Product quiz field guide

# Product quiz no-match fallbacks: a decision tree

Handle empty product-quiz results safely with a six-level severity ladder, constraint-release rules, recovery actions and worked examples.

20 September 2026·Product Quiz Guide

Answer: treat no match as a valid, designed outcome. Identify why the eligible set became empty, preserve hard constraints, explain the conflict in plain language and offer only safe next steps. Never silently recommend an excluded, unavailable or incompatible item merely to fill the result page.

## What is a product quiz no-match result?

A no-match result occurs when no product, service or plan satisfies every active eligibility rule. The empty set can be legitimate (for example, no in-stock item meets a required dimension), or it can reveal missing product data, an unavailable variant, a market restriction or a logic defect. It is not the same as a low-scoring result: a low score can still be eligible, while a no-match result has failed at least one gate for every candidate.

Start with the hard exclusions in the [quiz logic model](https://bestproductquiz.com/blog/product-quiz-logic), then use the [catalog-readiness checklist](https://bestproductquiz.com/blog/product-finder-catalog-readiness) to determine whether the empty set is real. The result page should still meet the explanation and recovery requirements in the [results-page checklist](https://bestproductquiz.com/blog/product-quiz-results-page).

## Use the No-match severity ladder

Classify the failure before choosing copy or a call to action. The ladder moves from temporary catalog conditions to failures that require a human or technical intervention.

LevelLikely causeSafe responseNever do1. AvailabilityEligible item or variant is temporarily unavailableShow the eligible item as unavailable, offer an alert or another marketSubstitute an ineligible item2. Catalog dataA required attribute is missing, stale or inconsistentFlag the data gap, suppress the uncertain recommendation and escalate catalog QATreat a blank value as a confirmed match3. Soft preferenceEligible products exist only outside a preferred price, style or delivery windowName one tradeoff and let the visitor choose whether to recalculateRelax several preferences invisibly4. Hard constraintEvery candidate conflicts with compatibility, an explicit exclusion or a required capabilityExplain the conflicting requirement and offer a revision or human reviewOverride the constraint to produce a result5. Unsupported requestThe request is outside the catalog, service scope or safe recommendation boundaryState the boundary and provide a suitable specialist or contact routeImply expertise or availability that is not present6. Logic or system failureAn impossible path, broken rule, timeout or version mismatch occurredPreserve answers, show a stable error state and log a technical incidentPresent the failure as a buyer preference conflict

## Follow the Constraint release matrix

Record each input as a hard constraint, a visitor-controlled preference or an operational condition. Only the visitor can authorize a preference tradeoff. A later recovery step must not erase an earlier explicit exclusion.

Constraint classAuto-relax?Allowed recoveryExampleSafety, medical, legal or regulatory boundaryNoStop automated matching and route to qualified reviewA request that requires a medical suitability judgmentTechnical compatibility or required featureNoShow the conflict or let the visitor revise the requirementA software plan without required SSOExplicit exclusion or accessibility needNoKeep the exclusion and offer only conforming optionsA fragrance-free requirementMarket, inventory or delivery conditionNo, unless the visitor changes itOffer another market, restock alert or named later dateA sofa unavailable by the requested dateBudget, color, style or brand preferenceVisitor-controlled onlyOffer one named tradeoff before recalculatingShow eligible options above the preferred budgetRanking preferenceNot neededKeep all eligible items and explain the closest rankingPreferred finish affects order, not eligibility

## Run the fallback decision tree

- Reproduce the empty set. Save the quiz version, catalog version and controlled input fixture. If the same answers produce different results, treat it as a system failure.
- Check candidate data. Confirm product status, market, inventory, variants and every attribute used by an eligibility rule. Shopify currently separates product status, publishing, variants and available inventory, so a product record alone does not prove that a buyable match exists.
- Find the first emptying gate. Apply hard constraints in a documented order and record the first rule that reduces the eligible set to zero. Do not infer the cause from the final score.
- Classify the cause. Use the severity ladder to distinguish availability, data, preference, hard-constraint, unsupported and system failures.
- Choose a safe recovery. Keep hard constraints fixed. Offer a restock alert, partial result, visitor-controlled soft-preference change, question revision or human handoff as appropriate.
- Explain and measure. State what conflicted, what remains fixed and what each recovery changes. Log a reason code and the recovery action, not a bundle of raw personal answers.

## Worked skincare case

A routine finder requires fragrance-free products for a visitor who reports sensitivity. The only cleanser and moisturizer that satisfy both rules are out of stock in the visitor's market. The correct result keeps the sensitivity and fragrance exclusions, explains that no complete in-stock routine is available and offers an eligible partial routine, a restock notification or a later retry. It does not fill the routine with fragranced products, and it does not turn the result into a treatment claim.

## Worked furniture case

A sofa finder has an exact maximum width and a preferred delivery date. The catalog contains sofas within the width but none available by that date. Width remains fixed because the visitor described a physical space limit. The result can offer “keep this width and see later delivery” or “revise the maximum width,” with the changed condition stated before recalculation. It must not silently widen the allowed dimensions.

## Worked B2B case

A plan matcher requires SSO, a named integration and a 90-day implementation window. No self-serve plan satisfies all three. The fallback identifies the first unmet requirement, keeps the required capabilities visible and routes the buyer to an enterprise review with the requirements attached. It must not recommend the highest-scoring self-serve plan while hiding that SSO is absent.

## Make recovery accessible

The no-match state must be conveyed in text, not color alone. W3C notification guidance says feedback should be concise, understandable and include simple instructions for resolving the problem. Place a clear heading before the recovery choices, preserve the visitor's answers, link or return focus to the exact question that can be revised and announce dynamic result changes through an appropriate live region. Do not move keyboard focus unexpectedly when an inline message is sufficient.

## Measure reason codes and recovery actions

Use a small controlled vocabulary rather than raw answer values. Google Analytics distinguishes recommended events from custom events; use an existing recommended event when its definition fits and a documented custom event only when no recommended event describes the interaction.

FieldExamplePurposeQuiz and catalog versionquiz_12 / catalog_204Reproduce the evaluated rule and data stateReason codeavailability_emptyCount stable failure classes without exposing the answerFirst emptying gatemarket_inventoryLocate the rule that reduced the candidate set to zeroRecovery shownrestock_alertAudit whether the response matched the causeRecovery actionalert_requestedMeasure what the visitor chose next

Do not send health, financial, demographic or other sensitive answers as analytics parameters. Keep the detailed fixture in an access-controlled QA system when it is necessary to reproduce a defect.

## Release checklist

- Create a known no-match fixture for every hard constraint.
- Test missing attributes, unavailable variants, market restrictions and stale catalog data separately.
- Verify that every hard exclusion survives each recovery action.
- Confirm that each soft-preference change is named before recalculation.
- Test screen-reader announcement, focus order, keyboard operation and retained answers.
- Check reason codes and recovery events without sending sensitive raw answers.
- Repeat the fixtures after any rule, catalog, integration or product-status change.

## Limitations

This decision tree is an original design and QA framework, not a claim of universal match accuracy or conversion lift. It cannot decide which medical, legal, regulated or safety-relevant constraints are safe to relax; those boundaries require qualified domain review. Product status and inventory behavior vary by commerce platform and integration, so teams must verify the live data path they actually use. The worked cases are reproducible examples, not measured customer outcomes.

## Sources

- [Shopify, product details, variants, inventory, status and publishing](https://help.shopify.com/en/manual/products/details/product-details-page)
- [RevenueHunt, troubleshooting quiz results](https://docs.revenuehunt.com/how-to-guides/troubleshoot-quiz-results/)
- [W3C WAI, clear and accessible user notifications](https://www.w3.org/WAI/tutorials/forms/notifications/)
- [Google Analytics, recommended events](https://support.google.com/analytics/answer/9267735)

Method: the original No-match severity ladder, Constraint release matrix and decision tree were applied to the Northstar skincare, furniture and Atlas B2B briefs. Primary sources were rechecked on 20 September 2026. No live quiz experiment was run and no performance benchmark is claimed.

Corrections: send the page URL, the exact statement and a current source through the [contact form](https://bestproductquiz.com/contact).

[Compare the eight product recommendation quiz builders](https://bestproductquiz.com/#comparison) or read the [full testing protocol](https://bestproductquiz.com/how-we-test).

---

Canonical: https://bestproductquiz.com/blog/product-quiz-no-match
