Supplier Evaluation Matrix

A VP of operations has been evaluating a contract renewal for three months. The incumbent vendor has performance issues the client can name precisely. Alternatives have been considered informally. The client keeps finding reasons to delay the decision — more information needed, timing not right, relationship complications. The real constraint is discomfort with the decision itself, not lack of data.

Frame as a forcing function, not a research tool. 'You have enough data. What you don't have is a structure that forces you to weigh it the same way across vendors. Let's build that.' The client who resists ('I know what I think about each vendor') is usually the client who most needs the side-by-side comparison — intuition without structure often serves to protect existing preferences rather than surface the best choice.

Watch for blank cells in the matrix. If one vendor has significantly more blank cells than others, the client either doesn't have the information to evaluate that vendor or is avoiding rating them. Ask which it is before proceeding. Also watch for criteria scoring: if a client scores every vendor 3-4 on every criterion, they're not differentiating — they're averaging.

Start with the totals, but spend more time on the two or three criteria where scores diverge most sharply between vendors. 'Here's where your ratings actually separate them. What do those gaps tell you?' Then move to the final decision field: 'Given this spread, what's the decision?' If the client resists making one even with the data in front of them, the obstacle is no longer informational.

A COO has a long-standing supplier relationship that has become a dependency. The vendor underperforms on reliability and pricing relative to alternatives, but switching feels difficult — the relationship is personal, the transition risk seems high, and the executive has built their internal processes around this vendor's specific capabilities. They want coaching help navigating the decision.

Position the matrix as a way to separate the vendor relationship from the vendor performance evaluation. 'The relationship exists whether or not this vendor is the right supplier going forward. Let's look at performance separately.' Include the incumbent as one of the four vendors rather than treating them as the default. Some clients resist this — insisting the incumbent is 'different' — which is precisely the bias the matrix is designed to surface.

Watch how the client scores the incumbent on the scalability and risk assessment criteria specifically. Clients with incumbent bias consistently under-score challenger vendors on criteria where they lack direct experience while over-scoring the incumbent on the basis of known-quantity comfort. If the incumbent's highest scores cluster on familiarity-related criteria ('communication,' 'reliability of existing service') and alternatives score higher on forward-looking criteria, name that explicitly.

Start with the incumbent's scores on the two or three criteria where they underperform relative to alternatives. 'These are the gaps you've identified in the existing relationship. What would it cost to close those gaps with the current vendor versus with one of these alternatives?' Then move to the relationship factor: 'Separate from performance — what does the incumbent relationship require to maintain, and is that cost worth it?'

A director who has recently taken responsibility for a function that manages vendor contracts needs to make their first significant procurement decision. They have no established framework for comparison and are uncertain how to weight different factors. The absence of a framework means the decision is likely to be driven by whoever presents most persuasively rather than by structured evaluation.

For this client, the criteria column is where the work begins — before filling in any scores. 'Before we evaluate vendors, we need to agree on what matters. Let's build your criteria list first.' Walk through the ten categories on the sheet and ask which ones are most important to this decision. Add domain-specific criteria if needed. The client who has never done this kind of evaluation often discovers during criterion-building that they don't agree with their stakeholders about what matters most.

Watch for criteria that the client cannot define precisely enough to score consistently. 'Cultural fit' and 'innovation capacity' are examples — they mean different things to different evaluators and different things to the same evaluator on different days. Push for behavioral definitions: 'What would a vendor have to do or show for you to score them a 5 on that criterion?'

After scores are in, start with the criteria the client weighted most heavily. 'You said X was most important. Looking at the scores there — what do you see?' Then explore the overall spread: 'If the scores are close, what does that tell you about how well-differentiated your vendors actually are? And if there's a clear leader, what would it take to act on that?'