Why marketers may still need to find a ‘true’ customer data platform?

With all the power of the major CRM packages, this may sound like a stupid question; but we have had multiple clients that use CRM vendors come to us recently because they are not getting the ‘true’ customer data platform, and hence single customer view, they need.

And these clients come from very different sectors including sport, financial services, and charity.

A single customer view that isn’t ‘true’ has some unfortunate consequences.

It becomes very difficult to accurately fulfil GDPR Subject Access Requests if Mr Smith appears several times in the system. It can lead to wasteful and irritating communications such as sending two emails to the same person at the same time. It will mean that you can be treating the same person as a high value customer and as one that has become dormant. It can mean you’re submitting an existing customer to a welcome programme because they changed their email address. It makes your dashboards describing customer recruitment and attrition inaccurate, and so on and so forth.

We believe that the problem has arisen for two key reasons:
1. the simple one is that the CRMs may let different users set up the same individual multiple times on the system without warnings, unless the contacts provided are absolutely identical
2. the more complicated one is that correctly identifying individuals is in fact far from simple

So, you may have an individual with a work email and the same individual with personal email, but they share the same mobile phone number and cookie ID.

They may use a different weekend and weekday name and address, but the same email. Or they may have several devices, hence cookie IDs, but a single email address. And so it goes on.

When we set about designing UniFida, our cloud-based customer data platform, we recognised that correctly identifying people is not only very important, but also very difficult.

So we decided to store as many different types of identifiers as we could. And most importantly to store the history of them, so that we never delete an identifier, unless of course someone is exercising their right to be forgotten.

It does mean that our system has to do a lot more work when new data arrives, because it has to look at all possible identifiers belonging to all personal records before deciding where to place new information.

And that can have interesting results. As well as bringing in new identifiers to an existing record, perhaps a new cookie ID, it can in some cases link together two people whom the system had been previously keeping separate. For instance two different email addresses can be found to have the same mobile phone number, and belong to the same person.

We call the process Purning, because we create a permanent URN, or unique reference number, for each individual, and that stays the same, even if over time all their identifiers may have changed.

As long as we can link a new identifier to an existing one, we know where to put the data that accompanies it.