Home/Journal/The Broker Who Walked Away From Onefinestay
Profile  ·  2026

The Broker Who Walked Away From Onefinestay: Why

A late-afternoon conversation on the 28th of March 2026 in a small office above a Provence square, with the founder of an independent broker that placed 48 villas through Onefinestay between 2017 and 2023. Onefinestay was acquired by Accor in 2016 for a reported $170 million. By December 2023 her last property had been delisted from the platform. We sat with her to walk the 14-month wind-down, the three policy shifts she says forced the decision, the specific incident in March 2023 that closed the door, and what replaced the book.

By The Villas For Kings desk

Onefinestay was founded in London in 2009 as a high-end shortlet platform, scaled through New York and Los Angeles in the early 2010s, and was bought by Accor in April 2016 for a price reported in the FT and Reuters at the time as $170 million. The platform's villa book grew rapidly into the late 2010s and then began to compress in 2021. By 2024 the platform was holding a meaningfully smaller villa inventory than it had in 2019, and the broker community that had fed the early growth was visibly thinning.

The broker we interviewed has agreed to be named in our autumn 2026 follow-up. For now we mark her and her agency. She is one of approximately 14 independent brokers who, by her count, supplied Onefinestay with more than 30 properties each at the platform's 2018 to 2020 peak. She estimates fewer than half of those 14 remain meaningful suppliers in 2026.

What follows is the wind-down as she described it: the three policy shifts after the 2016 acquisition that progressively misaligned the platform's incentives with the broker's, the March 2023 incident that closed the relationship, and the direct-let book that has now replaced it.

Shift I  ·  2018 to 2020

The commission band that moved twice.

"The original commission band was negotiable. When I joined the platform in 2017 the band on my book was 18 percent. The band reflected the fact that I was bringing inspected, staffed properties to the platform that needed only listing photography and pricing support. The platform agreed. The 18 was a fair number."

"The band moved to 22 percent in 2019. The platform argued that the increased marketing spend warranted a wider margin. Most of us absorbed it. Some passed it through to the owners. A few owners pulled from the platform at that point. We held the book."

"The band moved to 25 percent in 2021. The argument was the post-pandemic distribution cost. The argument did not hold. The 2021 demand was substantial. The platform's marketing spend per booking should have been falling, not rising. The owners who had absorbed the first move began to push back on the second. Three of mine pulled at the 2022 renewal. The platform did not offer to negotiate."

The commission creep is the pattern most independent brokers cite when explaining why they reduced their platform book. The platform did not present the increases as a take-it-or-leave-it choice but the owners read it as one. The broker absorbed the disputes in the middle. Our work on platforms that bury the cleaning fee shows the same upward-margin pattern on the per-booking line. The structural pressure is consistent.

Shift II  ·  2020 to 2022

The guest service centralisation that broke the local thread.

"The second shift was guest service. Until 2020 the platform's villa guests, on most of my book, were handled by a regional guest manager I knew and could reach by phone. The regional manager held context on the property and on me. If an issue arose in the first 24 hours of a stay, the regional manager called my local fixer. The fixer addressed the issue. The owner was insulated. The guest experience held."

"In 2021 the platform centralised guest service to a hub model. The hub did not hold context. The first call on an issue went into a ticket. The ticket sat. The local fixer was reached by email after a 6 to 14-hour delay. By the time the local response went out, the guest was in the second day of the stay and had escalated to the broker directly."

"I absorbed the escalations for two seasons. The brand still carried weight. By 2022 the guests were beginning to write to us before the trip about whether the platform's customer service was the platform's customer service, having read about the change online. The brand was no longer carrying. I started to ask the platform for a service-level agreement on first response. The platform did not commit to one. That was the moment I started planning the wind-down."

Shift III  ·  2022 to 2023

The payout-timing change that pushed working-capital risk back onto the broker.

"The third shift was payout timing. Until 2022 the owner payout was issued on a clear schedule: 50 percent within seven days of arrival, the balance within seven days of departure, subject to the standard deposit reconciliation. The schedule was predictable. The owners ran their working capital against it."

"In late 2022 the platform extended the post-departure window from seven days to 30 days, citing fraud screening. The change was applied retrospectively to bookings already on the books. The owner cash flow was disrupted. The broker, who had front-loaded marketing spend against the original schedule, sat with the carry. I had four properties whose owners called me directly about why their payout was late. Each of those four calls cost me approximately two hours of explanation and a week of trust."

"The platform did not invite a broker conversation about the schedule change. The change was a policy email. The brokers were treated as a distribution channel. By that point I had stopped placing new properties on the platform. The 48 in the book ran through their contracted seasons. They came off as the contracts expired."

The incident

The March 2023 stay that closed the door.

"A Provence villa, six bedrooms, week of the 11th to the 18th of March 2023. The guest paid in the region of €42,000 for the week. On arrival the heating system in the main building failed. The local fixer was on site within 90 minutes and confirmed the failure. The temperature in the master suite was 12 degrees. The guest had a two-year-old with them."

"I authorised a full refund on the night of arrival, in writing to the guest and to the platform's hub. The platform's policy was that the refund would be processed by the platform after the stay had ended. The guest left the next morning. The platform took 26 days to process the refund to the guest. The guest had to call their card issuer and dispute the charge in week three. During that period the guest's emails came to me, not the platform, because I had been the human voice in the booking."

"I was carrying a refund on a stay I had no operational responsibility for. The owner had been reimbursed by their insurance for the failed system. The guest had been refunded by their card issuer through the chargeback. The platform was reconciling the three lines on its own clock. I was the person the guest had spoken to for the duration. That was the incident. I notified the platform the following week that I would not be renewing any of the 14 contracts due in 2023, and I would be releasing the remaining 34 in 2024."

The incident has its parallels across the aggregator-platform era. Our work on the non-refundable deposit scam and the villa deposit disappearance pattern traces the same broker-as-shock-absorber problem across other platforms. The pattern is structural. The broker absorbs the timing risk the platform extracted from the guest.

What replaced it

The direct-let book she rebuilt in 2024.

"By the end of 2023 the platform book was gone. I rebuilt the book on a direct-let model. The owner contracts directly with the agency. The agency holds the booking and the payout schedule. The agency pays the owner on the original seven-day post-departure schedule. The agency absorbs the cancellation risk on a small float."

"The direct model carries higher operational cost per booking. The agency does the photography, the listing copy, the payment processing, and the guest screening. The trade-off is that the agency keeps 18 percent rather than passing 25 to the platform. The owner sees a higher net. The guest sees the same service level the agency was already providing."

"The book in 2026 is 31 properties, down from 48. Smaller. The smaller book is also tighter. The properties left are the ones where the owner and the agency are aligned on the model. The properties that left were properties where the owner wanted the platform's distribution. Those owners are now on aggregator listings. Some of them are doing fine. Some are calling me back."

The direct-let rebuild is the route a handful of independent brokers have taken in the past three years. Our coverage of Le Collectionist and The Thinking Traveller looks at the platforms that have managed to retain the broker-aligned model at scale. The category is small. The broker-aligned platforms charge their own margins. The trade-off remains.

The takeaway

What the buyer should take from this.

"The platform is the brand the guest sees. The broker is the person doing the work. When the platform's incentives stop aligning with the broker's, the broker either leaves or stops investing. The guest who books a platform villa in 2026 should ask, before signing, who placed this villa on the platform, and whether that broker still has a working relationship with the platform.

"If the broker has been on the platform for more than five years and is still placing new inventory there, the platform is functional. If the broker has been on the platform for more than five years and has stopped placing new inventory there, the platform is hollowing. Most guests will not get this answer. The guest who asks the question will learn something useful from the silence."

The question is the kind we now recommend a $40,000-a-week buyer ask any aggregator platform before booking. The platform is the distribution. The broker is the operations. A platform without engaged brokers is a directory. A directory at a 25 percent margin is a problem for the owner, the broker, and the guest, in that order. Our platform review hub tracks which platforms are still drawing fresh inventory and which are not.

FAQ

The Onefinestay walk-away, answered.

Who owns Onefinestay? Accor. The acquisition closed in April 2016 at a reported $170 million.

What was the broker commission band over time? 18 percent in 2017, 22 percent in 2019, 25 percent in 2021 on her book.

What was the March 2023 incident? A €42,000 Provence stay with a heating failure on arrival. A guest refund authorised on the night of arrival took 26 days to process. The broker absorbed the guest's calls.

What did the broker do next? Wound the 48-property book down through 2023 and 2024 and rebuilt 31 properties on a direct-let agency model.

What should a 2026 buyer ask the platform? Whether the broker who placed the villa is still placing new inventory on the platform.

The For Kings Network

Provence beyond the villa.

Our sister sites cover the hotels, restaurants, and bars that anchor the same Provence season the broker has worked for 19 years.

Newsletter

Get the buyer's guide.

One email a week. Broker interviews, platform audits, and the villas we pass on. Subscribe to the buyer’s brief.

Last updated 2026-04. We have not adjusted our editorial for the commission rate. See how-we-make-money for the full disclosure.