Callcredit Blog

Big data – Clean data?

Consumer Marketing Data

With “Big Data” being one of the most dominant themes in the industry today it must be remembered that however ‘big’ the data, it is worth very ‘little’ if any of the multiple database feeds are linked to out-of-date or un-contactable mailing addresses, telephone numbers or emails.

Having a process in place that cleans, screens and ultimately updates the data against the full range of national validation and suppression files available will ensure best practise and retain the true value of all data feeds.

Wikipedia defines “Big Data” amongst other things as “data so large and complex that it becomes difficult to process.” Because of this, it is often assumed that online data cleaning portals cannot process “Big Data”. But, that is not to say that these tools cannot process and update millions of records easily, quickly and cost effectively because  they can!

They can also take away the need to make upfront investments in the full range of industry suppression files, because all of these data resources are there to use, as and when you need them.

The best of breed online data cleaning tools also go beyond just cleaning data. They allow new variables to be added to an existing database, to make it bigger, better and more targetable. But one would not recommend appending any old data. It’s all about enhancing your database with valuable insights that can really add value to your data driven communications strategy. The best online tools will also help in this regard by allowing you to profile your current base (often for free!) and identify those key variables that really discriminate your file; those variables that will help drive response and those that will help drive sales and loyalty.

These could range from house price, interests, newspaper readership or geo-demographics through to contact channels such as email addresses and telephone numbers, which will help fill in the gaps in your data.

In deciding which are the best variables to append, an online profile of at least 1,000 records will identify the most predictive variables, but one would have thought that if you have “Big Data” then  more records would be available to provide a more representative profile.

So it’s not the case of “Big Data” being difficult to process online. If data is pared down to the necessary elements – full name, address and perhaps telephone number and email – online tools can be very effective at processing very large data files. They can also add up to 500 extra variables to help not only make “Bigger Data” but more “Insightful and Usable Data”.

Leave A Comment

Your email address will not be published. Required fields are marked *