---
title: "What if RFP questions and evaluation criteria do not align?"
description: "Build a two-way coverage map, preserve every explicit question and stop for clarification when a scored proposition has no permitted response location."
canonical: "https://zephior.com/insights/resolve-conflict-between-criteria-and-rfp-questions"
last-updated: 2026-09-04
---

# What if RFP questions and evaluation criteria do not align?

> Build a two-way coverage map, preserve every explicit question and stop for clarification when a scored proposition has no permitted response location.

By [Tony Kim](https://zephior.com/authors/tony-kim). Published 2026-09-04; updated 2026-09-04. 22 minute read.

## Definition

A criterion_question_coverage_map is a versioned, two-way record of how the buyer’s explicit response questions relate to its published evaluation criteria. It preserves each question contract, each criterion’s proof conditions, all permitted response locations and the relationship between them. The relationship may be direct, partial, one-to-many, many-to-one, an orphaned question, an orphaned criterion, an apparent tension, incompatible instructions or a missing route. The map keeps required but apparently unscored answers in scope and prevents scored material from being inserted into a prohibited or invented location. It does not decide the offer, rewrite the buyer’s method or treat a private assumption as clarification.

## Problem

Westmere Water Authority, a fictional buyer, is procuring an outage-reporting platform. Its response workbook asks for company details, an implementation plan, an operating model and an accessibility statement. The award schedule separately gives 25 per cent to continuity during migration, including rollback, transaction reconciliation and public notification. The implementation question asks for milestones and interfaces but never names those three proof points. A schedule annex is permitted, although its scope is not fully described. A further 10 per cent environmental criterion appears in the award table, yet no question, field or allowed attachment requests the evidence. The team is tempted to ignore the corporate question because it looks unscored and to hide environmental claims inside the implementation plan. Both shortcuts create an avoidable submission risk.

## Point of view

Read the question set and the evaluation model as two different structures that must be joined, not as substitutes for one another. Questions govern what the bidder must return and where. Criteria describe the propositions the buyer says it will assess. Preserve both at source level, decompose them only far enough to test coverage, and map them in both directions. A partial match calls for a precise response plan, not silent expansion of the question. A criterion with no proven route remains unresolved and may require a neutral clarification through the official channel. A mandatory question remains mandatory even when no score can be traced to it. Only an applicable amendment, clarification or precedence rule may change those conclusions.

## Keep the question tree and the evaluation tree separate

An RFP question is a delivery instruction. It may specify a prompt, a maximum length, a template cell, an attachment and the bidder entity that must answer. An evaluation criterion has another function: it identifies what the buyer will assess and may add a weight, threshold, descriptor or evidence expectation. The two often overlap, but their headings and boundaries need not match. Begin by preserving each structure as issued. Do not rename one until the original identifier, wording, version and location are recorded.

The distinction matters because compliance and score coverage can fail independently. A bidder can answer every visible question yet leave a scored proposition unsupported. It can also write persuasive material for every criterion while omitting a required declaration or putting it in the wrong file. UK guidance updated on 17 August 2026 says tenders are assessed against the published criteria and methodology, while procedural requirements in the tender notice or associated documents may still lead to a tender being disregarded. One checklist cannot stand in for both tests.

Freeze the comparison boundary before mapping. For Westmere, record the procedure identifier, lot, response stage, sole or consortium bidder, current workbook, instructions, award schedule, permitted annex list, amendments, clarification answers and the portal fields visible under the customer’s authorization. A criterion from an earlier notice and a question from the final workbook do not form a valid pair until the version chain proves that both still apply.

Use source identifiers that survive drafting. Q4.2 in the workbook, AC-2(b) in the award schedule and Annex S-1 in the submission rules are better anchors than “implementation section” or “continuity part.” Keep a human-readable locator and a machine-recoverable locator for each. The map can then survive a heading change, while reviewers can still find the issued text without trusting the normalized record.

**Two registers that must remain distinguishable**

| Register | Minimum source fields | Question it answers |
| --- | --- | --- |
| Question | ID, exact prompt, verb, subparts, limit, scope, location, version | What must the bidder return, where and in what form? |
| Criterion | ID, wording, subcriteria, descriptor, weight, threshold, version | What does the buyer say it will assess? |
| Response surface | File, field, section, annex rule, portal slot and cross-reference rule | Where may supporting material legally and procedurally appear? |
| Change source | Amendment, clarification, replacement object, effective scope and date | What altered or superseded the original relationship? |

## Compare question contracts with criterion proof conditions

Copying headings into a matrix produces false matches. Decompose each question into its answer contract: actor, requested act, object, conditions, time horizon, evidence, format and any explicit consequence. Q4 may ask the bidder to “provide the implementation plan, including milestones, interfaces and training.” Those are three requested components. It does not follow that the words “implementation plan” also request rollback criteria, reconciliation controls and public incident messages.

Now decompose the criterion without inventing hidden evaluator preferences. A proof condition is the smallest published proposition that could change the recorded assessment. Westmere’s continuity criterion names uninterrupted public reporting during transition, a tested rollback decision, reconciliation of queued records and notification during disruption. Preserve those as four conditions tied to the criterion source. If a descriptor distinguishes merely described from tested recovery, keep that distinction. Do not turn a broad policy objective into an unstated subcriterion.

Compare meaning, not keywords. A migration timeline may show when cutover occurs but say nothing about what happens when the cutover fails. A service desk response time may concern operation after go-live, not public notification during a migration incident. Conversely, one recovery rehearsal could support both continuity and implementation-risk propositions if the documents allow the same evidence to be referenced. The relationship must follow what the content proves in its stated scope.

EU rules illustrate why this discipline is useful. Under Article 56 of Directive 2014/24/EU, the authority verifies before award that the tender follows the applicable notice and procurement-document requirements. Article 67 requires criteria that allow the information supplied by bidders to be checked effectively. Neither provision tells a bidder which paragraph to use. Together they support a narrower control: preserve the disclosed assessment proposition and point to submitted evidence that can be examined against it.

**Coverage relationship vocabulary**

| Relationship | Evidence test | Response consequence |
| --- | --- | --- |
| Direct | One allowed answer establishes the whole criterion condition | Draft and review in that location |
| Partial | The answer covers some, but not all, published proof conditions | Add permitted coverage or keep a gap |
| One-to-many | One answer supports several criteria with the same bounded evidence | Reconcile claims and valid references |
| Many-to-one | Several answers jointly establish one criterion | Create a criterion-level synthesis path |
| Orphaned question | No published scoring link can be proven | Answer if required; do not discard |
| Orphaned criterion | No question or response route can be proven | Clarify or retain an unresolved stop |
| Apparent tension | Both instructions can be satisfied once function or scope is separated | Preserve both and document the distinction |
| Incompatible instructions | Controlling instructions require incompatible acts | Use authority rules or clarify |
| Route missing | Relevant proof exists but no permitted delivery location is established | Do not invent an attachment or field |

## Map forward from every question and backward from every criterion

A forward pass starts at each question. Record whether it is mandatory, conditional, optional or informative, then link it to relevant criteria only where the text supports the link. Give the required company-details question a response owner even if it is absent from the award table. It may support identification, eligibility or administration rather than merit scoring. The map should say “no award link identified” rather than “worthless” or “safe to omit.”

A backward pass starts with every criterion and subcriterion. For each proof condition, find the response location where an evaluator can encounter the support. A criterion is not covered because its noun appears somewhere in the bid. Record the exact answer, paragraph, table row, annex or portal field, and show how that material establishes the condition. If several answers are needed, name their order and the permitted cross-reference that joins them.

Run both passes before assigning writing work. A forward-only matrix rewards completion of the workbook and misses orphaned criteria. A backward-only score plan encourages writers to neglect unscored instructions. The difference is visible in Westmere: Q1 remains a required response in the forward pass, while the environmental criterion remains unresolved in the backward pass. Neither can be erased by prioritizing the other view.

Keep negative evidence. If the annex rules list only pricing, security and accessibility attachments, that list may be evidence against a proposed environmental annex. If the portal displays a generic “other document” slot, record it as an observed interface element until the procurement instructions establish its permitted purpose. Absence is harder to prove than presence, so cite the source boundary and checked time instead of asserting that no route can ever exist.

## Give scored evidence a permitted evaluator path

A partial mapping does not automatically authorize a new section. Start with the narrowest existing question whose contract can lawfully hold the missing proof. Westmere Q4 asks for an implementation plan and permits a schedule annex. Rollback decision points and reconciliation during cutover are implementation controls, so they can be planned under Q4 if the stated limit and annex rule accommodate them. Public notification may instead belong in the operating-model answer. The map should preserve that split rather than forcing all continuity evidence under one heading.

Cross-references must do more than save words. They need a permitted convention, a stable destination and enough context for the evaluator to understand the claim. “See elsewhere” is not a coverage route. A useful record names the source answer, target criterion, submitted file, final section, evidence proposition and fallback if cross-references are prohibited. When the same fact appears twice, choose one controlled statement and reconcile any necessary restatement before release.

Keep answer limits attached to the route. Eight pages for implementation do not become ten because the criterion has four subparts. A permitted annex may be excluded from the page count yet restricted to schedules rather than narrative. A portal field can truncate text or reject a file type even when the document rules allow the content. These are distinct constraints. If they cannot all be met, the problem belongs in the clarification or response-architecture queue, not in a writer’s private workaround.

FAR 15.204 is a useful US federal example of solicitation sections separating instructions from evaluation factors, while FAR 15.305 ties evaluation to the factors and subfactors in the solicitation. That familiar separation should not be projected onto every procurement. Some buyers combine questions, instructions and criteria in one workbook. The map follows the actual issued objects, not a preferred procurement format.

## Do not call every mismatch a contradiction

Questions and criteria often point in different directions without requiring incompatible acts. A question may request a chronological implementation plan while the criterion groups evidence by resilience, accessibility and staffing. The bidder can satisfy both by keeping the requested chronology and using permitted headings, tables or references to make the assessed propositions findable. This is an apparent tension between structures, not a conflict in what must be submitted.

A material conflict requires two applicable propositions that cannot both be satisfied. One instruction might prohibit any attachments while another requires a signed form to be attached. A response question might cap an answer at 500 words while a controlling schedule mandates a full table that cannot fit inside the field and forbids references. Preserve the two source statements, scope and consequence before applying any precedence rule. Similar discomfort or a difficult page budget is not enough.

Authority comes from the procurement, not from a general preference for the award table. An explicit document-order clause may settle a conflict, but only for the stated documents and issue. A later amendment may replace one instruction if its scope and effect are clear. An authoritative clarification may explain the route. Publication order, filename date, spreadsheet colour, a portal label or the internal belief that scored content “must win” cannot silently displace a question.

The Court of Justice decisions in Lianakis and TNS Dimarso address limits around undisclosed weightings, subcriteria and evaluation methods. They are not bidder-side precedence rules. Use them to understand why the disclosed evaluation basis matters, then resolve the live document issue through the procurement’s own sources and applicable legal review. The map records the operational decision; it does not make a legal determination.

**Decision test for a suspected mismatch**

| Test | If yes | If no |
| --- | --- | --- |
| Same procedure, lot, stage and version? | Continue | Separate the scopes before comparing |
| Same required act or proposition? | Test compatibility | Record a relationship, not a conflict |
| Can both instructions be satisfied as written? | Classify apparent tension | Preserve a conflict candidate |
| Does an applicable precedence rule decide it? | Record the rule and result | Check amendment and clarification authority |
| Does an authoritative change settle the issue? | Supersede with lineage | Keep unresolved and consider clarification |
| Would a private workaround change the delivered object? | Stop | Proceed only within the proven response route |

## Answer required questions even when no score is visible

No visible scoring link does not mean no consequence. Company details may establish bidder identity. A declaration may confirm acceptance, conflicts or exclusion information. A data table may feed a contract schedule. A question may be evaluated under a broader criterion that does not repeat its label. Classify the item from the issuing text and its stated consequence. If it is mandatory, plan a compliant answer unless an authoritative source removes it.

Give unscored work proportionate attention without pretending it earns points. Westmere Q1 asks for a 500-word company profile. The bidder should supply the requested legal identity, relevant delivery role and bounded organizational facts, then stop. It should not consume the full allowance with marketing claims. The scoring map can mark “award link not identified,” while the compliance register preserves the mandatory response and final location.

Separate conditions of participation from award criteria where the procurement does so. Current UK guidance states that participation conditions concern supplier capability and are assessed separately from the tender’s merits. The same piece of experience should not drift from an eligibility response into merit scoring unless the issued criteria request it there. This prevents both double counting and the mistaken deletion of a selection-stage answer.

If the purpose remains uncertain, keep the uncertainty narrow. “No criterion link found within the reviewed award schedule and clarifications as of the checked time” is supportable. “The buyer will ignore this answer” is not. The map is a planning and review instrument. It cannot know an evaluator’s private reading or convert silence into permission to omit a required return.

## Ask where to respond when the procurement never provides a route

An orphaned criterion becomes material when its published proof conditions could affect assessment and the reviewed response architecture provides no supported location. Confirm the source boundary first. Check the instructions, templates, listed annexes, amendments, clarification log and permitted portal view. Record inaccessible or missing objects rather than treating them as reviewed. Then test whether an existing free-form field expressly permits the content. A generic upload control does not answer that question.

Write the clarification around the missing procedural fact. For Westmere: “The award schedule lists Environmental Performance at 10 per cent. We have not identified a corresponding response field or permitted attachment in the current workbook and submission schedule. Please confirm the required response location and any applicable limit.” The question names the source, the observed gap and the decision needed. It does not reveal the bidder’s environmental method, evidence or intended score strategy.

Timing and authority still matter. Use the official clarification channel, observe its cutoff and obtain customer approval before external contact. If the buyer answers publicly, bind the response to its audience, date and affected documents. If the answer creates a new field or changes a limit, update the response architecture and every dependent map row. If the buyer does not answer, preserve the unresolved state and route the risk to the authorized bid and legal decision rather than improvising a submission object.

FAR 15.206 shows one formal amendment model in which changes to a solicitation are issued to affected parties. World Bank procurement regulations likewise require evaluation criteria and methodology in the request document. Those sources help frame a control, but the live procedure determines which communication changes Westmere’s instructions. An email, webinar comment or help-desk reply has no automatic authority outside the rules that govern it.

**Westmere criterion_question_coverage_map excerpt**

| Map row | Question route | Criterion proof condition | State and next action |
| --- | --- | --- | --- |
| W-01 | Q1 company details | No award link identified | Required orphaned question; answer concisely |
| W-02 | Q4 plan plus permitted schedule annex | Migration milestones and interface control | Direct; draft in Q4 |
| W-03 | Q4 plan plus schedule annex | Rollback decision and record reconciliation | Partial; add bounded controls and evidence |
| W-04 | Q6 operating model with reference to Q4 | Public notification during migration disruption | Many-to-one; verify cross-reference permission |
| W-05 | No proven field or listed annex | Environmental Performance, 10 per cent | Route missing; submit neutral clarification |
| W-06 | Accessibility statement and Q4 training plan | Accessible adoption during transition | Many-to-one; reconcile roles and dates |

## Release only when both directions reach the submitted evidence

Run the forward release review against the exact submission candidate. Every required or triggered question needs a completed response in its permitted destination, within its limit, using the current template. Conditional questions need their trigger decision. Optional questions need an explicit inclusion choice. An orphaned question can pass this review because its compliance treatment is known even when no scoring relationship is proven.

Then run the backward review. Start at every criterion and follow the map to the words and evidence the evaluator will receive. Confirm that the destination exists in the export, the reference resolves, the proof retains its approved scope and the relevant claim did not change during editing. A criterion with a route-missing or conflict state cannot be marked covered merely because the internal team has evidence on file.

Use bounded release states: complete for response planning, complete with review items, source boundary incomplete, route unresolved, instruction conflict, human review required or superseded. “Complete” means every declared source was reviewed and every material relationship has a supported disposition. It does not predict the score, certify legal compliance or promise that the buyer will accept a disputed route.

Bind the map to source and response versions. Reopen affected rows after an amendment, authoritative clarification, replacement template, criterion refinement, lot change, bidder-structure change or material portal change. An authorized tool may extract candidate questions, compare phrases, propose relationships and test references. People still confirm meaning, source authority, evidence scope, buyer contact and release. The final work product remains inspectable: a reviewer can see why each answer exists and where each scored proposition is supported.

## Useful outcomes

- Every explicit question has a response disposition, permitted location and accountable owner, including questions that appear unscored.
- Every published criterion and material subcriterion has at least one traced proof route or an explicit unresolved state.
- Direct, partial, one-to-many and many-to-one relationships are recorded without forcing a false one-question-one-criterion model.
- Unrequested scored material is placed only where the procurement permits it, with valid cross-references when allowed.
- Apparent tensions are separated from material contradictions in instructions, scope or permitted response location.
- Missing routes trigger a bounded clarification decision rather than an invented annex, portal field or exception.
- Changes to questions, criteria, templates or response rules reopen only the affected relationships and answers.
- The final response can be reviewed from question to criterion and from criterion back to submitted proof.

## Workflow

1. **Freeze the applicable package.** Identify the procedure, lot, stage, bidder configuration, current documents, amendments, clarifications, templates and permitted portal view.
2. **Build the question register.** Record every explicit prompt, declaration and requested return with its verb, subparts, limit, location, scope and consequence.
3. **Build the criterion register.** Preserve criteria, subcriteria, descriptors, weights, thresholds and stated assessment evidence with clause-level locations.
4. **Map in both directions.** Link each question to every relevant criterion, then test each criterion against all questions and allowed response surfaces.
5. **Classify each relationship.** Use direct, partial, one-to-many, many-to-one, orphaned, tension, conflict or route-missing states with supporting evidence.
6. **Design permitted coverage.** Assign proof to the narrowest allowed question, schedule, field or attachment and record any valid cross-reference convention.
7. **Clarify unresolved routes.** Ask the minimum neutral question when a material criterion has no proven response route or two controlling instructions cannot coexist.
8. **Reconcile the submission candidate.** Review forward and backward against the exact export, retain unresolved stops and reopen affected rows after a controlling change.

## Key decisions

- Which procedure, lot, stage, bidder configuration and source versions define the comparison?
- Is the item a response question, declaration, condition of participation, requirement, criterion, descriptor or delivery instruction?
- What exact proposition must the answer establish, and what evidence would satisfy the published criterion?
- Does one question cover the criterion fully, partly or only by an unsupported thematic resemblance?
- Does the criterion depend on several answers, or does one answer support several criteria?
- Where may the bidder place the material, and are cross-references or supplementary annexes expressly permitted?
- Is an apparently unscored question still mandatory, conditional, optional or merely informative?
- Do the documents create a true contradiction, or do they describe different functions that can both be satisfied?
- Can an applicable amendment, clarification or stated precedence rule resolve the issue without bidder invention?
- What change, missing source or approval reopens the mapping before release?

## Risks

- Assuming that the response workbook contains every proposition the evaluators may assess
- Deleting a mandatory corporate, declaration or compliance answer because no numerical score is visible
- Filling an orphaned criterion with an unauthorized appendix or an unrelated portal upload slot
- Treating shared vocabulary as proof that a question covers every subcriterion
- Writing the same scoring claim into several answers without reconciling scope, numbers or commitments
- Letting an internal cross-reference point to a working document rather than the submitted location
- Using a later filename, convenient reading or private convention as if it settled document authority
- Asking a strategic or leading clarification that reveals the planned solution rather than the missing route
- Closing the map while an inaccessible template, portal field or bidder message remains outside the checked boundary
- Failing to reopen answers after a criterion, question, attachment rule or clarification changes

## Metrics

- Explicit questions with a recorded response disposition and permitted location
- Criteria and subcriteria with a supported response route or named unresolved status
- Partial relationships with an approved coverage plan and owner
- Orphaned questions retained and classified rather than dropped
- Orphaned criteria sent to clarification or response-architecture review
- Cross-references tested against the final submitted identifiers and files
- Conflicts resolved only through documented buyer authority
- Unmapped submitted claims found in backward review
- Changes that reopened mapping rows and dependent answers

## Frequently asked questions

### Should every RFP question map to a scored criterion?

No. Questions may collect declarations, identity data, eligibility information, contractual inputs or administrative returns. Preserve and answer a required question even when no award link can be proven.

### Should every evaluation criterion have its own question?

Not necessarily. One question may support several criteria, and several answers may jointly support one criterion. Each criterion still needs a permitted evaluator path to the submitted proof.

### Can we add an appendix for a criterion with no response field?

Only when the submission rules permit that appendix and its purpose. A generic upload slot or an internal need for more space is not enough. Clarify a material missing route instead of inventing one.

### Is a partial question-to-criterion match a conflict?

Usually not. It shows that the question covers only part of the published proof condition. Map the remaining condition to another permitted location or retain it as an unresolved gap.

### Which takes priority, the questions or the scoring table?

Neither has automatic priority for every issue. Questions govern required responses; criteria govern the disclosed assessment basis. Use the procurement’s applicable precedence, amendment and clarification rules for a true contradiction.

### How should cross-references be used?

Use them only where permitted, with a stable submitted destination and enough context to show what the target proves. Verify every reference against the final export.

### What if the clarification deadline has passed?

Preserve the missing route and follow the procedure’s available communication and internal decision controls. Do not present a late private assumption as buyer authority or create an unauthorized response object.

### What can software do safely in this mapping?

It can inventory supplied sources, extract candidate units, suggest relationships, find orphan candidates and test final references. Human reviewers decide meaning, authority, evidence use, clarification and release.


## Primary sources

- [Guidance on assessing competitive tenders under the Procurement Act 2023](https://www.gov.uk/government/publications/procurement-act-2023-guidance-documents-procure-phase/assessing-competitive-tenders-html), Government Commercial Function
- [Official Procurement Act learning module on assessment and award](https://www.gov.uk/government/publications/the-official-procurement-act-2023-e-learning/module-7-assessment-and-award-of-contracts), Government Commercial Function
- [Guidance on assessment summaries and criterion reasons](https://www.gov.uk/government/publications/procurement-act-2023-guidance-documents-procure-phase/guidance-assessment-summaries-html), Government Commercial Function
- [Procurement Act 2023 section 23 on award criteria and assessment methodology](https://www.legislation.gov.uk/ukpga/2023/54/section/23), The National Archives
- [Procurement Act 2023 section 24 on refining award criteria](https://www.legislation.gov.uk/ukpga/2023/54/section/24), The National Archives
- [Directive 2014/24/EU, consolidated English text](https://eur-lex.europa.eu/eli/dir/2014/24/2026-01-01/eng), EUR-Lex
- [TED eForms business terms for award criteria](https://docs.ted.europa.eu/eforms/latest/reference/business-terms/index.html), Publications Office of the European Union
- [TED eForms schema usage for complex criteria](https://docs.ted.europa.eu/eforms/latest/schema/all-in-one.html), Publications Office of the European Union
- [TNS Dimarso judgment, Case C-6/15](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A62015CJ0006), Court of Justice of the European Union
- [Lianakis judgment, Case C-532/06](https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62006CJ0532), Court of Justice of the European Union
- [Antea Polska judgment, Case C-54/21](https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62021CJ0054_RES), Court of Justice of the European Union
- [Partner Apelski Dariusz judgment, Case C-387/14](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A62014CJ0387), Court of Justice of the European Union
- [FAR 15.203 on requests for proposals](https://www.acquisition.gov/far/15.203), Acquisition.gov
- [FAR 15.204 on the uniform contract format](https://www.acquisition.gov/far/15.204), Acquisition.gov
- [FAR 15.304 on evaluation factors and relative importance](https://www.acquisition.gov/far/15.304), Acquisition.gov
- [FAR 15.305 on proposal evaluation](https://www.acquisition.gov/far/15.305), Acquisition.gov
- [FAR 15.206 on solicitation amendments](https://www.acquisition.gov/far/15.206), Acquisition.gov
- [World Bank Procurement Regulations, seventh edition](https://thedocs.worldbank.org/en/doc/c84273d1b230aeb2b0b8134de5dc8cd7-0290012025/original/Procurement-Regulations-7th-Edition-Sep-2025.pdf), World Bank
- [UNCITRAL Model Law on Public Procurement](https://uncitral.un.org/sites/default/files/media-documents/uncitral/en/2011-model-law-on-public-procurement-e.pdf), United Nations Commission on International Trade Law
- [ADB guidance note on merit point criteria](https://www.adb.org/documents/merit-point-criteria-procurement-guidance-note), Asian Development Bank


## Related articles

- [Turn tender evaluation criteria into a response strategy](https://zephior.com/insights/turn-evaluation-criteria-into-response-strategy)
- [What to do when a buyer does not answer a tender question](https://zephior.com/insights/proceed-when-the-buyer-does-not-answer)
- [How do you map privacy records to an RFP question?](https://zephior.com/insights/map-privacy-controls-to-rfp-questions)
- [Which RFP requirements apply only in certain cases?](https://zephior.com/insights/map-conditional-rfp-requirements)