How the pitch decision actually gets made.
Watch the session
The setup
Agencies spend more money on new business than on almost anything else that is not payroll, and they spend it close to blind. The deck goes in. The team flies out. Two weeks later an email arrives that says the decision was "incredibly difficult" and the winning agency was "a slightly better fit." Nobody tells you that you lost in the first four minutes, or that a member of your team introduced himself to his own colleague on the Zoom while the client watched.
Rachel Huff built this session to close that gap. She is a pitch consultant, which means she sits between the client and the agencies and hears both sides of every conversation. She crowdsourced the questions agencies most often ask her and put them to two people who have sat on the buying side of the table for a combined several decades: a former CMO who has also run pitches from the agency side, and a UK procurement consultant who has written the RFPs agencies complain about.
40 minutes, no slides, and a card exercise in the middle where the panel had to call red flag or green flag on scenarios most principals will face this year. What came out was a fairly brutal inventory of the things clients notice, do not mention, and score you on anyway.
Who was on stage, and why they're worth listening to
Part One — Why agencies win
Rachel opened on the positive side, which turned out to be the more useful half. Her question: describe a time an agency won without being the flashiest or the biggest.
Deirdre's example is worth sitting with because the constraint was severe. She was a first-time CMO at a direct-to-consumer business running its first-ever brand campaign, with a board that had to be convinced brand spend was not a waste.
"The budget that they gave me for the brand campaign was 6% of the total media budget, and I needed to go big and prove to the board that brand was the right investment."
The brief asked for media and creative insight arriving together rather than in sequence. That is a requirement most agencies treat as a scheduling problem. Deirdre treats it as a character test.
"That requires you to work nicely in the sandbox, which I always say: don't make your problems my problems. Play nice in the sandbox."
The agency that won did not answer the brief as written.
"They came in with an idea that was not what we asked them for, but it delivered on the brief in a very innovative way, and made the budget that I have feel like four times the budget."
The tell in that story is which discipline led. It was not the brand agency. The media agency stitched the tactics together, and the result, she says, was one of the most strategic campaigns of a career that includes Google-scale budgets. A 6% media allocation beat multi-hundred-million-dollar work because somebody solved the actual arithmetic problem instead of presenting against the brief line by line.
Tina's example came from COVID, and it is smaller and stranger. An agency, pitching remotely, noticed how the client's own people were working and rebuilt its presentation to mirror it. She struggled to name what made it land, and the struggle is instructive: it was not the idea itself, which she rates highly, but the fact that the agency had put itself inside the client's working conditions before deciding how to present.
Then a detail that should worry anyone who thinks the scoring model decides these things. Tina, a procurement consultant, went back to the scorecard afterward.
"We had a scorecard to procurement. We got the scorecards, and it didn't really fit what the idea was on it. Actually, the ones didn't really matter."
The procurement person in the room is telling you the procurement instrument failed to capture the reason the winner won. Hold that thought for the section on folklore.
Rachel pushed for moments where an agency won the room in real time, and Deirdre produced the session's most transferable story. An agency arrived at her pitch having misread the brief. The standard response is to present anyway, because the deck exists and the flights are booked.
"This agency did the brilliant thing of saying, you know what, we missed the mark. It's clear that what we prepared to show you today is not going to meet your expectations. They called an audible."
Abandoning a prepared pitch mid-meeting and working the real problem live — calling an audible — only works because it proves something a deck cannot. Deirdre could see that the team knew her business well enough to improvise inside it, and she could see how the shop behaves under pressure, which is the thing she is buying.
"I know what it's going to be like at 11 p.m. when we're in a bind trying to figure out how we make this happen. I got to see it real time through that pitch process."
And the part that mattered most to her had nothing to do with the leader:
"I also loved that it wasn't just the most senior person in the room who was leaning in to say let's pivot. I actually got to see how the working team managed through that situation and how they got on their feet."
She is careful to note that not every client would allow this. But the risk calculation is worth doing, because presenting a deck you know is wrong guarantees the loss you are trying to avoid.
Rachel offered the counterweight, from a virtual pitch she ran recently. She joined the Zoom a few minutes early, as she always does. Two members of the agency team joined too, and introduced themselves to each other.
"It was so clear in the meeting that the agency team did not know each other well and that they had not really worked well together in the past. I gave that feedback to the agency leader."
The agency's defense was that this was an early, introductory call. Rachel's response is the line to write down.
"Doesn't really matter. These things show."
They did not make it to the next round. Not because they were unqualified, but because 90 seconds of pre-meeting small talk revealed that the team on the org chart was an assembly rather than a team.
Part Two — The card exercise
Rachel handed out red and green cards and ran scenarios. The instruction was to commit to a color before explaining. The disagreements are where the value is.
Pushing back on the RFP process. Tina defended the instrument on fairness grounds, and she is the person who has to administer it. Getting procurement involved, she argues, exists to make sure every agency is treated the same. Her caveat is about proportion rather than principle:
"My view is, as long as the RFP process fits the brief. That is my one caveat. It should fit the task."
Her example of the failure mode is a 200-page document attached to a brief that does not warrant it. Deirdre agreed on the substance and drew a sharper line about how to object.
"If there's something in the process that you think could be additive, then you can offer a suggestion. So it's not a complete pushback. Now, if they don't take your suggestion, then you've got to move on."
Rachel, whose job depends on the process working, sided with the agencies with a condition attached: bring a reason and a solution, not a complaint. She has seen it work.
Then she took a position that runs against how most searches are administered. Clients frequently distribute the answers from one agency's questions to every agency in the process, in the name of fairness. Rachel does not think that is fair at all.
"If they ask a really smart question, they should get the answer back."
Deirdre's build on group Q&A sessions is one line and it explains why your best questions go unasked:
"I feel like you're giving away your secret."
If you have been sitting on a differentiating question because the Q&A is a group call, that instinct is correct and both sides of the table know it.
No direct category experience. The only scenario where the panel came close to a consensus red flag, and only in two sectors. Rachel called it a red flag in pharma and financial services on the evidence of recent searches. Deirdre took the opposite view everywhere else:
"I have seen brilliant work by agencies who did not have experience at all in the sector that I was in, because they were bringing innovative thinking from other categories to my business. I am definitely of the school of thought more times than not not to be too narrow."
She has taken an experiential specialist and made it lead agency on a full campaign. Her qualifier is worth noting for anyone pitching a large brand: at Google-scale clients, that openness was rare. At businesses between $1 billion and $5 billion in revenue, she says, scrappiness was survival and the bets paid off.
Tina's UK view is that category experience is a proxy clients reach for instead of doing the work. Her market context is the reason:
"In the UK alone, we have 25,000 agencies."
With that many shops and a conflict rule that removes your best candidates the moment they have category experience, screening on sector becomes self-defeating. Her reframe of what the client should be asking:
"Clients should be asking: what is my business problem, and finding someone that can deal with that business problem. Not 'I want an agency that's X, Y and Z.'"
Deirdre, a year into living in Europe, asked Tina directly whether the UK is a "round hole, round peg market" compared with the US growth-mindset approach she was used to. Tina's answer was that it depends on the client, followed by a genuine worry about junior marketers who cannot see past a TikTok execution to the long-term brand build. The exchange did not resolve, and it is the most interesting unresolved thing in the session.
Introducing new team members at the final pitch. Rachel argued it can be green, when the addition is visibly a response to something the client said in an earlier meeting. Deirdre agreed only under that condition, and named the cost:
"You've lost all of this momentum. It's the process of building a relationship, understanding the atmosphere, seeing the rapport."
A less polished pitch than the competition. Both sides called this green, energetically. Tina's contribution is the most quoted thing in the session and it is a joke about hospitality theater, delivered by the person who signs off on cost. She recalled an agency that spent about £50,000 on a pitch, and an agency reception where the staff proudly described the in-house pitch theater team and the lighting rig. Her verdict on that: too much. Then the joke, which is also not a joke, about what procurement clocks on a visit: the biscuits, and the hand cream in the toilets. Small signals about how a shop treats people, read by someone trained to read cost.
The agency seems very hungry. Unambiguous green, and the panel's frustration was that agencies suppress it. Deirdre's version connects hunger to the part of the meeting agencies treat as throat-clearing:
"You can win a pitch in the intro if you present that thoughtfully and with the energy and excitement, because it shows that you've gotten into my business, really love it, and you want it."
And then the omission she cannot understand:
"A lot of the agencies also don't ask for the business at the end. It always surprises me. So ask for the business."
The agency says no to part of the assignment. Deirdre's answer was "it depends," and the dependency is entirely about how the no is constructed.
"No without context for why, and no without even bringing me an alternative. I've been media trained, right? You can answer the question without answering the question. I prefer you do that."
A former CMO is telling agencies to use media training on their clients. The point is not evasion. It is that a flat refusal transfers your problem to the buyer, and a redirected answer does not. Tina's frustration is about timing rather than substance: an agency that goes all the way through a process and then declines has burned everyone's time, and she says so with more profanity than this document needs. Her conclusion is more forgiving than her tone: saying no is hard for any agency, and a no delivered at the right moment does not hurt you.
The agency won't engage with procurement. Both called it red, briefly, and then Deirdre flipped the scenario in the direction agencies never hear:
"I get really frustrated if it's procurement only. At a certain point, the client needs to stay engaged, and you need to represent the needs of your team, because procurement doesn't know your business better than you do."
Rachel's role in that gap is the most portable idea in the session:
"I often will say to my clients, use me as the procurement translator. I'm the go-between to help them understand what you need, but also understanding their needs."
The agency can start next week. Tina, red, and her reasoning should change how you write your availability line:
"It just makes me a bit nervous. Maybe because they've got resources, but are they not managing the resource?"
Full availability reads as either idle capacity or a conflict you have not disclosed. Say why you are free.
The uncomfortable truth
Rachel asked what matters more than agencies realize. Deirdre prefaced her answer by saying it might be unpopular, and then made the session's least comfortable point.
"This might be unpopular, but I would say casting. The number of times I've been in a room where I am a Black female CMO working in a female category, and the entire pitch team is a room of white guys."
The frustration is not abstract. It is about what a client is forced to conclude in the absence of any explanation.
"The silent business team is more dangerous, because then I'm drawing my own conclusions about what your agency looks like."
Her argument is commercial before it is anything else. If the agency cannot assemble a team that resembles the customer base, the client has to ask how the work will speak to that base. She puts the standard plainly: the best agencies mirror what the world looks like.
And then the part that will not appear in anyone's DEI statement. Tina noted that the conversation has gone quiet in the US. Deirdre confirmed it, and confirmed that the quiet changes nothing about the decision.
"Now in the US, you can't even really talk about it. But I'm going to tell you, whether I can talk about it or not, I'm making a decision based on the casting that's in the room. It's the reality."
The political weather has removed the language, not the scoring. Agencies that read the retreat of the conversation as the retreat of the criterion are misreading their buyer. This is the passage of the session that most deserves to be repeated inside agencies that have quietly stopped talking about who is in the room.
The related watch-out came immediately after, and it costs agencies work every week:
"Another watchout is when the senior-most person in the room talks over the team. I know it's hard, and we feel like they might not be answering the question the way that you would answer it. But be really mindful of how you interject, because it shows to us a lack of confidence in the team."
If the leader cannot let a team member finish an answer, the client concludes the leader does not believe the team can do the work. And Deirdre added the version of this that nobody will ever tell you to your face:
"It also might bring to question, like, he always talks over the woman. Things that we notice that probably people won't tell you. I will tell you that that is something that I notice."
Rachel's variant is structural rather than hierarchical. Someone answers a question, then a second person adds to it, then three more pile on, none of them adding anything. She calls the standard "meaningful builds." If your build is not one, do not make it.
Part Three — Why clients fire agencies
Rachel moved the session to the losing side, and the answers reframe how agencies think about account risk.
Tina's is the phrase most likely to survive this Blueprint:
"They don't fire you overnight. It's a series of small disappointments. Lack of controlling the cost, lack of transparency, senior management not continuing. It builds up over a period of, I would say, up to six months."
There is no single moment to defend against. There is a slope, and it runs for two quarters before anyone tells you.
Deirdre's exception is the one thing that skips the slope entirely.
"The one thing that will get you fired immediately is if I catch you lying. Just don't lie. The whole agency-client relationship is built on trust, and I will fire an agency without a really good backup plan."
Rachel's addition is that most trust erosion is a communications failure rather than a performance failure. She has just handled a case where a client discovered its agency was working with a competitor and had not been told. The more common version is a senior departure that goes unacknowledged.
"Sometimes they don't have a solution right away, and that's okay. But to stop and have the conversation, to say we know this was a really important team member, we know you're really going to feel this. This happens all the time where they just don't talk about it. They don't address the elephant in the room, and it erodes trust."
Deirdre's structural fix is to build the meetings before you need them.
"There's nothing worse than when you see the emergency meeting on your calendar. It's better if it happens during a regular."
She recommends stakeholder mapping from the agency side, matching the right people to the right regular check-ins, so the hard conversation has somewhere ordinary to happen.
Tina extended that to the function agencies stop calling the day the contract is signed.
"Procurement do the pitch, and then you never see them again. Have quarterly meetings with them. Procurement is a client as well now."
Her reasoning is that scope creep and cost drift are exactly the small disappointments that accumulate, and procurement is the function positioned to catch them early. Keeping procurement in a quarterly review turns your most dangerous stakeholder into an early-warning system.
Deirdre's money principle is short enough to put on a wall:
"Not spending my money like it's your money."
The specific behavior she means is the agency that waits until halfway through to reveal the work cannot be done inside the agreed budget. And the last thing she said in that stretch was the thing she says she did not understand until she crossed the table:
"Do not underestimate how small a part of your client's time whatever you guys are working on is. I left the agency always frustrated, thinking I don't understand why clients aren't making time for this. And then I became a client. Oh my gosh, you guys are 10% of all the nonsense that I have to deal with on a daily basis."
90 percent of your client's working life has nothing to do with you. An agency that treats client time as the scarce resource it is has an advantage that no case study can buy.
Part Four — Folklore worth retiring
Rachel closed by asking what pitching mythology should be put down.
Tina went first, at her own function's expense:
"I would say procurement's only there to negotiate. Sometimes it is, to be fair. But I do think agencies don't do enough work, because there are some that will be looking at models and structures."
Her counter-example is Bayer's global pitch, which she describes as an inspirational piece of work on the strength of what it did with the ecosystem and the operating model rather than the rate card. Her point is that agencies who assume procurement is a price function walk past the part of the process where procurement is designing how the relationship will work, and where an agency with a view could shape it.
Deirdre's build inverts the assumption agencies make about what procurement wants:
"I always worry if it's too cheap sometimes. It's rare that I've chosen the cheapest solution. That can be challenging with procurement sometimes, because I feel like this feels too good to be true."
Followed by the preference that follows from it: give the full view on costs up front rather than surprising the client with an extra of this and an extra of that, because the client is planning around a budget.
Rachel's target was the vocabulary itself.
"The word pitching gives the connotation that the agencies are up there talking at the client, and I hate that idea. It should just be more of a conversation."
She reports the same allergy to "RFP" and substitutes "the brief" wherever she can. Both panellists agreed, and both admitted they use the old words daily.
Deirdre raised a live trend she is still forming a view on: over the last three years, agencies increasingly decline to pitch at all, preferring to start with a paid project. She is comfortable with it in some circumstances and uncertain about it in others. Tina's response was that refusing to pitch is a luxury most agency businesses cannot afford, and then she told the story that undercuts her own caution. Years ago, an agency skipped the process entirely, walked in and said it wanted the business, and won it. The pitch never happened. What won was the wanting, made visible.
The session's last word was Tina's, and it reframes everything above it:
"The final pitch isn't where the final decision is. It's the build-up."
Including, she added, the agency website with the generic hello on the homepage.
What you can use
Every session from Indies @ Cannes 2026 — plus every other Blueprint, the tools, and the surveys — opens up with one sign-in.
