Sales & PartnershipsFree
Vendor Evaluation Matrix Response Template
Their scoring sheet filled in with your evidence, and the criteria written so that nobody but the incumbent can pass them.
River's matrix response works inside the buyer's own spreadsheet rather than converting it into yours. Paste their criteria, their weights and their scale, and it comes back completed row by row, each answer carrying the specific evidence behind it and written in the language the criterion uses. Alongside it comes a short document naming the rows where you score badly, the ones where the score cannot move, and the reframing worth offering on each.
Before any of that it scores you. Every row gets an honest mark on the buyer's own scale and a realistic ceiling, and the difference is the points actually in play there. Two things fall out. Most rows have no points in play at all, because you already score full marks or cannot move by the deadline. And some rows are capped for everyone except the incumbent, which is a different problem and needs a different answer.
Built for the bid manager with four days and thirty-four rows, the account executive who has just been sent a spreadsheet with no covering note, and the proposal lead deciding where the writing time goes. Run it the day the matrix arrives, while the clarification window is still open and a badly written criterion can still be questioned rather than worked around. The criteria themselves come out of the document with the requirements matrix, and whether to bid at all is the bid decision.
Why some rows cannot be answered well
A criterion that names a product rather than a capability is the tell, and the one buyer whose rules on this are published is the federal government. Agencies are told to state requirements in terms of functions to be performed, performance required, or essential physical characteristics, and to include restrictive conditions only so far as their needs require. Commercial buyers are bound by none of that, which is exactly why their matrices carry so many of these rows.
The same rules give you the reply. Where a brand name is used, the description must also carry a general description of the salient physical, functional or performance characteristics an equal item has to meet. Every wired criterion has a real requirement underneath it, and the reply is to name that requirement and answer it. Native integration with a named platform is a request for reads and writes at an acceptable latency, which is answerable.
Calderbank Mutual, 34 criteria across six sections, weights summing to 100, scored out of five, shortlist at 80. An honest self-score comes to 62.2. Write perfectly on every row you can move and you reach 71.6, still 8.4 short. Four criteria worth 22 points cap everyone but the incumbent at one out of five, so they take 4.4 of a possible 22. Restate those four as characteristics and score four, and the total is 84.8. Four sentences beat thirty-four answers.
How it works
Paste their matrix
Criteria, weights, the scoring scale and the threshold, exactly as the buyer wrote them.
River scores you first
An honest mark and a realistic ceiling on every row, before a single answer.
Find the wired rows
Criteria nobody but the incumbent can score on, and the weight they take off the table.
Answer and ask
The rows written where points move, and the clarification questions on the rows that are capped.
What you get
- Their matrix returned complete, in their format, with a source named on every row
- An honest self-score per criterion, on the buyer's own scale, before anything is written
- Points actually in play per row, so the writing goes where the score moves
- Criteria capped for everyone but the incumbent, with the weight they lock up
- A restated version of each wired criterion, written as a characteristic you can meet
- The rows you genuinely lose, and the reframing to offer rather than a claim
Common questions
Should we really tell them a criterion is unfair?
Not in those words, and not as a complaint. Ask a clarification question that offers the characteristic behind the requirement: whether evidence of reads and writes at a stated latency would satisfy a native integration row. Buyers accept that more often than people expect, and where the evidence has to be produced rather than cited, a sized pilot is where it comes from.
How can you score us before we have written anything?
Because the score is a property of the evidence, not the prose. A row asking for a current audit report is a five if you have one and a two if yours expired in March, whatever the paragraph says. Scoring first shows that eleven of the thirty-four rows in the worked example were already at full marks. The narrative sections around the matrix are the proposal pack.
We write the same care into every row. Is that wrong?
It is expensive. In the worked example fourteen of the thirty-four rows had no movable points, holding thirty-one of the hundred weight. At two hundred and fifty words a row, that is three and a half thousand words of an eight and a half thousand word response that cannot change the outcome.
What counts as a wired criterion?
Anything whose ceiling depends on who you are rather than what you can do. Years serving one buyer segment, a named platform, a radius from head office, a certification tied to one product family. The test is simple: could the incumbent score five, and could anyone else, however good, score more than one?
The matrix arrived as a locked spreadsheet with no weights.
Then the first clarification question is the weights, and asking for them is normal. Without them every row looks equally important, which is the condition wired criteria depend on. Where weights are genuinely withheld the response assumes them equal, says so on the face of the sheet, and ranks rows by how much of the submission they occupy instead.
Does it reuse our existing answers?
Where they fit the criterion as written, yes, and where they do not it says so rather than bending them. An answer written for a different buyer's question scores badly against this one even when the underlying fact is the same. Keeping that library usable is what the answer library handles.
What if the arithmetic says we cannot win?
Then you have learned it on day one rather than after four days of writing. In the worked example the four wired rows put the shortlist out of reach on the numbers, and the only move worth making was a clarification question. If it stays out of reach, the loss is at least documented for the debrief.
Vendor Evaluation Matrix Response Template
Fill in the form and your workspace opens with the work already underway.