Evaluation Matrix

An Evaluation Matrix is the practical scoring tool that sits at the heart of any multi-criteria tender evaluation. It translates the conceptual framework of award criteria and weightings into a concrete grid where each compliant tender receives a numerical score for each criterion, enabling objective comparison across submissions.

What is an Evaluation Matrix?

The Evaluation Matrix is not specifically defined in EU procurement directives but is standard practice across contracting authorities in Europe as the mechanism for implementing the scoring methodology described in the procurement documents. A typical evaluation matrix is structured as follows:

Rows: Award criteria and sub-criteria. Each top-level award criterion and each award criteria sub-criterion appears as a row, with its associated weight expressed as a percentage or score out of the total available points.

Columns: Compliant tenders. Each tender admitted to the award stage appears as a column. Tenders that have been found non-compliant or excluded do not appear in the award evaluation matrix.

Cells: Raw and weighted scores. Each cell contains the raw criterion score awarded by the evaluation panel (for example, a score from 0 to 10 or from 0 to 100) and the weighted score produced by multiplying the raw score by the criterion weight. The weighted scores sum to produce the overall tender score.

Summary row: Overall scores and ranking. The bottom of the matrix shows each tender's overall weighted score and its ranking relative to other compliant tenders.

The matrix is completed by the evaluation panel during the evaluation period. Panel members may complete individual matrices before moderation, with the moderated consensus scores entered into the final evaluation matrix that forms part of the evaluation report.

In absolute-scoring systems, each cell is scored independently of other bids. In relative-scoring systems (sometimes called "normative" scoring), the highest-scoring tender on a given criterion sets the ceiling, and other tenders are scored relative to that benchmark. The methodology used must be specified in the procurement documents.

Why the Evaluation Matrix matters for bidders

The evaluation matrix is the mechanism by which your bid narrative becomes a number. Understanding how the matrix will be completed helps you structure your submission so that panel members can score it efficiently and consistently.

A well-structured bid response that explicitly addresses each award criteria sub-criterion makes the evaluator's task straightforward: the relevant evidence for each matrix row is easy to find, reducing the risk that content is overlooked or attributed to the wrong criterion. A discursive response that blends multiple criteria together forces the evaluator to make judgements about what counts where, introducing inconsistency.

The weighting of award criteria structure embedded in the matrix tells you exactly what your overall score will be for any given combination of criterion scores. You can model the arithmetic before you submit: if you expect to score 7/10 on methodology (weighted 30%), 8/10 on quality assurance (15%), 6/10 on social value (15%), and 8/10 on price (40%), your projected overall score is (7 x 30 + 8 x 15 + 6 x 15 + 8 x 40) / 100 = (210 + 120 + 90 + 320) / 100 = 7.4/10.

Example

A Czech ministry evaluation matrix for an IT systems integration contract lists six rows: solution architecture (20%), implementation methodology (15%), testing and quality assurance approach (10%), transition and migration plan (15%), service management and support (10%), and price (30%). Five compliant tenders are scored across all criteria. The matrix shows that Tender C has the highest architecture score but is outperformed by Tender A on implementation methodology and price, resulting in Tender A achieving the highest overall weighted score despite not leading on any individual criterion.

Frequently Asked Questions

Is the evaluation matrix shared with bidders?

Not usually in full, because it would disclose other bidders' scores. However, in a debrief, the authority should provide your own criterion scores from the matrix and the overall score of the winning tender (and often the winning tender's criterion-level scores where this does not reveal commercially sensitive information).

Can panel members use different raw scores that are then averaged?

Yes. In many evaluation processes, panel members score independently before moderation. Individual scores may be averaged to produce a consensus raw score, or the panel may discuss and agree a single moderated score. The process used should be described in the evaluation methodology document.

What is the difference between an evaluation matrix and an evaluation scorecard?

The terms are used interchangeably in most procurement contexts. Some authorities use "scorecard" to refer to the individual panel member's working document and "matrix" to refer to the consolidated panel record. Where an authority uses both terms, check whether it intends to distinguish between them in its evaluation documentation.