Your shortlist is ready and every finalist looks credible. How UK organisers score the final three speakers and get the decision signed off without stalling.
Link copied
The hardest part of booking a keynote speaker is not building the shortlist. It is choosing between two or three finalists who all look credible on paper, and being able to explain the choice to everyone who has to approve it.
Most guidance stops before this point. Our own complete guide to hiring a keynote speaker takes you through objectives, sourcing and evaluation; the companion piece on choosing the right speaker for your audience covers how to narrow the field. Both end where this article begins: three names, one slot, and a decision meeting on Thursday.
By the time a shortlist reaches three, the obvious filters have already done their work. Everyone left is available, broadly on budget and demonstrably experienced. What remains are differences that spreadsheets do not capture and that different stakeholders weigh differently.
This is not a failure of process. It is what multi-person decisions look like. Gartner's 2025 research on B2B buying groups found that 74% of buyer teams show unhealthy conflict during the decision process, with buying groups running from five to sixteen people across as many as four functions. Speaker selection rarely involves sixteen people, but the pattern is familiar: marketing wants reach, HR wants relevance, the sponsor wants a name the board recognises.
The same research points at the way out. Groups that reach consensus are 2.5 times more likely to report a high-quality outcome, and content framed around what the *group* is trying to achieve lifts consensus by 20%. Framing built around individual preferences has a 59% negative effect. Applied to a speaker decision, that means one thing: stop comparing the finalists to each other and start comparing each of them to the session you are trying to deliver.
A side-by-side comparison of three speakers produces an argument about taste. A comparison of each speaker against a fixed set of session requirements produces a decision.
Write the requirements first, before you look at the names again. Five lines is enough:
Requirement | What you are actually asking |
Audience fit | Does this speaker recognise the room's daily reality, not just its industry? |
Objective fit | Does the talk move people toward the decision or behaviour the event exists for? |
Slot fit | Opening, post-lunch or closing? Energy demands differ and they are not interchangeable |
Evidence of delivery | Showreel and references from a comparable room, not a bigger one |
Adaptability | Will the content be shaped to your organisation, or delivered as it stands? |
Score each finalist out of five on each line. The exercise usually resolves itself in the fourth and fifth rows, because that is where a well-known name and a well-matched name stop being the same thing.
The evidence on what audiences value is unambiguous, and it does not favour star power.
In Freeman's Learning Trends research, 59% of attendees said they prefer to hear industry experts at general sessions; the report's authors note that celebrity keynotes are not resonating. The gap between expectation and delivery is where the useful detail sits. 87% of attendees say it matters that a speaker gives real-world challenges and examples, but only 75% find speakers effective at it. 84% want takeaways they can implement, and only 64% find speakers effective at delivering them. Just 47% say general sessions add to their event's overall return.
Two of your finalists will look identical against the first figure and separate sharply against the last three.
There is a second gap worth knowing about, because it explains why organisers and attendees often disagree in the debrief. Freeman's Unpacking XLNC research found that attendees who experience a peak moment are 85% more likely to return, but only 40% report having had one, while 78% of organisers believe they are delivering them. Organisers point to keynotes and gala moments; attendees point to relationships, learning and networking.
The practical reading: the speaker who builds a session around what the room can use tends to outperform the speaker who builds one around what the room will recognise.
Some things are worth more than a low score. They should take a name off the list regardless of how strong the rest of the profile looks.
None of these are about seniority or fee band. They are about whether the session is being built for you.
What a full session reveals that a showreel cannot is the shape of the argument. Julian Treasure's talk on how people listen is a useful reference point for the difference:
How to Speak So That People Want to Listen | Julian Treasure | TED
It helps to see the finalists as three different kinds of session rather than three individuals. Speakers in our UK portfolio illustrate the distinction: a business transformation speaker such as Hamish Taylor frames change decisions for a leadership audience; a communication specialist such as Julian Treasure works on how people are heard inside an organisation; a data and mathematics speaker such as Hannah Fry reframes how a technical subject is understood by a general room.
These are examples of matching logic, not recommendations, and no availability or fee is implied by their appearance here. The point is that the three finalists on your list are almost certainly answering three different questions. Decide which question your event is asking, and the shortlist stops being a tie.
Once you have a preference, the delay usually comes from the approval step rather than the decision itself. Three things shorten it.
Bring one recommendation and two alternates, not three equals. A panel asked to choose will debate; a panel asked to approve or challenge a recommendation will decide.
Lead with the session requirement, not the speaker. "We need the post-lunch slot to hold a room of 200 regional managers who have already sat through three hours of numbers" makes the case before a name is mentioned.
Put the cost in context. UK keynote fees sit in recognisable bands, from around £1,000-3,000 for rising and regional voices up to £30,000 and above for global names; the fee guide sets out what moves a speaker between them. A finance stakeholder who can see the band and what sits inside it approves faster than one who sees a single number.
The decision is not finished when the name is chosen.
Book the briefing before the contract is signed. The brief is where a good match becomes a good session, and it is the step most often left until the fortnight before. Our speaker brief guide and template covers what belongs in it.
Keep the runner-up warm. Confirm availability with your second choice and tell them honestly where they stand. A shortlist of one is a plan without a fallback.
Agree how you will measure the session now. Amex GBT's 2026 Global Meetings and Events Forecast, based on a YouGov survey of 601 meeting professionals across eight countries, found that only 24% of organisations include ROI metrics in their meetings policy and only 26% have guidelines for post-event satisfaction surveys. Without a measure agreed in advance, the next shortlist will be built on the same instinct as this one.
If you would rather not run the final comparison alone, talk to us — we can put the finalists side by side against your session requirements and tell you where they differ.
---
Score each one against the session rather than against each other. Use five fixed criteria: audience fit, objective fit, slot fit, evidence of delivery in a comparable room, and willingness to adapt the content. Differences that feel subjective usually resolve in the last two criteria, because that is where a well-known speaker and a well-matched speaker separate.
A showreel made only of applause moments rather than argument, no questions asked about your audience, references drawn exclusively from much larger events, talk titles unchanged for three years, and reluctance to join a 30-minute briefing call before the booking is confirmed. None of these relate to fee band or seniority; they indicate the session will not be built for your organisation.
The evidence favours subject-matter expertise. In Freeman's Learning Trends research, 59% of attendees said they prefer industry experts at general sessions, and the report notes that celebrity keynotes are not resonating. Recognition attracts registrations; relevance is what audiences report valuing once they are in the room.
Present one recommendation with two alternates rather than three equal options, and open with the session requirement instead of the name. Gartner's research on B2B buying groups found that framing built around what the group is trying to achieve raises consensus by 20%, while framing around individual preferences reduces it. Adding the relevant fee band gives finance stakeholders the context they need to approve quickly.
Before you sign with your first choice. Confirm the second name's availability and tell them where they stand. Speaker diaries move quickly during the September to November season, and a shortlist reduced to one leaves you without a fallback if circumstances change.
Agree the measure before the event, not after. Only 24% of organisations include ROI metrics in their meetings policy and only 26% have post-event survey guidelines, according to Amex GBT's 2026 forecast. A single question asked two weeks later — which part of the session changed something in your work? — gives you more than a satisfaction score and improves the next shortlist.
After Dinner Speakers
Contents
