How this guide was built
This guide combines a dated review of Pew Research Center questionnaire guidance, W3C Web Accessibility Initiative multi-page-form guidance, and U.S. Web Design System complex-form patterns with an original fifteen-question inventory, purpose-and-action matrix, worked B2B fixture, and twelve question regression tests. It is documentation-led guidance, not a usability study, authenticated conversion test, universal discovery script, or sales-methodology certification.
What is a lead qualification question?
A lead qualification question collects one piece of evidence needed to choose a useful next step. Its answer may change the path, score, result, owner, workflow, preparation content, or whether a manual review is required. The question is not justified merely because the answer might be interesting later.
A useful question has a stable answer ID, a declared purpose, an optional-or-required rule, a blank-state policy, and at least one downstream action that can be tested. The visible wording may improve over time, but the stored identity and action contract should stay governed.
Define the decision before writing the question
Write the allowed routes first: book now, send to a specialist, request manual review, provide a self-service result, or nurture until a declared condition changes. Then list the minimum evidence needed to choose among those routes. If two different answers produce the same experience, record, and action, the question may belong in later discovery instead of the funnel.
Pew Research Center advises careful question wording, pretesting, attention to question order, and awareness that answer-option order can influence responses. W3C recommends splitting long forms into logical stages, indicating progress, and making optional stages recognizable and skippable. Those principles improve interpretability and accessibility; neither source promises a conversion lift.
Sources: Pew Research Center: Writing survey questions, W3C Web Accessibility Initiative: Multi-page forms
Use this fifteen-question library
Select the smallest set that can change a declared decision. The examples below are prompts to adapt, not a recommendation to ask all fifteen questions in one funnel.
| Evidence | Example question | Ask when | Action it may change |
|---|---|---|---|
| Goal | What outcome are you trying to achieve? | Everyone | Path and result |
| Current state | How do you handle this today? | Everyone | Gap explanation |
| Trigger | What changed recently? | Timing matters | Urgency evidence |
| Consequence | What happens if this stays the same? | Complex purchases | Priority, not fear copy |
| Scope | Which team or process is affected? | Multi-team offers | Specialist route |
| Volume | About how many cases occur monthly? | Capacity-sensitive offers | Package or review |
| Environment | Which relevant system is already in place? | Integration-dependent offers | Feasibility path |
| Constraint | Which limit matters most? | Tradeoffs differ | Recommendation |
| Timeline | When must a workable solution be live? | Delivery has lead time | Timing route |
| Budget range | Which approved range is realistic? | It changes delivery | Commercial route |
| Authority | Who needs to approve the next step? | Multi-stakeholder sale | Preparation content |
| Owner | Who would own implementation? | Execution risk matters | Readiness score |
| Required capability | Which requirement cannot be compromised? | Options differ | Product or service fit |
| Contact | Where should the result be sent? | Value is explained | Identification |
| Consent | May we send the stated follow-up? | Before marketing follow-up | Workflow permission |
What should a qualification funnel skip?
Skip sensitive, speculative, or invasive fields that do not control a declared decision. Do not ask visitors to diagnose themselves using internal jargon. Avoid combining two facts in one question, presenting overlapping answer options, forcing an answer when 'not sure' is valid, or asking for an exact budget when a range is sufficient.
Do not treat an unanswered optional field as negative intent. Do not infer protected or sensitive traits from proxy answers. If the information is not needed for the promised result or a clearly explained next step, remove it or defer it to a qualified human conversation.
Sequence from simple context to justified sensitivity
The U.S. Web Design System recommends progressive disclosure, moving from simpler to more difficult questions, breaking complex forms into meaningful chunks, explaining why information is collected, and helping users recover from errors. Use those principles to place low-effort context first and potentially sensitive constraints only after their relevance is visible.
Show progress, preserve corrections, and disclose optional steps. When earlier answers hide later questions, confirm that back navigation recalculates the path without silently discarding information the visitor still needs to review.
Sources: U.S. Web Design System: Complete a complex form, U.S. Web Design System: Progress easily
Keep fit, readiness, and route separate
Fit asks whether the use case matches the offer. Readiness asks whether the buyer can act. Route asks which next step is useful now. A strong fit can be early, a ready buyer can be wrong for the service, and a low numerical score can still deserve a useful self-service result.
Store the three layers separately before combining them into a visible result. This makes overrides explainable and prevents one convenient total score from hiding whether the actual constraint is product fit, timing, ownership, permission, or missing evidence.
| Layer | Question answered | Example action |
|---|---|---|
| Fit | Does the need match what can be delivered? | Match, specialist review, or referral |
| Readiness | Can the buyer act under current constraints? | Book, prepare, or nurture |
| Route | Which next step is useful and permitted now? | Owner, workflow, resource, or manual review |
Worked example: an implementation consultancy
A hypothetical consultancy offers guided setup, a partner referral, and a readiness plan. Its funnel asks goal, current process, team scope, deadline, existing CRM, implementation owner, and contact details. A compatible system plus a named owner and near-term deadline routes to guided setup. A compatible system without an owner returns a two-step readiness plan. An incompatible requirement routes to specialist review instead of a false rejection.
The same stable answer IDs, route ID, and rules version appear in the result, contact record, and follow-up. This fixture is illustrative; it is not evidence that this question set improves conversion or lead quality.
Run twelve question regression tests
Test fixed fixtures before release and after any wording, option, branching, scoring, consent, or routing change. Keep expected and actual path, result, stored answer IDs, and actions together in the evidence record.
| Area | Tests |
|---|---|
| Purpose | 1. Every question has one documented purpose; 2. each answer changes a declared state or is removed; 3. options are intelligible and include a truthful fallback |
| Path | 4. Required and optional status is visible; 5. skip logic never hides evidence needed by the result; 6. back navigation and corrections recalculate |
| Decision | 7. Boundary answers select the declared band; 8. contradictions reach a safe review path; 9. contact gating preserves the promised result |
| Action | 10. Consent controls only the actions it authorizes; 11. stored answer IDs, result, and action agree; 12. keyboard, focus, errors, and progress are understandable |
What are the limitations?
This library is not a universal discovery script, legal basis for data collection, predictive model, conversion benchmark, or substitute for seller judgment. Regulated or consequential decisions require appropriate governance, qualified review, and testing for unfair or harmful outcomes.
Question behavior, source guidance, and downstream product configuration can change. Pretest with the intended audience, validate every rule and stored field in the actual implementation, and submit corrections through the site contact page.
The decision in one paragraph
Define the possible routes first, then ask for the minimum evidence that can change one of them. Keep fit, readiness, and route separate; place sensitive questions only where their purpose is clear; store stable answer identities; and prove with fixed fixtures that corrections, consent, contradictions, and optional paths lead to the declared result and action.