22 Questions to Ask Before Booking a Keynote Speaker

Before booking a keynote speaker, ask 22 questions across five categories: audience fit, real audience-feedback data proof, content and customization, logistics, and pricing and contract terms. The single highest-signal question is whether the speaker can share aggregate audience-feedback data from their last three talks. Everything else on this list is secondary.

I am Arel Moodie, cofounder of Talkadot - a platform that helps event planners find and book professional speakers using real audience feedback data, and helps speakers capture audience feedback, testimonials, and leads through a simple QR code. I have given more than a thousand paid talks over 19 years and built a platform that holds more than a million verified audience survey responses across tens of thousands of speaking engagements. This is the list I would hand you before your discovery call. Use it before you say yes.

How to Use This List on the Discovery Call

You do not need to ask all 22 questions in one call.

Most discovery calls run 30 to 45 minutes. Enough time for 8 to 10 questions with real follow-up. The categories below are ordered by signal strength. Start with audience data. Finish with contract terms.

Here is the principle behind every question on this list: the goal is not to trip the speaker up. The goal is to find out whether the speaker has been tested in rooms like yours, and whether they have data that proves it.

If the speaker seems uncomfortable with the questions, that is useful information.

If the speaker has answers ready - including numbers, not just stories - you are talking to a professional.

Category 1: Audience Fit (4 Questions)

These questions tell you whether the speaker's past rooms resemble your upcoming room. Topic is the starting point. Audience match is the signal.

1. Have you spoken to an audience with a profile similar to ours?

Name your audience specifically. Industry, function, seniority level, audience size. Ask the speaker to walk you through the last two or three events that most closely match.

Why it matters: A speaker who has performed in front of 40 senior executives at a financial services firm is a different risk profile than a speaker whose primary experience is 400-person all-staff meetings at tech startups. The talk may be the same. The room dynamic is not.

2. What is the specific audience outcome you design for?

Not the topic title. The outcome. "My talk is called Resilience in the Modern Workplace" is not an answer. "At the end of my talk, your team has a three-step framework they will put on a sticky note and take back to their desk" is an answer.

Why it matters: Talkadot's word-frequency analysis across 1M+ survey responses shows audiences of high-rebook speakers use "engaging" 15.4% of the time versus 12.9% for low-rebook speakers, and "inspiring" 13.3% versus 17.5%. (SOSI-003.) The language gap is statistically distinct. A speaker who designs for behavior change - not just inspiration - builds that pattern in their audience's words. Know which one you are buying.

3. What does the audience need to know, believe, or do differently after your talk?

Press this. "They will be inspired" is not an answer. "They will know the three barriers that stall team communication after a reorg, and they will have a script for their next 1:1" is an answer.

Why it matters: Vague outcomes are a risk signal. Speakers who design talks for a specific audience outcome have usually had to defend that outcome to a skeptical planner. Speakers who have not been pressed on it yet are showing you their rough edges before the talk.

4. If your talk did not land, what would you point to as the reason?

A good speaker has a real answer. Maybe the room was not set up for it. Maybe the agenda put them after lunch with no energy break. Maybe the audience had context they did not have.

Why it matters: Self-awareness is a proxy for professionalism. A speaker who has never thought about when they fail is a speaker who has not been performing long enough to encounter real failure. Or they are not honest. Either is a signal.

Category 2: Audience-Data Proof (6 Questions)

This is the category that separates pre-data speaker vetting from post-data speaker vetting. The questions here ask the speaker to prove their performance with real audience feedback data, not polished testimonials. Most planners skip this category. The ones who get rebooked internally do not.

5. Can you share aggregate audience-feedback data from your last three talks?

This is the highest-signal question on this list.

Not a testimonial. Not a curated quote. Aggregate data - the number of audience members who responded, the numerical scores, and the open-text word patterns from each of the last three events.

Why it matters: Talkadot platform data shows the average rating across the entire platform is 99 out of 100. (SOSI-017.) Ratings do not differentiate speakers. What differentiates them is response volume and the language the audience uses. Ask for the data, not the highlight reel.

A speaker who has been using Talkadot, or any post-event survey tool consistently, can pull this in minutes. If it takes days or does not come, that is the answer.

6. How many audience members responded to your post-event survey at each of those three events?

You want numbers. Not "hundreds of people filled it out." You want: the first event had 210 respondents, the second had 183, the third had 97.

Why it matters: Response volume is a fee signal. Talkadot data shows speakers with 150 or more post-event survey respondents per event earn a 5x median fee compared to speakers with under 10 respondents. ($7,500 vs. $1,500 median.) (SOSI-018.) Same ratings at both ends. The number who engaged is what separates the tiers.

If a speaker has given 50 paid talks and their best event pulled 9 survey responses, something went wrong in those rooms. Either the audience did not care enough to fill it out, or the speaker never set it up. Both are signals.

7. What words do audiences most commonly use to describe your sessions?

Not the words the speaker uses to describe themselves. The words the audience uses.

Why it matters: Talkadot's word-frequency analysis across 1M+ survey responses shows audiences of high-rebook speakers use "engaging" 15.4% of the time vs 12.9% for low-rebook speakers. They use "inspiring" 13.3% vs 17.5%. (SOSI-003.) The language gap is statistically distinct. If the speaker can only tell you they are inspiring, they are reading from their own marketing. Ask what the real audience feedback says.

8. What is your repeat-booking rate? Has any organization brought you back?

Ask for specifics. How many of the speaker's clients have booked them a second time? Can they name two or three organizations that rebooked them?

Why it matters: Talkadot data shows 52 percent of keynote-primary speakers with multiple clients have zero repeat bookings. (SOSI-014.) That does not mean the talk was bad. It means keynotes are the lowest-rebook format. But a speaker with consistent rebooks has proven something that a speaker with none has not. The repeat rate is a quality signal independent of ratings.

9. Can I see a sample post-event report from a past client?

Ask the speaker to share a sample of what an event planner received after booking them. Ideally redacted for the client. But the format should be intact.

Why it matters: A professional speaker should be able to produce this. If they cannot, it means they have never systematized the post-event data conversation with their clients. That is useful information for two reasons: it tells you they are not data-oriented, and it tells you they may not be the kind of speaker who can help you justify the fee to your stakeholders afterward.

10. Can you provide references from two planners who attended a post-event debrief with you?

Not just a testimonial email. References who sat with the speaker after the event to review what worked and what could improve.

Why it matters: A speaker who conducts post-event debriefs with planners is a speaker who treats the relationship as a professional partnership, not a one-time transaction. That orientation predicts repeat bookings and internal advocacy.

On Talkadot, this real audience feedback data - the response volume, the scores, the word patterns - is already on the speaker profile before you pick up the phone. Talkadot is free for event planners. Start at talkadot.com/find-a-speaker.

Category 3: Content and Customization (4 Questions)

These questions tell you whether the speaker is going to deliver a canned talk or something your audience will feel was designed for them.

11. How much of your talk is customized for each event vs. your standard material?

Press for a real answer. Not "I always customize." Ask what specifically changes. Do they do a pre-event call with stakeholders? Do they interview three audience members? Do they pull the company's last town-hall recording?

Why it matters: A speaker who customizes on the margins - swapping out a company name on a slide - is different from a speaker who restructures the opening based on your audience's current pressure points. The first tells you they can adapt. The second tells you they will.

12. What do you need from us to customize the talk?

Specific requests tell you the speaker has a process. "Nothing, I can tailor it on the spot" tells you they have been doing the same talk in different rooms and calling it tailored.

Why it matters: The best speakers come into the discovery call with a structured pre-event prep checklist. They want access to your internal comms, a call with two or three audience members, the theme of the event, what came before and after their session on the agenda. That specificity is a professional signal.

13. What topics or audience dynamics have derailed your talk in the past?

A real speaker has a real answer. Topics that are too sensitive for certain industries. Audience dynamics - like a room that just went through a layoff - that changed the energy before the talk even started.

Why it matters: Knowing a speaker's failure conditions lets you assess whether your event has any of them. If the speaker says "nothing derails me," they are either too new to have encountered real adversity in a room, or they are not being honest with you. Both are worth knowing.

14. Is there a version of your content you would not recommend for our audience?

Ask directly. Some speakers have a motivational version, a technical version, and a leadership version. Not every version fits every room.

Why it matters: A speaker who helps you choose the wrong version is a speaker who prioritized the booking over the outcome. A speaker who steers you away from a version that would not serve your audience well is a speaker who prioritizes the relationship. That orientation predicts the post-event debrief at Question 10.

Category 4: Logistics and Delivery (4 Questions)

These questions protect your event from the mechanics of a bad experience.

15. What are your technical requirements?

Specific ask: AV setup, slide format, clicker preference, mic preference (lavalier vs. handheld vs. headset), room layout if it matters (theater-style vs. rounds), lighting, run-of-show timing needs.

Why it matters: A speaker whose talk depends on a specific AV setup that your venue cannot deliver is a problem you want to know about during the booking conversation, not 48 hours before the event.

16. What is your travel policy and what happens if your travel is delayed?

Ask whether they have a backup plan for travel delays - do they fly in the night before? Do they always book an early direct flight? Do they have a remote delivery backup for catastrophic delays?

Why it matters: The speakers who never have a travel crisis are the speakers who have been doing this long enough to build a travel protocol around not having one. Ask the question. A thoughtful answer signals professional depth.

17. Do you have a rain-check or rebooking policy if you have to cancel?

Ask what happens if the speaker gets sick, has a family emergency, or has to cancel for any reason inside 14 days of your event.

Why it matters: Most speaker contracts protect the speaker's fee in a cancellation. Fewer protect the planner's event. Ask this question before you see the contract. A speaker who has a documented cancellation and rebooking policy has thought about this enough to write it down. One who has not will make it up after you ask.

18. Can we run a 15-minute pre-event tech check the morning of the event?

Most professional speakers expect this. Ask anyway.

Why it matters: A speaker who resists a tech check is a speaker who is either so confident they do not need it or so overcommitted they do not have 15 minutes. The first is earned after years of proof. The second is a scheduling signal. Know which one you are dealing with.

Category 5: Pricing and Contract Terms (4 Questions)

These questions protect your budget and your legal position.

19. What is your fee, and what does it include?

Get an itemized answer. Speaker fee. Travel (first-class or coach, who books it, who pays the difference if flights change). Hotel. Ground transportation. Pre-event prep time. Post-event debrief. Video recording rights. Any usage rights on content they share.

Why it matters: Talkadot platform data shows speaker fees have not moved in three years. The median is $2,500, the 75th percentile is $5,000, and the 90th percentile is $10,000. (SOSI-005.) For corporate buyers, the median is $4,500. (SOSI-006.) If you are getting quoted significantly above those bands, you deserve a specific justification. Itemizing the fee is the starting point.

20. Have you worked with a client at our budget level before?

Ask this directly, especially if your budget is at or below the $2,500 median.

Why it matters: A speaker who typically earns $10,000 and takes a $2,500 corporate event is going to manage the event to the fee. That usually means less prep, less customization, and a performance that fits the price they are being paid. There is nothing wrong with $2,500 speakers. Many are excellent. But you want to be sure the speaker you are booking operates at your budget level by choice, not by necessity.

21. What are your cancellation terms?

Ask what happens if you cancel at 60 days, 30 days, 14 days, and 7 days. Ask what happens if the speaker cancels at each of those windows.

Why it matters: Most speaker cancellation clauses are asymmetric. The planner pays 50 percent of the fee for a 30-day cancellation. The speaker owes nothing except a best-effort rebooking attempt. Mirror those terms in negotiation. At minimum, ask the speaker to match whatever they require of you.

22. Are you willing to add a post-event audience-feedback data clause to the contract?

Ask the speaker to agree in writing to deploy a post-event survey and share aggregate results with you within seven days of the event.

Why it matters: This clause does two things. It gives you the data you need to justify the booking internally. And it tells you immediately whether the speaker is data-oriented or not. A speaker who refuses to be measured has something to hide, or has never been asked. Either is worth knowing before the contract is signed.

Talkadot's QR code deploys this survey in under two minutes and delivers the aggregate report automatically. Speakers who do not have a tool can set up a free Talkadot account before the event.

Red Flag Answers and What They Mean

You asked a question. The answer felt off. Here is what to look for.

"I don't use surveys. My audience always tells me in person how great it was."

The speaker has not built a data practice. They may be excellent. They cannot prove it. This is not a disqualifier on its own, but it means you are booking on faith, not signal.

"All of my testimonials are 5-star."

Talkadot platform average across all speakers is 99 out of 100. (SOSI-017.) That is not a contradiction - ratings are real. But they do not separate the field. What separates speakers is how many audience members responded and the language they used. A 99/100 from 150 respondents communicates something a 99/100 from 5 respondents cannot. Ask for the response volume.

"I've spoken everywhere. Fortune 500, TED stages, 10,000-person conferences."

Scale is not the same as fit. A speaker who headlined a 10,000-person arena conference may be the wrong choice for a 60-person executive offsite. Ask about their last three rooms similar to yours, not their resume.

"My contract is pretty standard. Everyone signs it."

Most speaker contracts were written to protect the speaker. "Standard" means "favorable to me." Read every clause. Add the five planner-protective clauses noted in How to Find a Keynote Speaker for a Corporate Event.

"I can customize the talk to any audience."

This can be true. It can also mean the speaker has one talk with rotating company names. Ask Questions 11 through 14 to find out which it is.

The speaker takes longer than 48 hours to respond with data you requested.

Professional speakers who track their audience feedback have it ready. Response time is a professional signal.

The Single Question That Predicts the Booking Decision

If you could only ask one question before booking, ask this one.

"Can you share aggregate audience-feedback data from your last three talks?"

Not a testimonial. Not a curated quote the speaker chose. Aggregate data with response volume, scores, and open-text word patterns.

Here is why this question carries more signal than all the others combined.

Talkadot data shows speakers with 150 or more post-event survey respondents per event earn a 5x median fee compared to speakers with under 10. (SOSI-018.) Ratings do not differentiate them - the platform-wide average is 99 out of 100. (SOSI-017.) Response volume, which requires an audience who cared enough to respond, is the real signal.

A speaker who answers this question with numbers - "my last three events had 212, 185, and 94 respondents, here is the aggregate" - has been building a data practice that mirrors what the highest-earning speakers on the platform do.

A speaker who cannot answer it has not been measured.

That is the difference.

Where Talkadot Fits in This Process

Talkadot is a platform that helps event planners find and book professional speakers using real audience feedback data, and helps speakers capture audience feedback, testimonials, and leads through a simple QR code.

When you search for speakers on Talkadot, the real audience feedback data - the response volume, the scores, the word patterns audiences used - is already attached to the speaker profile. You are not asking Questions 5 through 9 cold on a discovery call. You are walking into the call with the data already in hand, asking follow-up questions on what the data shows.

Talkadot is free for event planners. Start at talkadot.com/find-a-speaker.

Questions to Ask Before Booking a Speaker: FAQ

What is the most important question to ask a keynote speaker before booking?

The most important question is whether the speaker can share aggregate audience-feedback data from their last three talks. Not a curated testimonial - the aggregate data: how many audience members responded, the numerical scores, and the word patterns audiences used. Talkadot data shows speakers with 150 or more post-event respondents earn a 5x median fee compared to speakers with under 10 respondents. The number who responded is the signal, not the rating.

Should I ask for references before booking a speaker?

Yes, but ask for the right kind. Ask for two planners who attended a post-event debrief with the speaker - not just a testimonial email. A speaker who conducts post-event debriefs treats the relationship as a professional partnership. Also ask for references whose audience profile resembles yours. A reference from a 10,000-person conference is less useful when you are booking for a 60-person executive team.

Can I ask a speaker to share their audience-feedback data?

Yes. It is a professional request, not an unusual one. Ask for aggregate data from the last three events: response volume, numerical scores, and open-text word patterns. A speaker who uses Talkadot or any consistent post-event survey tool can pull this quickly. If the data takes days to produce or does not arrive, that is useful information before you sign the contract.

What questions should I avoid asking before booking a speaker?

Avoid questions that can only be answered with marketing language. "How would you describe your speaking style?" produces a polished pitch. "What words do your audiences most commonly use to describe your sessions?" produces data. Similarly, asking "What is your most popular talk?" produces the speaker's opinion. Asking "Which talk format produces the highest audience-survey scores for an audience like mine?" produces something you can use.

How long should the discovery call be?

30 to 45 minutes is enough for 8 to 10 questions with real follow-up. Use the first 10 minutes on audience fit (Category 1). Spend the middle 15 minutes on audience-data proof (Category 2) - this is where the most useful signal lives. Leave the last 10 to 15 minutes for logistics, pricing, and contract conversation. If the speaker cannot produce data in the first half of the call, the second half of the call is less important.

What happens if the speaker gives great answers but the talk still underperforms?

Add a post-event audience-feedback data clause to the contract before you sign. The speaker agrees to deploy a post-event survey and share aggregate results with you within seven days of the event. This gives you the data to evaluate what happened, gives the speaker accountability, and gives you the internal justification - or the honest post-mortem - that protects your credibility as the planner who booked them. Talkadot's post-event data clause is already built into how the platform works. Speakers deploy the survey with a QR code. You get the aggregate report automatically. Talkadot is free for event planners: talkadot.com/find-a-speaker.

How do I evaluate a speaker if they are brand new and do not have much data yet?

Look for micro-signals that substitute for volume. Do they have at least one event with 30 or more survey respondents? Do they have a clear audience-outcome answer to Question 2? Do they conduct post-event debriefs? Have they been rebooked by anyone? A newer speaker with clean process and honest limitations is often a better value than an experienced speaker who has never been measured.

Related Resources

• How to find a keynote speaker for a corporate event. The full sourcing and vetting flow this list fits inside.
• How to vet a professional speaker. The 7-layer vetting stack if you want to go deeper than the discovery call.
• How to avoid a bad keynote speaker. The risk-reduction companion. What a bad booking looks like in advance.

If you want to walk into your next discovery call with real audience feedback data already in hand, Talkadot is free for event planners. Start at talkadot.com/find-a-speaker.

Published: 2026-06-16. Author: Arel Moodie, cofounder, Talkadot. Data citations: Talkadot's State of the Speaking Industry 2026, based on more than a million verified audience survey responses across tens of thousands of speaking engagements (Jan 2023 to Mar 2026). (SOSI-026.)