Where creator–brand deals actually break.
Between September and November 2026 we wrote to 135 creators, freelancers, studios, agencies and the brand managers who hire them, and asked one question: where does a paid collaboration fall apart — agreeing the price, agreeing what will be delivered, or getting paid afterwards. Twenty-eight of them answered in substance. This page is what they said.
Nobody was paid and nothing was sold. We asked because we are building the thing, not to collect testimonials. Several answers contradict how we had designed it; those are in here too.
Published 10 November 2026. PortalBunny is in its pilot.
Related: how much to charge for a sponsored LinkedIn post · how a deal works · how we differ.
Who we asked, and what this is not.
The sample. 135 first messages sent on LinkedIn from two founder accounts, one person at a time, each written for that person. 28 substantive replies. Respondents sit on both sides of the deal: creators, UGC specialists, photographers, videographers, ghostwriters and studios on one side; brand managers, influencer-marketing leads, agency owners and partnership managers on the other.
Countries represented in the answers. United States, United Kingdom, Portugal, Germany, India, Italy, Spain, Norway, Ukraine, Serbia.
The question. One question per message, and always the same shape: when a brand approaches you now, which step most often falls apart — the price, the scope, or the payment?
What this is not. It is not a survey and the sample is not representative. It is self-selected: people who answer a stranger’s message are, by definition, people who had something to say. The value is not in the proportions, it is in the fact that the same specific sentences came back from people who have never met each other and work on different continents. Where we give a number, it is the number of conversations, not a percentage of a market.
Names. Quoted by name only where the person gave explicit permission. Everyone else appears as a role and a country. We did not ask for permission in advance, so most of this is reported without names on purpose.
Five patterns.
In the order of how often they came back, not in the order of how interesting they are.
01Scope grows because refusing is the expensive option
The single most useful answer we received, and it reframed our own product. A brand marketer who is also a creator told us that the work expands after the price is agreed — and not because the contract was weak. It expands because saying no to a client damages the relationship, so doing the extra thing quietly is cheaper than defending the boundary.
That makes the fix a process problem rather than a legal one. A contract does not help if nobody wants to invoke it. What helps is an agreed route for additional work, so that asking is a normal step instead of a confrontation. The same shape came back from a photographer in India, who described cancellations and reschedules the same way: the terms exist, nobody wants to be the one to raise them.
02The brief is where the money is actually lost
Asked where things break, more respondents pointed at the brief than at the price or the payment. A brand strategist in Lisbon put it best: a list of deliverables survives the process, but the reason behind the idea disappears first — three reels get made, and the feeling the brand actually wanted does not.
An agency SMM in Lisbon named the mechanical version of the same thing: even with a clear brief, the date slips if revisions go in circles and nobody knows who gives the final yes. A media strategist with fifteen years at large agencies added that creative work is judged differently by different people, so checkpoints during the work matter more than precision at the start.
The practical consequence runs through everything else on this page: a requirement that cannot be checked is not a requirement. “Make it feel premium” and “SEO-optimised” have no resolution except stubbornness. “Mention the free trial”, “use this phrase”, “include the ad label” do.
03The advance is a local custom, not a standard
We had assumed paying before the work starts was a universal improvement. It is not universally read that way.
A copywriter and concept writer in Germany told us German clients are cautious about paying everything up front and 50/50 is the habit — not a dealbreaker, a habit. A photographer and videographer in India said whether an advance is normal depends entirely on the client; it is contextual, not standardised. A UK freelance journalist and UGC creator, George Priestman, who gave us permission to use his name, said the opposite problem: asking a new client for a deposit inside a contract puts small clients off, while money that is already there would help a small creator.
And a video studio owner in India, Abhay Bhattacherji, also by permission, told us his clients have no payment problem at all — he takes no deposits and works on trust, because his clients are long-standing. Which told us something we needed: an established studio with repeat clients is not the person this solves a problem for.
04Approval has no clock
A LinkedIn ghostwriter for founders in Delhi named a failure nobody else had: founders sometimes take weeks to approve a draft, and the finished work sits unpaid in the meantime. Not a dispute, not a disagreement about quality — just silence with no deadline attached to it.
It is the quietest item on this list and the one with the clearest fix. A draft step without a time limit is not a safeguard. It is a waiting room.
05The thing creators cannot produce is proof
A ghostwriter told us her hardest problem is not finding clients or agreeing a price: it is proving to the next client that the last job went well. An AI video producer with a 100% job success score on a freelance platform told us the score does not travel — it stays on the platform that issued it, and a serious client still has to be convinced from zero.
Several creators described their actual workaround for the payment half of the same problem: a watermark, or a low-resolution file, held back until the money lands. That is a self-made deposit, and it costs the creator the first impression of their own work.
What we changed because of these answers.
Published here because a research page that changed nothing is marketing.
A clock on approval. The buyer has two days to respond to a draft, with a reminder a day before. Silence counts as approval, and the time the buyer spends deciding is added to the publication deadline so the creator is not made late by someone else’s delay.
Complaints only against the brief. An issue can be raised only against a specific requirement from the accepted brief — the post is missing, it materially differs from the approved draft, a required tag or ad label is absent. Taste is not grounds for rejection. That is a direct answer to finding 02: it forces requirements to be written in a form that can be checked.
The record, not a rating. A finished deal leaves a record on both sides — how many deals, with how many different people, posted on time, accepted. No stars and no reviews, because a review is a story someone chose to tell and a record is what happened. That is finding 05.
What we did not change. We did not make the advance optional. Finding 03 says the custom is local, and we accept that this costs us some creators in markets where 50/50 is the habit. We think a price agreed and paid before the work starts removes a worse problem than it creates, and we would rather be wrong about this in public than quietly.
What we still do not know.
Whether a fixed price survives contact with work that is priced per word, per day or per month. A freelance SEO writer in Lisbon prices per word and sells retainers; a figure calculated from an agreed count is still fixed, but a retainer is not a deal with an end, and we do not do retainers.
What a brief field should be called so that the reason behind the idea actually gets filled in, rather than left as free text nobody reads twice.
Whether a public record pushes work away from a platform rather than onto it — raised by someone who works on creator monetisation at scale, and still unanswered.
If you work on either side of these deals and any of this is wrong in your market, we would rather hear it than not: services@portalbunny.com.
Citing or writing about it?
The findings on this page may be quoted with a link to this page. If you are writing about the creator economy and want the underlying material, the question wording, or an interview, write to services@portalbunny.com — we will answer in writing. Applications for creators and brands are open on the home page.