An evaluation-criteria clarification asks the buyer to make a published assessment basis usable where the tender leaves a material uncertainty about what is evaluated, the relative importance, required evidence, score boundary or aggregation. It points to the exact source and asks for neutral meaning or treatment. It does not request inside guidance on how the bidder should win, propose a favorable scoring rule or ask the buyer to pre-evaluate the bidder’s planned response.
Teams often say a criterion is vague when they have not separated the actual ambiguity. One person questions the weight, another the evidence and a third the meaning of “innovation.” Their combined question becomes long, strategic and easy to answer partially. Other teams ask, “Will full marks be awarded if we propose X?” That reveals the solution and tries to turn clarification into scoring advice. If the buyer replies only to the surface wording, the bid still lacks a controlled basis for effort, evidence and response design.
Diagnose before asking. Trace criterion, subcriterion, question, instruction, score descriptor and price treatment, then name one decision the documents do not permit. Classify it as scope, relative importance, evidence, scoring boundary, calculation or process. Draft a short neutral question that quotes or cites the conflict and requests the missing rule for all bidders. Test that the answer would change a material bid decision and that the question does not disclose the proposed solution.
Diagnosis
Locate the ambiguity in the evaluation chain
Build a compact chain for the affected item: response question, required content, evaluation factor, significant subfactor, relative importance, score descriptor, formula and award method. Include definitions and amendments. Many apparent ambiguities disappear when a term in the specification or a formula note is connected to the criterion. Others become sharper: the question requests transition evidence while the stated criterion evaluates only steady-state service, or the weights sum to a different total in two schedules.
Classify what remains. Scope ambiguity concerns which content or entity is assessed. Evidence ambiguity concerns the record needed to substantiate performance. Scoring ambiguity concerns the boundary between ratings or whether a threshold is pass-fail. Weighting ambiguity concerns relative importance. Aggregation ambiguity concerns normalization, rounding or combining price and quality. Process ambiguity concerns presentations, demonstrations, moderation or stages. One issue can touch several classes, but the question should identify them separately.
Distinguish missing disclosure from desired detail. Some regimes require factors and relative importance without requiring every internal rating technique to be published. FAR 15.304, for example, says factors, significant subfactors and relative importance must be stated clearly, while the rating method need not be disclosed. A bidder should ask for the information needed to prepare an assessable response, not assume it is entitled to every evaluator worksheet. Applicable rules and tender terms control the boundary.
| Class | Unresolved question | Possible bid effect |
|---|---|---|
| Scope | Which content, service, entity or period is assessed? | Solution and response coverage |
| Evidence | What proof is required or permitted? | Evidence collection and page use |
| Score boundary | What distinguishes threshold, acceptable and higher quality? | Response depth and review target |
| Weighting | What is the relative importance? | Effort and page allocation |
| Aggregation | How are scores normalized, rounded or combined? | Commercial model and scenario analysis |
| Process | Which stage or artifact is evaluated? | Team, demo and submission planning |
Question design
Ask for an assessable rule, not help writing the answer
Use a four-part structure: source, observed gap, decision that cannot be made and neutral request. Example: “Schedule 2 assigns 15 percent to implementation, while Appendix B assigns 10 percent. Please confirm the weighting that will be used in evaluation.” For a qualitative term: “Criterion 4 evaluates ‘local capability,’ but the tender does not define whether subcontractor personnel are included. Please confirm the entities and personnel included in that assessment.” Both can be answered for every bidder.
Quote only enough text to locate the issue. Do not attach an essay about why your interpretation is better. Avoid double questions such as “Please confirm the weight and whether three case studies will receive full marks.” Split weight, evidence quantity and score treatment because the buyer may have different authoritative sources for each. Use defined terms from the tender and ask for confirmation, definition, precedence or correction rather than inventing new vocabulary.
Test the requested output. The answer should be capable of becoming one controlled statement: “subcontractor personnel are included,” “the applicable weight is 15 percent,” or “the criterion is pass-fail.” If the likely answer is merely “please follow the tender,” the question may still be too broad. If the requested output is a critique of your solution, it is not a clarification. Redraft until the buyer can resolve the published basis without seeing the planned response.
- Cite the exact schedule, criterion and version.
- Describe one conflict or missing boundary.
- State the assessment decision that cannot be made.
- Request a rule that applies equally to all bidders.
- Keep the proposed solution and target score out of the question.
Disclosure control
Avoid scoring hypotheticals and strategic leakage
Do not ask, “Will we receive maximum points if…” The buyer cannot fairly pre-score an unseen response, and the question may reveal the exact feature, partner, evidence gap or pricing concern that shapes your strategy. Ask what characteristic is assessed, not whether your planned characteristic wins. Replace “Will our AI-based approach score highly?” with “Please clarify whether the innovation subcriterion assesses novelty, measurable outcome, implementation feasibility, or a combination of these.”
Remove facts that are unnecessary to resolve the rule. If the uncertainty is whether consortium experience counts, there is no need to name your consortium member or admit that the prime lacks experience. If the question concerns treatment of zero values in a formula, do not disclose your expected price. Preserve confidential strategy while giving the buyer enough context to identify the published defect. Assume that the question and answer may be shared with all bidders.
Do not use clarification to lobby for a new criterion, lower threshold or preferred technical architecture. A legitimate question may expose a contradiction or an assessment basis that cannot be applied consistently. State that issue neutrally and leave the correction to the buyer. Public procurement principles and FAR evaluation rules both emphasize assessment against stated criteria. The bidder’s job is to request clarity, not to write the authority’s source-selection method.
| Weak pattern | Problem | Controlled request |
|---|---|---|
| Will X receive full marks? | Requests pre-scoring and reveals strategy | What characteristics distinguish the published score levels? |
| What do you mean by quality? | No source or decision boundary | Does criterion 3 assess process, output, or both? |
| Can you reduce this threshold? | Seeks a favorable requirement change | Which of the conflicting thresholds controls? |
| Please explain the whole formula | May be too broad to resolve | How are tied or rounded values treated in cell C7? |
| Our partner has X; will it count? | Discloses bid facts unnecessarily | Is subcontractor evidence eligible under criterion 5? |
What good looks like
Useful outcomes from clarify ambiguous tender evaluation criteria
- Every clarification addresses one identifiable gap in the published evaluation basis.
- Buyer answers can be applied consistently to response structure, evidence and effort.
- Questions do not ask the buyer to endorse a planned solution or award a hypothetical score.
- Weighting and score calculations are distinguished from qualitative meaning.
- The team preserves the authoritative wording and records exactly what changed.
- Unanswered ambiguities receive a controlled bid treatment rather than private interpretations.
Operating model
How to run the work
- 01
Reconstruct the published evaluation chain
Link the response question to criterion, subcriterion, weight, descriptor, instructions, price treatment, formula and relevant clarification.
- 02
Name one unresolved assessment decision
State what a reasonable bidder cannot determine and classify the uncertainty by scope, evidence, scoring, weighting, aggregation or process.
- 03
Test materiality and neutrality
Confirm that an answer changes a real bid decision and that the question neither reveals the solution nor asks for preferential coaching.
- 04
Draft the minimum sufficient question
Cite the source, describe the inconsistency or missing boundary and request the authoritative treatment in terms applicable to all bidders.
- 05
Apply the answer across the bid
Update the controlled criteria map, response plan, evidence standard, score model, effort and approvals without expanding the answer beyond what the buyer stated.
Evaluation
Questions that change the decision
- Which exact criterion, descriptor, instruction or formula creates the ambiguity?
- Is the uncertainty about what is assessed or how assessment results are combined?
- Can the documents be reconciled through a defined term or explicit precedence rule?
- Would the answer change solution design, evidence, page allocation, price or pursuit appetite?
- Does the draft question expose a differentiator or ask the buyer to score a hypothetical response?
- Can one question be split into independent decisions that the buyer can answer precisely?
- What is the final date and permitted channel for evaluation clarifications?
- Who approves the bid basis if the buyer does not answer?
Failure modes
Where teams lose control
A broad question may receive a broad answer while the material uncertainty remains.
The bidder may reveal a planned differentiator or concern to every competitor.
A question may coach the buyer toward a scoring method favorable to one solution.
The team may ask for information already defined elsewhere and lose credibility.
Weighting, threshold and score aggregation may be mixed into one calculation error.
An answer may be treated as changing more than its exact wording supports.
Different authors may apply the clarification to only some affected sections.
The team may wait for a reply beyond the point where a bid basis must be approved.
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 questions tied to one source and one decision gap
- clarifications answered without follow-up interpretation
- questions removed because the documents already resolve them
- material bid decisions changed by an answer
- criteria-map and response-plan updates completed after clarification
- unanswered evaluation ambiguities with approved treatment
- conflicting applications of one buyer answer found in review
Questions
Common questions
Can a bidder ask the buyer what is needed for full marks?
Ask for clarification of published score descriptors or missing boundaries, not for a pre-score of your intended response. Phrase the request so the answer defines the rule equally for all bidders.
Should every vague term become a clarification question?
No. First check definitions, specifications, criteria, amendments and ordinary context. Ask only where a material assessment decision remains unresolved and the answer would change bid treatment.
May the buyer keep the detailed rating method confidential?
Rules differ. Some require clear factors and relative importance without disclosure of every internal rating technique. Seek qualified advice on the applicable regime and request only what bidders need for a fair, assessable response.
What if two evaluation schedules conflict?
Cite both locations and versions, state the precise difference and ask which controls or whether an amendment will follow. Do not choose the more favorable schedule without an approved bid basis.
Sources
Primary references
- FAR 15.304 Evaluation Factors and Significant Subfactors Acquisition.gov
- FAR 15.305 Proposal Evaluation Acquisition.gov
- Directive 2014/24/EU on public procurement EUR-Lex
About the author
Tony Kim
Tony writes about applied AI, dependable product engineering and the systems that turn complex response work into controlled delivery.
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.
