TLDR: A real estate brokerage whose enquiries arrive mostly through listing portals does not own its pipeline; it rents it, on terms the landlord can change. When I mapped our enquiries by source and ran the forecast against a portal fee increase and a ranking change, the exposure looked less like a marketing preference and more like a supplier concentration risk. The hedge was building enquiry channels we own: the website's buyer question pages, the buyer email list, and referrals.
The head of sales keeps a whiteboard behind his desk with the month's viewings on it, and for a long time nobody asked where the names on the board had come from. I asked, one quiet Thursday in the off-season, mostly because I was new. I work in-house on the digital side of a luxury real estate brokerage in Marrakech, and I had been hired to look after the website. What I found that afternoon was a finance question wearing a marketing costume.
We went name by name. This one came from a portal. This one from a portal. This one from a former buyer's cousin. This one, again, from a portal. By the end of the board the pattern was obvious to both of us, and neither of us liked it. Roughly two enquiries in three, by my own count that day, had started on a website the brokerage did not control and paid to appear on.
A Supplier, Not a Channel
The portals are usually discussed in the marketing budget as a line item, next to photography and the paid ads. That framing hides what they are. A listing portal that supplies most of your enquiries is a supplier of your most important raw material, and you have one of them, maybe two, and they set the price.
Michael Porter's five forces give the language for it: a powerful supplier captures more of the value in an industry by charging more or by changing the terms. A portal can raise its listing fees, introduce a paid tier for placement, change the ranking so that the largest advertisers sit at the top, or start selling the same buyer's enquiry to three brokerages at once. Portals in other markets have done several of these things already. None of them require our consent, and none of them appear in the forecast until they have already happened.
That is the balance-sheet argument. The brokerage's revenue depends on a flow of enquiries. The flow depends on an asset the brokerage does not hold. An income stream that rests on someone else's asset is a liability waiting for its trigger.
What the Risk Analysis Looked Like
I am not a finance professional, so I kept the exercise simple enough that the head of sales and the owner could argue with every number. It had three parts.
First, share of enquiries by source, for the previous two years, taken from the CRM and the inbox rather than from memory. Portals, direct website, referrals from past buyers, referrals from the notary and other professionals, walk-ins at the office, and paid ads. I would not publish the exact split, but the shape was what the whiteboard had suggested: portals dominant, the brokerage's own website a small and unstable share, referrals steady and under-counted.
Second, the conversion of each source to a viewing and to a signed purchase, so the analysis was about revenue and not about volume. Portal enquiries were the most numerous and the least likely to close; referrals were the reverse.
Third, two scenarios run against the coming year's forecast:
- A fee increase on the main portal, at a size we judged plausible from its own pricing history. The cost line moved, the enquiry line did not, and the margin on portal-sourced deals thinned to the point where the sales team's commission structure stopped making sense on those deals.
- A ranking change that pushed our listings below the larger agencies and the aggregators. The enquiry line fell, and because portal enquiries were most of the pipeline, the forecast for signed purchases fell with it, by more than the owner had assumed possible from a change he would not even be notified about.
The exercise did not produce a precise number I would defend in a boardroom. What it produced was a conversation the brokerage had never had: how much of next year's revenue is subject to a decision made by a company we pay? The Harvard Business Review's collection on risk management is full of larger companies asking that question about suppliers and platforms. A brokerage with a small team of agents is asking it about a listing portal, and the logic is the same.
The Hedge: Channels the Brokerage Owns
Once the exposure was named, the strategy was less a marketing plan than a diversification plan. We were staying on the portals. We wanted a bad decision by a portal to move the forecast by an amount the brokerage could absorb, and no more.
Three owned channels came out of it, in the order we could build them.
The first was the website, rebuilt around the questions foreign buyers ask before they ever contact an agency. Our buyers mostly live abroad, in France, Belgium, Switzerland, the Gulf and West Africa, and their first questions are about foreign ownership, the notary's role, currency and repatriation, and which neighbourhood suits a family that visits six weeks a year. When I audited the site, of about sixty pages, twenty-six were effectively invisible to Google, and the keyword map pointed to several hundred pages worth building over time against sixty-three that existed. A buyer who arrives from one of those question pages on the brokerage's website has not been sold to us by a portal. Nobody can raise the fee on that enquiry.
The second was the buyer email list. Every enquiry, whatever its source, was asked whether they wanted the monthly note on new properties and on changes to the rules that matter to foreign buyers. A portal can stop sending us a buyer; it cannot take back the ones who chose to hear from us directly.
The third was referrals, which the analysis had shown to be the highest-closing source and the least managed one. A simple routine followed: a call to every completed buyer at six months, an introduction to the notary and the currency-transfer specialists we work with, and a standing thank-you for any name passed on.
What Changed in the Forecast
I will not claim the exposure is gone. Roughly, and this is my own estimate rather than an audited figure, the owned share of enquiries has grown to the point where the ranking-change scenario now moves the forecast by an amount the brokerage can plan around rather than one that would force a decision about headcount. The portals still matter. They matter as one supplier among several, which is what a supplier should be.
Two things I would tell a finance lead at any small business that sells through a platform it does not own:
- Trace every sale of the last two years back to its first touch, from the records and not from memory, and put the result in front of the people who approve the platform spend. The spend is usually discussed; the dependency almost never is.
- Run the platform's most likely bad decision through your forecast before it happens, and size the owned channels so that the result is a bad quarter rather than a crisis. Owned channels are slow to build and cheap to keep, which is the opposite of a platform.
The whiteboard is still behind the head of sales' desk. The names on it now come from more places, and he writes the source next to each one without being asked.