This is sort of a continuation of last Monday’s blogologue where I wondered if the sourcing nation has a Prozac problem. While thinking about why the new solutions are being ignored when they should be the target of evaluations (and positive feedback, so that even if they don’t work for you today, they will tomorrow), it occurred to me that this is because there is a larger problem – the market is blind to what the true problems are, and most of the big vendors aren’t doing much on the user-education front. The reality is that if you don’t understand you have a problem, and in some cases a big problem, you won’t be looking for a new technology to solve it. Thus, the elephants in the room are often going unnoticed while the vendors focus your attention on the fuzzy bunnies.
In an attempt to make sure they don’t go unnoticed again, I’m going to pull down the blinds they’re hiding behind and expose you to three of the biggest elephants that your vendors might not want you to know about.
Optimization is not a set of reports that tell you the lowest cost supplier by unit cost or landed cost, the ability to calculate the cost of a random sample of award simulations and select the best one, or heuristically simplifying a model until you can run it in the framework provided to you by a third party provider that provides you with an engine – it’s the ability to allow an end-user to build a model that realistically models their situation, allows them to account for all of their costs and business constraints, and then solves that model using sound and complete optimization algorithms (such as those based on MILP) to come up with the optimal award across suppliers. Furthermore, it lets the user create multiple what-if scenarios to see what would happen if certain constraints were relaxed or new constraints were added and then lets the user compare those results side by side.
It’s not electronic invoice presentment and payment or supplier networks or e-Procurement that saves you time and money, it’s compliance – and that requires the ability to match invoices to items to contracts before an invoice is accepted and approved for payment. The fact that an invoice is from a vendor contracted to provide the items in question is not enough – office supplies distributors are notorious for overcharges and electronics vendors are notorious for not adjust pricing downward over time to insure their customers get the “best market price” that is in the contract.
Finally, when it comes to spend analysis, it’s the analysis! I don’t care how much cleansing, categorizing, and enrichment the latest product offers – it doesn’t mean diddly squat if you can’t do the analysis you need to do. Static top n vendors, top n categories, and top n commodity reports on one inflexible cube doesn’t cut it. You need to be able to build multiple cubes on the fly, on arbitrary dimensions of interest, and segment by range as well as rank. It’s not the top 10 suppliers with the greatest percent of spend, it’s the top 10 suppliers with the greatest variance in spend – chances are these are the ones overcharging you. It’s not the top 10 commodities with the greatest percentage of spend – it’s the top 10 commodities being bought off contract as this is where you need to curtail maverick spend first. It’s not the top 10 categories with the greatest percentage of spend, it’s the top 10 categories with the greatest variance in spend across departments or channels – as these are the ones that need better sourcing strategies.
So the next time someone tries to sell you a comparative reporting toolset on top of a simulation engine and pass it off as optimization, ask them these questions. The next time someone tries to sell you a supplier network or EIPP solution as the answer to all your procurement compliance problems, ask them how it automates three-way matching between invoices, goods receipts, and contracts and tags those invoices for human review that are questionable. And the next time someone tells you it’s all about the cleansing, say “no thanks, I’m looking for solutions, not religion“.