We do like a bit of data segmentation!
Data segmentation and selections seem to be a constant thorn in the side of fundraisers. Why is that?
Over the years I’ve seen plenty of dead people who’d been mailed by some of our clients, and I’ve often seen that the right people had been selected but they were in the wrong segment. These are just two of the most common problems, but there are many others and we have cracked them all.
First off, I want to dispel a myth. Not matter what you might have been told, it is often true that only the simplest selections can be done straight from the database – such as “mail all known females”, or “mail everyone in Bishopsgate”. There are exceptions to this, but even they have some caveats. If you want to use more than just one or two segments and you need segments based on granular transactional history, then the data will need to be manipulated outside of the database.
But before you try and work out the where, you have to think about the what and this is where the data brief comes in.
In fact this is where the detailed, accurate and unambiguous data brief comes in.
Think of a data brief as a point of reference for everyone involved. If person A refers to an active cash supporter a properly constructed brief should make it clear to person B what that actually means.
But here are some of the problems I see with many data briefs:
- they are poorly constructed
- they contain very little detail
- they are often an amended previous brief with erroneous stragglers left in
- they are written in a fundraising-style vernacular
So, a question: how can it be possible to ensure that each selection is consistent when the terminology, which is key to our understanding, is inconsistent?
And the answer: it isn’t possible.
Back in the day, at Shelter (when I had a proper job!); we realised this was a problem, so we compiled a glossary; a definitive list of terms relating to supporter groups and how they were interpreted in data terms, this glossary introduced a common understanding across the whole fundraising team, for instance; “active” meant a donation made in the last 24 months, “cash” meant donations to cash appeals, website, social posts, unsolicited, but not donations/payments for raffle tickets or sponsorship or community group fundraising…and so on.
Remember the old chestnut (well, my old chestnut), you want to ask the right people, the most appropriate thing, in the best way. This is how you maximise your fundraising efforts.
The Glossary took a while, but it was time well spent. These definitions are the backbone of your selections, and they don’t change.
When Geoff from The Special Team comes strolling over and asks for a random pool of data (even he’s not really sure what he’s asking for), he can be directed to the wonderful world of the brief and the segment glossary.
“Wow!” says Geoff, “…this makes my life so much easier and if I use these clearly defined segments I can analyse the response properly and benchmark it against other activities, which means I’ll be able to see if I invested in the right activity.”
The best briefs I see use terminology from the agreed glossary. Questions can be asked and answered, using the glossary, as the brief is being drafted. “Regular givers; are these direct debits and standing orders? And what about regular CAF donations, do they count?” All the information will be at your fingertips – well in the glossary!
Drafting subsequent briefs will get quicker over time, you won’t need to revisit many of the questions as you will have that agreed understanding – in writing! And it helps to build team spirit – but that is for another day! For now, though, this process will give your data brief the best possible chance of producing the segments you need to get the best out of your fundraising budget – and then you can get onto the analytics!
