March 2026 was the best sales month many developers had seen in years. Across the seven largest markets, about 5,400 units were sold. BIK reported a 80.5% year-on-year rise in the value of mortgage enquiries and more than 63,000 applicants — the most since July 2008, that is, before the collapse of Lehman Brothers.
And that is exactly why this is the most dangerous moment to decide that the sales problem has been solved.
Because that record isn’t the start of a healthy boom. It’s pull-forward demand — a panicked acceleration of buying decisions ahead of worsening financing conditions. Clients and mortgage advisers rushed to file applications on the old terms, and they were right: the median fixed-rate mortgage offer jumped from 5.78% in early March to 6.5% in April. In a single month.
That one fact changes everything about how you should read your own sales funnel. Because if the cost of money rises faster than your clients’ ability to close the purchase mid-process, then your record reservation result masks a risk you can’t see in any form report.
That is the thesis of this text, and it’s uncomfortable: in 2026 a reservation stopped meaning a sale. And most developers’ marketing-and-sales systems are still built as if it did.
Why the classic “lead funnel” stopped being enough
The standard funnel picture used by most agencies and marketing departments ends at the form:
traffic → click → form → lead → meeting → reservation
In this picture the reservation is the finish line. Marketing delivered the enquiry, sales closed the reservation, the report shows the result, the board is happy.
The problem is that in development the money doesn’t appear at the reservation. It appears at the notarial deed — and between the two lies the longest and most fragile stretch of the whole process, which the classic funnel doesn’t see at all:
reservation → mortgage decision → development contract → deed
This is the stretch where the real drama unfolds in 2026. The client reserved an apartment with a mortgage offer of 5.78%. Before they got the bank’s decision, the market moved them to 6.5%. Their borrowing capacity fell. The instalment rose. The reservation that sales counted as a result quietly turns into a withdrawal — while in the board’s forecast it’s entered as an almost-sale.
This phenomenon has a name: revenue leakage. And it’s more dangerous than the classic “not enough leads” problem, because it’s invisible. A missing lead shows up in the report. A leaking reservation shows up only in cashflow, a quarter later, when it’s too late to react.
Seven places where the result leaks
Below are seven leak points, ordered along the full revenue path — from the first click to the deed. For each I give the symptom (how to tell you have the problem) and the test (how to measure it yourself, usually in half an hour and with no new tools).
I deliberately don’t give ready repair instructions here. The reason is simple: fixing each of these points requires decisions calibrated to a specific project, team and CRM — and that’s work, not a checklist. The point of this text is that after reading it you know where and how much is leaking. That’s the first step of any serious audit.
1. Traffic without intent
Symptom: the number of leads grows, but the share that progress to a meeting falls. Marketing boasts about CPL, sales complains about “quality”.
Test: take the last 100 leads from one channel. Count how many reached a meeting. If the channel with the lowest CPL has the lowest conversion to a meeting — you’re paying for traffic, not for sales.
This is the foundation of everything else. The overriding metric in development isn’t cost per lead (CPL), but the cost of a qualified lead — one with intent and capacity. A cheap lead without intent isn’t a cheaper asset. It’s a cost, because it eats the time of a salesperson who could be talking to someone ready to buy.
2. The ad doesn’t match the page
Symptom: high bounce rate from campaigns, short time on page, leads “don’t know what they clicked on”.
Test: click your own ad like a client. Is the promise from the ad the first thing you see on the page? If the ad talks about price and the page about “prestige” — you’ve broken continuity and you’re losing part of the traffic you paid for.
3. The page doesn’t address the client’s risks
Symptom: there’s traffic, few forms. Page conversion below the market sanity check (the median landing page is about 4.2%; good pages exceed 11%).
Test: show the page to someone outside the industry and give them 30 seconds. Let them say what the investment is, where, for whom, at what price and what they get. If they can’t — the page doesn’t reduce the client’s uncertainty, it leaves it with them.
Buying an apartment is one of the most expensive and risky decisions in a client’s life, often made against something that doesn’t yet exist. A page that doesn’t show construction status, schedule, documents and price logic asks for contact before it gave a reason to trust.
4. The form collects contact but doesn’t qualify
Symptom: sales receive leads with no context — there’s no knowing whether the client has capacity, cash for a deposit, or any real intent.
Test: check how much qualifying information you have on the last 20 leads. If you only know “name, phone, email” — your team only qualifies over the phone, wasting the first, most valuable contact on establishing basics.
5. The CRM doesn’t measure response time or statuses
Symptom: you can’t answer the question “what is our median time to first contact” without counting by hand.
Test: send an enquiry from your own site as a mystery shopper. Measure how many minutes (or hours) to a call back. Then check in the CRM whether that time is recorded anywhere.
I do this regularly — I pose as a client and send an enquiry to an investment to see what happens on the other side. From one I got a reply only after four days: I sent the enquiry on Wednesday, they got back to me on Sunday, by email instead of phone, and straight away with a clean offer — even though in the enquiry I hadn’t even said which unit interested me. At another developer, a large and recognisable one, where I’d expect a completely different standard, no one responded at all. At all.
These aren’t exceptions. Response time is critical in this process, and most developers react in hours or days, not minutes — and, worse, they’re convinced they do it “right away”, because no one measures it. A CRM without measurement of response time, statuses and loss reasons isn’t a sales system. It’s a database in which leads die in silence.
And often there isn’t even that. Walking into a company for the first time, I may just as well find no CRM at all, an Excel not updated for months, or a database with no recorded lead sources and no history of phone contacts — that is, no knowledge of where the client came from and what they were told.
6. Reservations are treated as sales
Symptom: the board’s forecast rests on gross reservations. “Soft” reservations sit in the same bucket as development contracts.
Test: take the reservations from the last quarter and count how many reached a deed. If you don’t separate soft reservations from development contracts, your forecast is a fiction that lulls the board to sleep.
This is the point that hurts most in 2026. A reservation at a mortgage rate of 5.78% and a reservation that has to pass through a mortgage decision at 6.5% are two completely different assets. Counting them the same isn’t optimism. It’s blindness to operational risk.
7. No analysis of loss reasons between reservation, mortgage and deed
Symptom: when a reservation falls apart, no one records why. “The client changed their mind” closes the topic.
Test: check the last 10 lost reservations. Is there a loss reason on each one — mortgage refusal, drop in capacity, change of mind, competition? If not, you’re flying blind on the most important stretch.
Without loss categories you don’t know whether you’re losing clients through the bank, through the product, or through the competition. Each of these causes requires a different response — and you don’t know which one to switch on.
The most expensive leak isn’t in the system. It’s in the salesperson’s head.
All seven points above can be measured and put into a number. But there’s one more leak, invisible in any metric, that I’ve seen with my own eyes in developers’ CRMs — and it’s more expensive than anyone supposes.
It’s the notes salespeople leave on leads.
“Old guy, poor, probably won’t buy.” “No idea what this Ukrainian woman is, I don’t get it.”
I quote it verbatim, because that is exactly what gets written into the system — and because as long as it sounds abstract, no one sees the problem. And the problem isn’t a matter of manners. It’s financial.
That “old guy, poor” is very often a parent or grandparent buying an apartment for their child in cash — no mortgage, no bank decision, that is, the safest transaction there is. That “Ukrainian woman the salesperson doesn’t understand” is just as often an entrepreneur with capital whom no one took seriously, because she didn’t fit the image of a client. In both cases the salesperson, judging the client by appearance or accent and giving up, made a pricing decision on the board’s behalf — and got it wrong, because they had no idea who had the capacity and the cash.
This is revenue leakage in its purest form. Not through lack of budget, not through a weak product — through a prejudice that turns a ready client into a lost sale before anyone has a chance to qualify them. And unlike the other seven points, this one leaves no trace in any report. The money simply walks out the door, and no one knows it ever stood near it.
That’s why a funnel audit that makes sense doesn’t end at the system’s data. It reaches into what salespeople actually think and write about clients — because there, in a single dismissive note, there can sit more lost margin than in an entire badly set up advertising channel.
Methodology: how a developer’s funnel is properly audited
So that this text isn’t just another list of “mistakes”, I’ll show the principle by which it’s done seriously. A funnel audit that makes sense rests on three rules.
Rule one: we measure the whole revenue path, not its convenient beginning. A funnel that ends at the reservation exists to look nice in a report. A funnel meant to protect cashflow has to reach the notarial deed. That means the unit of analysis isn’t the “lead”, but the transition: lead→contact, contact→meeting, meeting→reservation, reservation→mortgage decision, mortgage→deed. We measure the leak on each transition separately, because each has a different cause and a different lever.
Rule two: every observation must be tied to a consequence in money. A “low show-rate” isn’t a problem — it’s a symptom. The problem is: how many meetings × what unit value × what meeting-to-reservation conversion = how much revenue empty chairs burn per month. An audit without conversion into money is an opinion. An audit with it is a board-level decision.
Rule three: we don’t fake a precision the data doesn’t give. This is where most “data-driven” analyses in development discredit themselves. If you have 18 reservations in a quarter, you won’t build a predictive model on that — and saying “conversion will rise by 23.4%” on such a sample isn’t statistics, it’s theatre. A serious audit distinguishes where the numbers are enough to base a decision on, and where you have to write honestly “too little data, let’s gather it first”. False precision is more dangerous than admitting you don’t know, because it leads the board to bad decisions with a sense of certainty.
What to do about it
If you’re reading this as a CEO or sales director and recognise three or more of these symptoms in your own operation — you have a revenue leak you can’t see in the current report, because the report looks at leads and reservations, while the leak happens further on.
The good news is this: on most of these transitions you don’t need to increase the budget to recover the result. You need to remove friction. The same traffic, better handled, with faster response and harder qualification, delivers a result that others would spend the same amount again on media to get. But the condition is to first know where and how much is leaking — and you can’t do that from the campaign level. You have to see it from the level of the whole revenue system.
In 2026 developer marketing stops being a department for reach and leads. It becomes a system for controlling revenue quality. The winner isn’t the one who shouts loudest about records, but the one with hard control over the transition from the first click to the notarial deed.
If you want to see where this leak is in your own operation, start with two minutes: take the short funnel self-check — seven questions, a result straight away, you send nothing. It will show you on which stages you are most likely losing the result. And if you see what I suspect you’ll see — let’s talk about a full diagnosis. I don’t start by increasing the budget. I start by counting how much you’re losing today — and where.