Friday Rant: What do 3-Way Matches, Spend Analysis and Recall Elections Have in Common?

When I first moved to the south in 2003, I saw firsthand where the old expression around things "going South" came from. Now I think the direction has turned. "Go North" is the new go South. Well, at least in election years.

In talking about going North, I am referring to the recent stories from Madison, WI. Yes, those elections, the ones with above 100% voter turnout (the equivalent of classifying 101% of your spend correctly on the first pass -- reminds me of something that a vendor might have tried to claim, but hold that thought for just a minute).

Getting back to "vote early, vote often, vote twice" in Wisconsin, I must admit that I thought such turnouts were only seen in Chicago and in the former Soviet Union? For those who claim that last minute voter turnout led to this, let me suggest that it is statistically speaking a completely unrealistic result -- it's so many standard deviations away from the norm that I think it is safe to say that this is nothing short of voter fraud. And sure enough it is, despite local political ploys to "get out the vote" by all means. Straight from the horse's (or Badger's, given the locale) mouth:

The law requiring voters to show photo identification at the polls is under injunction, and voters will not be required to show ID at the polls on Tuesday, June 5

How brilliant is this? Get lucky Bucky! Vote early, vote often, vote 'til you drop. The procurement analogy is that it's like prohibiting matching invoices to purchase orders and shipping documents before cutting checks to suppliers (at the risk of offending suppliers who wanted to do things right, but somehow typed in a wrong PO on the invoice -- no need to hold them responsible, we say, by Wisconsin logic). Assuming Wisconsin comes to its senses and starts to validate voters, it would be interesting to see a graph with turnout stats pre- and post-photo ID checks.

Just as PO and invoice matching has great benefit to driving compliance and avoiding fraud, the procurement analogy continues when you begin to apply spend analysis thinking to the voting outcomes:

  1. Dedupe the results: remove those who voted multiple times
  2. Purge the vendor master of inactive records: in this case, conduct some "digital reinterment" (i.e., return the dead to their resting ground, buried dangling chads and all)
  3. Apply parent-child linking: in the sense of identifying where those who voted actually live! I see a social media role (e.g., Foursquare) here to help us put neighboring state voters back on the bus to wherever they came from and prevent them from "checking in" in the first place -- or something like that

See how sound supply chain methodologies are applicable everywhere -- especially when shedding daylight on Wisconsin, "Going North" get-out-the-vote strategies? At the end of the polling day, it certainly feels like those UW-Madison Badgers are devilish little creatures now, aren't they, when it comes maverick voting.

Perhaps this is a case for the equivalent of Ariba, Basware or someone else who really gets a 3-way match to come in and clean things up after Emptoris, Bravo, Spend Radar, Zycus or whomever "classifies" the voter fraud to start. I say bring on "voting" transparency through the equivalent of P2P and spend analysis processes! And let's avoid things from ever "Going North" again. At least until the next generation of Daley is elected to fill Rahm's shoes in Chicago...

- Thomas Kase (and Jason Busch)

Discuss this:

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