Around 40 percent of Singapore’s registered property agents do not close a single deal in a year.
Every industry with that shape hands training to its veterans, on the theory that whoever has done it longest knows it best. Christian Oh’s position is that the useful trainer is the one who still remembers earning nothing, and that the memory expires.
He was eighteen months past his first closed deal when six people started working under him.
He had no teaching to do. What he had was a system that had just started working and that he had written down in full: the phone script, the objections he kept hearing and the answer that moved each one, which blocks he had walked and what he had found there.
Training meant proximity rather than instruction. He knocked, they watched, then they knocked and he stood behind them and told them what he had heard.
I wasn’t the guru. I was eighteen months ahead.
The gap is the qualification. Someone ten years in has forgotten what it is to have nobody pick up the phone; Oh had spent a full year there, at two hundred doors and a hundred calls a day, and had made zero dollars across the whole of it.
I wasn’t the guru. I was eighteen months ahead.
He is also a self-described introvert who used to look at the floor instead of people’s eyes, which is why he believed the system would transfer. If it worked for him it was not talent, and a system is the one thing that hands over.
The regulator has reached a version of the same reading from the other side. From January 2027 the Council for Estate Agencies runs registrations on three-year cycles requiring three transactions or a refresher exam, and exempts new entrants for their entire first year.
Singapore has written the zero year into the rulebook. Oh’s argument is that the person one year out of it is the asset.
The model broke anyway, and where it broke is the part worth copying.
He was sleeping two or three hours a night across classes, viewings, calls and whoever needed him that evening, on the belief that leading meant never dropping. His chest seized in his early twenties and he was admitted with a suspected heart attack.
What he lost in the ward was the belief that he could be the system. Everything he had built ran only while he was in the room, so the six people depended on his availability rather than on anything that worked without him.
That’s not a team. That’s a bottleneck with people attached to it.
The fix was structural. The default in the industry is solo: one agent runs prospecting, buy side, sell side, paperwork and follow-up, which makes the agent the whole company and therefore the ceiling.
That’s not a team. That’s a bottleneck with people attached to it.
Oh and his mentor split the work into three, one group on buy side, one on selling, one on the larger commercial deals, so nobody had to be good at all three and everybody had to be excellent at one. That split became JNA Salesforce, and Oh dates the brands, the structure and the people to it.
The price arrived immediately, in the only currency an agent counts. Splitting the work splits the commission, and personal revenue dropped.
The second price took longer to show. The rule fails when the selection fails, and Oh’s miss was a person taken in on eagerness who turned out to be keeping a private tally of what was fair; nobody argued, they went quiet, and by the time the conversation happened they had decided to leave.
His revised filter is the most transferable thing in the account, because it costs nothing to run. In the month a deal a candidate touched pays somebody else, Oh watches whether they get smaller or bigger.
The one keeping a ledger of what they are owed leaves the first time the ledger fails to balance, and in a split model it never balances, because a good month is often somebody else’s commission.
Eighteen months ahead is enough to teach the work. Choosing who gets taught takes the longer look, and the tell arrives on the day the money goes to somebody else.1
Footnotes
-
From 2027 the Council for Estate Agencies also collects monthly commission data from agencies, in part so prospective agents can see what the career pays before entering it. Oh sat with CÈ’s AI interviewer on 2 September 2026, by text; quotes lightly normalised for print, wording untouched, transcripts on file. ↩