New Techniques to Keep Data Clean
For many organisations, the process of validating information in databases – phone numbers, addresses, emails – has not moved on since they first began using digital marketing. Jonathan Erwin, Managing Director of internet marketing company Aspect Web Media, looks at some of the new techniques to increase database effectiveness as well as some best practice tips for email hygiene.
Despite the dramatic growth in internet marketing, many companies still use outdated tools for validating information in their databases. They still use the traditional tools like PAF validation to check postal addresses or rely on fuzzy logic and simple algorithms to check the accuracy of email addresses and phone numbers. Unsurprisingly these tools can leave you with false positive rates of up to 11% as well as other inaccuracies and omissions.
For example, we recently used our new Pure List email database screening service to clean a file containing 1.2 million active consumer email addresses. We discovered 157,373 potential threats, 24,570 known complainers and 2,602 bounces. The list was consequently reduced by 15% and this contributed to a reduction in potential deliverability issues with major ISPs such as Hotmail or Yahoo.
Clearly it’s important for companies to be confident that their databases contain a high proportion of valid email addresses. The reasons why are numerous but principally include reduced strain on IT resources such as storage and bandwidth consumption as well as costs savings from not delivering to duplicate, invalid or dangerous email addresses. Many marketing managers we speak to agree good email data hygiene makes the difference to avoiding damage to their ISP reputation and thus email delivery.
A new breed of sophisticated tools keep you safe
Traditional data hygiene tools are simply too limited in their capacity to clean data properly and have not kept pace with the growth and sophistication of the internet marketing industry. There is now a new generation of tools which not only do more they enable you to, for example, distinguish between hard and soft bounces. This means you can tell whether an email address really exists or if it is returning an automated response such as an ‘out of office’ message meaning it is still valid.
Importantly these new tools also remove threats in the email database such as serial ‘complainers’, spam traps, honeypots, seeds and click bots – so your reputation with your ISPs remains intact.
Validate before rather than after
Companies should look at the value of validating data from a data capture form (‘front end validation’) as opposed to the more traditional method of validating data once it is in the customer database (‘back end validation’).
We recently compared 500 ‘back end’ leads from a well known lead generation company employing online coregistration surveys to a similarly sized sample of ‘front end’ leads from our own coregistration survey. The back end leads yielded nine (9) customers who went on to buy from one of our clients indicating a lead to sale conversion rate of 1.8%. Whereas the front end leads yielded twenty-four (24) customers who went on to buy from the same client. With a lead to sale conversion rate of 7.6% we had effectively quadrupled the conversion rate for our client.
We note that companies often use back end validation to clean data because it is seemingly easier to collect leads from an online form this way. However, whilst implementing front end validation adds a small amount of extra cost the benefits far outweigh the costs as evidenced by the quality of leads being produced for our client.
Don’t use it? Lose it!
Missing data; stale or inactive files; enterprise silo data; inaccurate data entry – all of these “sins” can creep into a database and wear down the success of prospect and customer engagement initiatives. Regularly cleaning customer databases to root out old data is essential to retaining existing customers.
The process of appending is an effective means of removing old data. It works by comparing the information of two opt in databases to consolidate recent, up-to-date information. In other words, if the databases contain information on the same customer, it is possible to flag out of date information and replace it. Although appending is not a new concept, this technique is being employed in more advanced ways such as with the inclusion of behavioural data and is particularly effective when performed with data protection laws in mind.
Of course we shouldn’t underestimate the value of keeping in contact with the customers on your database either. It’s important to keep in touch with your lists on a regular basis – every two to four weeks at a minimum – by sending out newsletters, small surveys or topical tips. Remember, a period of inactivity could lead to opt-outs and complaints from consumers, resulting in the loss of good prospects.