Amazon is the standard against which other online experiences are measured.  When donors say “I want to have my information saved on your site so I can donate without re-entering it” or “I want to get my acknowledgment from you rapidly with no errors,” the “like Amazon” that could follow these statements isn’t any less potent for being unspoken.

To these entreaties, we sigh.  We say “if I had a data infrastructure like Amazon’s, of course I could deliver better experiences.”  So many opportunities to become more donor-centered fall by the wayside because our systems aren’t ready.

Here’s my news flash: Amazon’s systems weren’t ready either.  When Amazon started stocking kitchen equipment, they were doing so in warehouses set up to handle books.  As a result, carving knives shot down chutes into sorting machines with all the mayhem you’d expect.

And in Amazon’s much-vaunted database, each knife had to be coded: is it a hardcover knife or a paperback knife? *

Yes, making technology do what you need it to do is a challenge.  I once inherited more than 20 databases that each had a different sliver of data on an organization’s constituents.  (That is, there were 20 until I started killing them off one at a time, like weekend house guests in an Agatha Christie novel.   There’s a decent guide to this process here.)

As with Amazon, though, neither database design nor other technological foibles should deter your donorcentricity designs.

With your technology, you are constantly striving for a better state of broken. Nothing will ever be perfect, as we are continually striving for more and better while a flood of data comes in.

Like the duck, however, you want to seem calm on the surface even as you paddle mightily below.

So here are some arguments to marshal the next time your attempts to collect feedback, learn donor identities, or understand your donors’ commitment to your organization are delayed because the data are too messy:

We don’t want to pave the cow path.  Cows are capable of walking through a pasture in a straight line, but don’t.  They meander.  Pretty soon, a path is worn in the pasture that may be the least efficient path between two points.  That’s why IT folks, especially those from Wisconsin, warn against paving the cow path – making inefficiencies more permanent.

This means you need to be setting the requirements for your database and what you are going to collect first, before custom fields are set in stone (or, at least, more expensive to change).  It would be illogical to set up a database that didn’t have a last name field, then add it in later – why would you do so with vital marketing information like whether the constituent has the disease your organization combats or is a cat or dog person?

These data are more valuable to segmentation than transactional information.  If you had to choose between knowing how committed to your organization a donor says they are and the details of their last donation, you’d pick the former.  There’s a great free webinar up here from DonorVoice UK about how Amnesty Belgium was able to increase their F2F six-month retention rates from 60% to 80%.  In it, they found that self-reported commitment scores were more predictive of future giving that current giving.

Similarly, as I talked about in The Agitator yesterday, you’d rather have donor identity at the top of your segmentation tree than RFM analysis.

Yet your databases record the transactional easily and the self-reported at gunpoint.  This is the case to make to help flip that around.

It’s worth the pain. With these large-scale data projects, there’s often a period of suck.  For example, to get two databases to talk to each other, for the near term, you must export a CSV file over here and import it over there (until the API gets built).  Volunteer to be that person.  With a little knowledge of Excel VLOOKUP functions, you can even combine disparate files to create what you need.  Yes, it’s pain in the short-term, but when it comes time to build the API, you’ll know exactly what fields you need where.

Or, as Amazon might have put it, it doesn’t matter if you classify the knife as hardcover or paperback, as long as the knife ships.

* If this is a story that interests you, I’m a fan of Brad Stone’s excellent The Everything Store: Jeff Bezos and the Age of Amazon.