Reading public tender evaluation criteria is the disciplined reconstruction of how a contracting authority says it will admit, compare and select tenders. The analysis separates exclusion and eligibility conditions, minimum technical or response requirements, scored award criteria, price or cost treatment, tie-breaks and procedural steps. It preserves the buyer’s exact language, hierarchy, weights and evidence requests, identifies ambiguity and maps every criterion to an owned response and proof plan.
Teams often copy a weight table and call the evaluation understood. They overlook a pass/fail requirement in an annex, treat company eligibility as a scored strength, or assume a high-weight heading reveals how subcriteria will be judged. Price formulas are modelled from memory. The response spends words on broad value while the criterion asks for named people, a method and evidence. When the documents do not disclose a detail, the team invents a hidden scoring model and optimizes against its own fiction.
Build the evaluation model only from the complete current tender documents and jurisdiction-specific rules. Separate gates from comparative criteria before estimating score opportunity. Reproduce disclosed wording, weight, scale, formula and evidence without “improving” them. Decompose each criterion into observable response obligations and identify the proof and decision owner. Use scenarios for disclosed mathematics and uncertainty, but never present an inferred evaluator preference as fact. Raise material ambiguity through the authorized clarification process.
Evaluation layers
Separate admission, compliance and comparative evaluation
Create distinct layers before interpreting score. Exclusion grounds concern whether an operator may participate. Eligibility or selection conditions concern capacity, qualifications, financial standing or experience. Technical specifications and response instructions define minimum offer and format requirements. Award criteria compare admissible tenders. A procedure may also contain minimum scores, oral presentations, demonstrations, verification, clarifications or tie-breaks. The governing documents and applicable law determine the exact structure, so use these categories as an analysis aid rather than legal conclusions.
Record the consequence stated for every item: exclusion, inadmissibility, correction opportunity, zero score, reduced score or evaluator discretion within a published scale. Search beyond the main table. Requirements can appear in forms, annexes, draft contracts, price schedules and portal fields. Do not assume an eligible company receives evaluation credit for its general experience unless that experience is part of a stated award criterion. Conversely, do not ignore a low-visibility mandatory declaration because it carries no weight. Gates protect entry; criteria govern comparison.
| Layer | Main question | Failure can mean |
|---|---|---|
| Exclusion | May the operator participate? | Exclusion |
| Eligibility | Does the operator meet required capacity? | No admission |
| Minimum specification | Is the offered solution compliant? | Inadmissible offer |
| Response form | Was the offer submitted as required? | Rejection or correction |
| Award criterion | How does the admissible offer compare? | Relative score |
| Threshold or tie-break | Does the score advance or separate? | Elimination or ranking |
Scoring model
Rebuild only the evaluation model the buyer has disclosed
Transcribe the hierarchy exactly. A forty-percent quality heading may contain several separately weighted subcriteria or one holistic assessment. Capture weights, ranges, order of importance, scoring scale, qualitative descriptors, minimum points and rounding. For price or cost, record the exact formula, reference price, treatment of taxes, options, quantities, life-cycle elements and abnormally low offer rules where stated. Reconcile the arithmetic and run controlled examples. Preserve source references beside every model input.
Distinguish calculation from judgment. A published price formula can be modelled precisely with scenario assumptions. A qualitative scale can support an evidence-strength assessment but not an exact predicted score unless enough official descriptors exist. Do not invent competitor prices or hidden evaluator preferences and present them as facts. Use ranges and name uncertainties. If two documents conflict, a scale is missing or the relationship between parent and subcriterion is unclear, submit a neutral clarification within the buyer’s rules rather than privately choosing the most favorable interpretation.
- Preserve criterion hierarchy and exact buyer wording.
- Validate that weights and subweights reconcile.
- Model price only with the disclosed formula and stated inputs.
- Separate calculable values from judgment assumptions.
- Raise material conflicts through the authorized channel.
Response planning
Translate each criterion into observable evaluator evidence
Decompose the criterion grammar. “Quality of implementation approach, including governance, transition and risk control” asks for at least four observable elements: the approach, governance, transition and risk treatment. Identify required deliverables, named staff, dates, methods, metrics, examples, certifications or demonstrations. Map each element to the response location and limit. State whether the evidence is current, applicable to the offered scope and authorized for disclosure. A criterion receives an accountable content owner and the specialist or authority needed to validate commitments.
Allocate effort by decision value and evidence gap, not weight alone. A high-weight criterion with mature proof may need less new work than a medium criterion requiring a new delivery model. Mandatory low-weight areas still need complete compliance. Define what a strong response can honestly demonstrate and where the organization should narrow a claim. Use the buyer’s terms accurately, make required information easy to locate and reference appendices precisely. Reviewers should be able to trace every planned claim to the criterion and every material criterion to visible proof.
| Field | Question | Control |
|---|---|---|
| Criterion element | What exactly will be judged? | Buyer wording |
| Response location | Where will it be found? | Section and limit |
| Claim or method | What will the offer demonstrate? | Owner draft |
| Evidence | Why should the evaluator believe it? | Current scoped source |
| Commitment | What future obligation is created? | Authorized decision |
| Review | Who confirms completeness and truth? | Named reviewer |
Bid decision
Use evaluation analysis to test attainability, not just write more
Assess the credible position for each criterion: strong current evidence, adequate evidence with adaptation, material gap with a recovery plan or unattainable condition. Add the response effort and delivery exposure. Test disclosed price-quality scenarios with ranges rather than one assumed competitor. The purpose is not to predict the award exactly. It is to determine whether a compliant and competitive path exists, which choices matter and where management must accept investment or risk.
Revisit the model after buyer clarifications and amendments. A changed weight, minimum score, required demonstration or price quantity can alter both pursuit and response architecture. During review, inspect the offer in the same criterion structure and verify mandatory conditions separately. Record residual gaps and authority. This guide is operational, not legal advice: the applicable jurisdiction, procurement documents and official communications control, and material legal interpretation should be handled by qualified counsel.
- Rate evidence strength and recovery plan by criterion.
- Combine attainable score, response effort and delivery exposure.
- Test disclosed price-quality mechanics across ranges.
- Reapprove pursuit when evaluation conditions materially change.
- Use qualified legal advice for jurisdiction-specific interpretation.
What good looks like
Useful outcomes from public tender evaluation criteria
- Exclusion, eligibility, minimum compliance and comparative scoring are visibly separated.
- All criteria, subcriteria, weights, scales and price rules are traced to source locations.
- The team can identify where failure means rejection and where quality changes relative score.
- Each scored element has a response location, owner, evidence and planned review.
- Price and quality scenarios use the disclosed method and explicit assumptions.
- High-weight criteria receive appropriate decision effort without starving mandatory items.
- Unknown evaluation details remain labelled rather than converted into false precision.
- Qualification considers attainable score, delivery exposure and effort together.
Operating model
How to run the work
- 01
Inventory the governing material
Collect the notice, instructions, specification, eligibility forms, award-criteria tables, price schedules, contract, annexes, questions and amendments. Record the current version and where the evaluation method appears. Resolve contradictions through document precedence or clarification.
- 02
Separate gates from comparison
Classify exclusions, eligibility, mandatory specifications, response-form rules and minimum thresholds separately from scored award criteria. Record the consequence of failure and required evidence. Do not assume a mandatory requirement earns points.
- 03
Reconstruct the disclosed scoring model
Capture criterion hierarchy, weights or order, scoring scale, descriptors, price or cost formula, normalization, minimum scores, tie-breaks, presentations and arithmetic. Test totals and scenarios without filling undisclosed gaps with invented rules.
- 04
Map criteria to response and proof
Break each criterion into the explicit elements an evaluator can observe. Assign a response owner, evidence source, format, space, decision and reviewer. Align named staff, methods, timelines and claims with the exact proof requested.
- 05
Use the model for pursuit and review
Estimate evidence strength and response effort by criterion, test price-quality scenarios and identify weak or unattainable areas. Revisit the pursue decision when a gate or score path changes. Review the final offer criterion by criterion in the buyer’s format.
Evaluation
Questions that change the decision
- Which document and version controls each evaluation rule?
- Is the item an exclusion, eligibility gate, mandatory specification, threshold or scored criterion?
- What exact consequence follows from failure or a low score?
- Which weights, scales and price calculations are explicitly disclosed?
- Which criterion elements require a claim, method, named resource, commitment or evidence?
- Where does the team have strong proof and where only an aspiration?
- Which ambiguity is material enough for a formal clarification?
- Does the attainable score and delivery position still justify the bid?
Failure modes
Where teams lose control
Selection criteria can be treated as award criteria and consume persuasive response space.
A minimum requirement can be missed because it sits outside the main criteria table.
Weights can add to a parent criterion while the team double-counts subcriteria.
A price formula can be approximated incorrectly and distort commercial scenarios.
A low-weight criterion can still contain a mandatory threshold.
The team can infer scoring descriptors that the authority did not disclose.
A response can address the heading but omit a required method, person or proof.
Claims about future delivery can be presented without an authorized commitment.
An amendment can change criteria or dates without updating the evaluation model.
Optimizing for stated score can obscure an unacceptable contractual or delivery exposure.
Measurement
Measure the finished job
Measure the completed workflow, including review effort and exceptions. Output volume on its own is not evidence of a better process.
- evaluation rules linked to current source locations
- gates with evidence, owner and pass condition
- scored criteria with response and proof mapping
- weights, scales and formulas that reconcile arithmetically
- undisclosed assumptions labelled and owned
- clarification questions raised before the buyer cutoff
- criteria with weak or unavailable supporting evidence
- planned response effort by disclosed criterion weight
- amendments propagated into the evaluation model
- final review findings by gate and scored criterion
Questions
Common questions
What is the difference between selection and award criteria?
Selection or eligibility criteria generally test whether the operator has the required capacity to participate. Award criteria compare admissible tenders. The applicable rules and tender documents define the exact treatment in each procedure.
Does the highest weighted criterion deserve the most proposal time?
It deserves strategic attention, but effort also depends on evidence maturity, complexity, dependencies and mandatory work. Allocate resources by evaluation value and the work required to produce an honest strong response.
Can a bidder calculate its likely tender score?
It can model disclosed formulas and use ranges for qualitative evidence, but it should not invent competitor inputs or undisclosed evaluator behavior. Keep calculable results separate from assumptions.
What should happen if evaluation documents conflict?
Record the exact sources and versions, check document-precedence rules and amendments, then raise a neutral clarification through the authorized channel when the conflict is material.
Sources
Primary references
- Directive 2014/24/EU on public procurement EUR-Lex
- Federal Act on Public Procurement Swiss Confederation
Zelius
Managed tender intelligence and bid execution for teams that want the commercial outcome.
Suppliers, founders and commercial teams pursuing public or private opportunities. Start with the workflow, constraints and evidence you already have.
See Zelius→