such information being revealed.
Call centres also rely on their agents being consistent, and individuals differ in their decisions and opinions. Telemarketers can also introduce errors into a database by mis-keying or misunderstanding expressed preferences.
When it’s best used: Businesses are best advised to use telemarketing techniques to improve data quality when their database is small, product unique or the service unusual; therefore, the price of data is relatively high.
2. Outsourcing datasets to specialist bureaux
How it works: A copy of database content is transferred electronically to a specialist bureau. The bureau checks it against a reference database such as those created and maintained by the Royal Mail, Experian, Equifax, etc. (which the bureau is licensed to use) then, once records are matched and errors are removed, the clean accurate database is returned to the customer.
Pros: Their access to comprehensive business data allow bureaux to add value to a database by creating new fields such as CTPS registration, turnover and number of employees. Martin Jaggard, account director at The Database Group, says: By adding company registration numbers we can create a corporate tree and make the links between companies much more apparent.
Bureaux work quickly and allow normal work to carry on as usual at the originating business.
Cons: Subjective information such as named employee contact details are more difficult to check because of the volatility of the market. Bureaux may also be busy and unable to clean data exactly when businesses require it.
Reputable firms enforce confidentiality and non-disclosure agreements but those handling sensitive information might be unwilling to let third parties handle it.
When it’s best used: Sarah Bearder, CEO at Datactics comments: Although not often possible, cleansing is most useful as close to the origin of the problems as possible. For instance ‘fuzzy’ searches, turnback of malformed input, validation, etc. is best at point of entry.
Bureaux often act as database auditors and advisors suggesting improvements and useful additional fields, depending on the nature of the business.
3. Specialist cleaning applications
How it works: Specialist cleansing applications typically work by matching data and comparing elements such as spelling, post codes and surname. Different fields have different levels of importance when matching, and these are indicated by a point score. Once the match is completed, the scores of the two sets of records are compared. If the score of the records being checked is high enough, the assumption is made that the company or business is the same and other pieces of information can be transferred to make a fuller record. As Chris Cuffe, managing director at HelpIT, says: What people are paying for with matching software is the quality of the matching algorithms and the sophistication of the evaluation.
Pros: Using a business’s own resources can be more cost-effective than using a bureau; businesses retain control of their own data, can clean it whenever they wish and transfer errors to a third party don’t occur. According to HelpIT, the average B2B database holds 10-20,000 records and matching using standard defaults could take as little as 10 minutes.
Costs can be very low as packages aimed at SMEs are particularly inexpensive. Once a business has purchased the software, it can use it as often as it chooses.
Cons:Despite training packages and online help, businesses purchasing software can feel isolated from expert assistance. The initial learning curve may slow the process at the start and customising the software may be difficult.
When it’s best used:Large firms with expert IT staff, small firms with a limited budget and businesses handling sensitive information may prefer to handle data cleansing inhouse.
4. Matching against suppression files
How it works: Records from a client database are compared with one where records are flagged if the business concerned has moved, ceased trading or otherwise no longer exists at that address, or if the data can be clarified some other way.
Pros: Suppression services avoid inappropriate mailings and so prevent waste, save money and preserve the good name of the brand. More effective targeted communications should also to have a better return on investment.
Cons: Suppression files risk identifying businesses as moved or ceased trading when they might still be trading, perhaps under a different name.
A suppression service alone cannot maintain data quality across all fields because it does not cover the breadth of information that, for example, a specialist data bureau would handle.
When it’s best used: Creative marketing material is costly. Suppression file matching ensures that the marketing budget is not wasted on DM material that doesn’t reach the intended recipient.
Purchasing data
How it works: If a database is compromised, businesses may consider scrapping it and starting again with data purchased from a third party. Before doing so businesses should audit their data or get a specialist bureau to do so and work out costs.
The analytics team from D&B, for example, profiles the best target companies and identifies the prospects in the client database that match them. Businesses can then purchase complementary data and avoid paying for information they don’t need.
Even the best database, once purchased, will decay unless properly maintained.
Pros: New data should be clean, appropriate and up-to-date. Bennett of Experian comments: Using it should result in little wastage, so you save money and prompt a high response and make money.
Cons: If they scrap their entire database, businesses could lose important and tailored information although if they can’t find the information anyway it’s of no use to them.
Businesses are totally in the hands of their data supplier. Purchasing data is costly, so businesses also need to be confident that they can merge additional data seamlessly into their own database.
When it’s best used: If a database is old, has rarely been cleaned, and has not evolved with the business, then companies might consider buying in. Corrupted databases may also be more easily replaced than repaired.
Complicated business mergers might be best served by a new database, although it’s likely that a specialist can merge existing databases into a new and useable form and retain useful details.
Looking at the facts
Bennett of Experian estimates that there has been a growth of about 12 per cent in businesses that actively maintain their data quality, while Jaggard at The Database Group feels that the number of new entrants to the bureau market is slowing down.
Certainly, D&B customers want their data to help them predict demand for a new product or identify those customers likely to upgrade. Rather than just servicing campaigns, such clients want to use their data to make important strategic decisions about where to focus their time and resources. Techniques are certainly becoming more sophisticated, but many businesses are not aware of what’s available and so aren’t using them to their full potential.
Business data quality is not only about software and hardware, it’s about attitude. D&B sees companies organising their data at the beginning of marketing campaigns so that they deliver the right message, to the right people, in the right way.
Staff are beginning to take responsibility for the quality of the data they use, but there is still a gap between what people say is important and what they actually do about it.
Martin Jaggard at The Database Group explains: Data quality is so much about confidence. If people don’t trust the data, they will start keeping their own records, which means A. the company doesn’t benefit and B. when they leave they take their records, and possibly your customers, with them.
The statistics are frightening: business data decays at an estimated rate of 35 per cent per year.
Companies House claims that one business closes every three minutes; TRG Strata suggests that 60-70 per cent of B2B mail contains one or more errors and DQ Global that 42 per cent of CRM failures are due to poor quality data. For data quality to be properly maintained, its importance needs to be built into the values of businesses. According to a PWC Global Data Management Survey, business executives attribute 37 per cent of their organisations value to data.
Unless someone takes responsibility for this valuable asset, the best kit in the world won’t help that much.
