Sometimes we’re in meetings where we’re chatting with really good engineers, and we’ll get comments such as “Well, we’re pretty good engineers — what you’re doing doesn’t seem that hard, in fact we’ve done a lot of the work already”.
Totally fair point, and often even true. Frakture, and more generally automation, really isn’t super-secret sauce. It doesn’t have the network effects of fax or Facebook, or proprietary ranking algorithms like Google (no, they don’t just use PageRank).
But a great example today came up that really got me thinking about WHY it’s valuable.
We’re working with a client automating data between a payment processing system and their CRM. Pretty common scenario. BUT, the database design of the CRM is …. horrible. Limited character database names. Inscrutable columns. And since this is the real world, forcing the client to change their CRM is unrealistic.
So, our engineer had to HUNT. He had to look in every table, every column, for the (undocumented) location of a unique identifier that he could use as the core of the automation. If you’ve ever hunted through Google developer documentation, or searched for the ever elusive QuickBooks unique id, you’ve got some sense of this — there’s SO MUCH there that the hunt becomes the long pole in the tent.
With automation, much of that hunt has already been completed …. and it’s our role to figure out which are the long poles going forward, so you can focus on the stuff that matters.
And not look for the column “UTLCNO” in the table “OOUTLOEP” (the actual answer).
Start automating, and stop hunting UTLCNO’s.