Award the business to the cheapest compliant quote. It’s the default rule in most procurement functions because it’s defensible, quick, and easy to explain to finance. It’s also the reason a lot of manufacturers are carrying suppliers who cost them far more than they save.
The Costs the Quote Doesn’t Include
A supplier who is three percent cheaper but delivers four days late on a third of orders isn’t cheaper. You’re paying for that lateness in expedited freight, safety stock, rescheduled production, and occasionally a missed customer commitment that costs you the account.
The same applies to quality. A two percent reject rate on an inbound component means inspection time, return logistics, replacement lead time, and the risk that a defect gets through to a finished product. None of that appears on the quote. All of it appears in your cost base.
Then there’s the softer stuff that’s harder to quantify but very real: responsiveness when you need to change an order, willingness to hold buffer stock, how quickly they escalate a problem versus hoping you don’t notice.
What a Model Does With This
If you’ve been transacting with suppliers for a few years, you’re already sitting on the data needed to score them properly: purchase order dates versus goods receipt dates, quantity variances, inspection results, quality claims, price history, response times on quotes and queries.
A scoring model turns that history into a total-cost estimate rather than a unit price. It can weight the factors by how much they actually cost you — which is different for a fast-moving consumable than for a critical long-lead component. And it updates continuously, so a supplier whose performance is slipping shows up as a trend rather than as a surprise at the annual review.
For new suppliers with no history, models are less useful and you’re back to references, audits, and trial orders. That’s fine. The point of scoring your existing base is partly to know when a new supplier is worth the switching cost.
Using it Without Being Ruled By It
A supplier score is an input to a decision, not the decision. Strategic considerations — geographic diversification, developing a second source, a relationship you want to build for capacity reasons — will legitimately override the score sometimes.
What the score does is make those overrides conscious. If you’re awarding to a supplier scoring below an alternative, you should be able to say why. That’s a much healthier procurement conversation than either blind price-chasing or unexamined loyalty.
Where to Begin
Start by measuring on-time-in-full properly for your top twenty suppliers by spend. Most manufacturers think they know these numbers and are wrong by a wide margin. Once the measurement is real, the scoring follows naturally.
ticktick.ai builds supplier scorecards from your actual transaction history — delivery, quality, price, and responsiveness — and factors them into sourcing recommendations.
