Post-webinar survey questions: a six-question starting core

What this six-question core is for

The router is additional to the six questions below. The set is a practical starting instrument, not a scientifically validated universal scale. Product demonstrations, public briefings and accredited courses may require different measures. Add an item only when a named person owns the decision it informs. CDC: building an evaluation plan

Decide what each answer will change

Before opening a form builder, write down the action attached to each result. If relevance ratings fall, the content owner reviews topic level and audience targeting. If the session did not match its invitation, the presenter and marketer compare the published promise with the delivered scope. If people report access failures, the producer repairs the joining instructions before the next event.

The CDC separates process questions, about content, format and delivery, from outcome questions, about learning and transfer. That distinction prevents a smooth login and a likeable presenter from being treated as proof that people learned something. CDC: building an evaluation plan

Plan each item with four short notes:

1. Decision: What could change?
2. Evidence: What answer would inform that change?
3. Owner: Who reviews and acts on it?
4. Trigger: What pattern prompts action?

Set the trigger before results arrive. It may be a recurring theme, a change across several comparable webinars or one serious access failure. Avoid inventing a universal benchmark after seeing the numbers. A stable internal trend is useful only when wording, audience, timing and delivery method remain comparable.

Add an attendance router before the six questions

Begin with one separate routing item: How did you take part?

• Live
• Recording
• Some of both
• Registered but did not attend

Do not ask a no-show to rate the speaker, content or interaction. Instead, branch to a short attendance form: why attendance did not happen, whether the person still wants the recording and whether another time would have helped. Treat those answers as attendance evidence, not content evaluation.

Replay viewers can judge relevance, clarity and usefulness. They should skip live joining and interaction questions unless they experienced those parts. Branching shortens the form and keeps answers tied to something the respondent actually saw.

Record the number invited, registered, attending live, using the recording and responding in each route. A result describes respondents, not automatically everyone invited or everyone who attended. If 20 of 200 attendees reply, a strong rating is evidence about those 20 responses. It does not prove that the other 180 agreed. Comparisons also need stable wording, population, timing and survey mode. AAPOR best practices for survey research

The six-question starting core

Use the following questions for people who attended live, watched the recording or did some of both. The eligibility note beneath each item is part of the design. Adapt the nouns to the webinar's actual purpose and pretest the complete form before launch.

1. Ask about intended use before supplying ideas

Question: What, if anything, do you plan to use from this webinar?

Response: Optional open text.

Use: Tests whether the central idea translated into an intended action. It does not show that the action later happened.

Place this item before detailed ratings. Closed questions can introduce ideas that people then repeat in a later open response. Asking about intended use first gives respondents a better chance to answer in their own frame. Pew Research Center: writing survey questions

2. Measure relevance to the reason for attending

Question: How relevant was this webinar to the reason you attended?

Response: Not at all relevant; slightly relevant; moderately relevant; very relevant; extremely relevant; or not applicable, I did not attend with a specific need.

Use: Informs topic level, audience targeting and the framing of the next invitation.

3. Check whether the scope matched the invitation

Question: How closely did the webinar cover what the invitation led you to expect?

Response: Did not match at all; matched slightly; matched moderately; matched very closely; matched completely; or I did not see or cannot remember the invitation.

Use: Reveals a gap between promotion and delivery. Keep the opt-out separate so a forgotten invitation does not become a negative rating.

4. Ask about format only when interaction was available

Question: How was the balance between presentation and interaction?

Response: Too much presentation; about right; or too much interaction.

Eligibility: Show only to live or partly live participants when the session actually included an interactive element. Skip it for no-shows, recording-only viewers and presentation-only events.

Use: Informs the next run sheet. This item is adapted from the CDC's recommended question about the balance of lecture and interactive learning. Its options indicate a direction, unlike a generic overall-quality score. CDC: recommended postcourse questions

5. Diagnose friction with a route-specific list

Question: Which, if any, of these problems affected your experience? Select all that apply.

Response options:

• Could not access the live session or recording
• Audio problem
• Video problem
• Captions or transcript missing or inaccurate
• Slides or text hard to read
• Pace too fast
• Pace too slow
• Content too basic
• Content too advanced
• Interaction tool difficult to use
• Other
• None of these

Make "none of these" exclusive. Show only relevant options: a recording viewer does not need a live joining-link choice, and a presentation-only event does not need an interaction-tool choice. Follow one selected problem with one short diagnostic branch when the answer can lead to a repair.

Use: Separates technical, access, pace and level problems without asking one rating to carry all of them.

6. Leave room for the improvement you did not predict

Question: What one change would most improve the next session?

Response: Optional open text.

Use: Captures an improvement outside the organiser's predefined options. Keep it only if someone will read and code the responses.

Do not call reaction data proof of learning

An immediate survey can measure experience, relevance, perceived usefulness, intention and obvious friction. It is much weaker at proving learning, retention or behaviour change.

The CDC's recommended postcourse guide states that satisfaction-focused evaluations provide inadequate evidence of training effectiveness. It also warns that perceived learning is not a strong substitute for an objective test of understanding. If a webinar promises a skill, use a knowledge check, worked example, observed task or other measure tied to the learning objective when the claim matters. CDC: recommended postcourse questions

A retrospective pair such as "Rate your knowledge before" and "Rate it now" can describe perceived change, but it remains self-report. A higher confidence score is not evidence that the webinar caused durable competence.

Recommendation questions have a narrower job. "How likely are you to recommend this webinar?" may be useful when the owner has decided to track advocacy or compare a stable event series. It does not reveal what people learned, what they will use or why a score changed.

A later survey still measures self-reported use

At the end of a webinar, people can report what they intend to do. They cannot yet report whether they used the material. The CDC therefore distinguishes immediate intention from delayed questions about use, barriers and support. CDC: recommended postcourse questions

Send a short follow-up only after the audience has had a realistic opportunity to act. The delay depends on the task: a template might be tried quickly, while a budget decision or workplace procedure may take longer. Ask:

1. To what extent have you used what you learned? Use a small, fully labelled response scale.
2. What did you use? Offer optional open text.
3. What helped or prevented you from using it? Show barriers such as time, resources, opportunity, support or need for more training.

Call the result self-reported use. A later answer is closer to application than immediate intent, but it is not independent proof of behaviour. When behaviour evidence matters, add an appropriate work sample, task record, system log, observed performance or other objective measure, with the necessary permission and safeguards.

Keep delayed self-report separate from immediate reaction data. The populations may also differ because some respondents will not answer twice. Report both denominators rather than treating the second survey as a continuation of the first score.

Rewrite questions that measure several things at once

The UK Government Analysis Function's questionnaire guidance, which is currently marked as under review, recommends short, clear, balanced questions and warns against leading and double-barrelled wording. Current AAPOR guidance similarly recommends asking about one concept at a time and using response options that are mutually exclusive and cover reasonable answers. UK questionnaire design guidance AAPOR best practices for survey research

The supplied question bank asks, "How would you rate the clarity and quality of the audio and video?" That combines at least four judgments. A person may have clear audio and frozen video, but the form allows only one answer. Split the item when each answer leads to a separate action. Otherwise ask whether a technical problem affected participation, then branch to its type. Supplied post-webinar question bank

The same bank asks, "Were the speakers courteous, well behaved and interactive?" This also combines concepts and uses loaded wording. If interaction is a real decision, ask neutrally whether there was enough opportunity to ask questions. If nobody will act on a courtesy rating, remove it. Supplied post-webinar question bank

Label every scale point so a future organiser knows what a "3" meant. If wording, options, order, audience or delivery method changes, mark the break rather than presenting the new result as directly comparable with earlier surveys.

Pretest the form and plan the analysis

A survey can look clear to its author and still confuse the audience. Pew and AAPOR both recommend pretesting new questions. Ask a few people similar to the intended respondents to complete the form while you observe where they hesitate, misread a term or need an option that is missing. Pew Research Center: writing survey questions AAPOR best practices for survey research

Test branching on a phone, with a keyboard and with a screen reader. Confirm that the invitation explains why feedback is collected, how long the form is likely to take and whether responses are anonymous, confidential, pseudonymous or linked to registration data.

Open responses need an analysis owner. Decide before launch who will read them, how themes will be coded and how rare but serious access or technical problems will be escalated. The CDC guide advises omitting open-ended questions when there is no capacity to analyse them. A neglected comment box creates work for the respondent but no evidence for the organiser. CDC: recommended postcourse questions

Be precise about anonymity and personal data

The European Commission's GDPR guidance highlights purpose limitation, data minimisation and storage limitation. The rules can apply to an organisation based in the EU and to organisations outside the EU when they process personal data of people in the EU in circumstances covered by the GDPR. European Commission: GDPR principles

Removing the name field does not automatically make a response anonymous. Login records, unique survey links, IP data or platform metadata may still identify a person. Call responses anonymous only when the organisation cannot reasonably reconnect them to an individual. Otherwise explain whether they are confidential or pseudonymous and who can access them.

When a respondent may request a reply, collect contact details through a separate opt-in route where possible. Adding an email address to the same form makes that response identifiable. Link to the full privacy notice for the legal basis, recipients, retention period and individual rights rather than treating a short survey introduction as the complete notice.

Close the loop before sending the next survey

After analysis, record the decision, owner and change. A short note before the next webinar might say that the invitation now states the expected level, the joining guide was rewritten or more time was added for questions.

Then test the evaluation itself: can the team state how many people were invited, attended and responded by route; explain what each question changed; and distinguish immediate reaction, self-reported use and objective evidence? If not, fix the instrument before asking another audience to complete it.