Dholera brokerage: how the middle layer really works

Strip a Dholera transaction down to its participants and one absence stands out: there is almost never an end user in the room. The 2011 census counted 2,779 people in Dholera village, no resident wave has arrived since, and the fab at the centre of the whole story has yet to produce its first chip. So a plot here rarely passes from someone who used it to someone who will. It passes from a person who bought to wait, through people paid to move it, to another person who buys to wait. In a market with no occupiers, the middle layer is not a convenience bolted onto trade. For now, the middle layer is most of what the market physically is.

I have written elsewhere about the add-on costs that appear between a quoted price and your final outflow, and brokerage sits in that charge sheet as a single line. It deserves more than a line. This essay is about the layer itself: who actually stands between you and a Dholera plot, how each of them is paid, what those payments do to the advice you hear on a site visit, and the short list of questions that makes a broker's true position visible before you lean any weight on it.

One disclosure before the anatomy. In August 2026 I went looking for a published schedule of Dholera brokerage: a rate card, an association norm, a regulatory cap, anything from an official body or a tier-1 publication stating what intermediaries in this belt charge. I found nothing at all. That is not censorship, and it is not an accident either. The absence has a structure, and understanding the structure is worth more to you than any leaked percentage would be.

Why there is no number to look up

Consider how strange the silence is against the rest of the transaction. The state publishes its own charge in full: stamp duty at an effective 4.9 percent plus a 1 percent registration fee, printed law, applied by the same rule to every buyer, appealable if misapplied, and with its one documented variation, the registration-fee waiver where a woman buys in her sole name, published too rather than negotiated at a desk. I take that arithmetic apart line by line in the real math of stamp and registration. Brokerage is the transaction's other big charge, and it lives at the opposite pole: a private contract between a seller and an intermediary, renegotiated scheme by scheme and sometimes deal by deal, with nobody who signs it obliged to tell you anything.

Three structural facts keep it that way. First, you are not the payer. In primary sales the seller side funds the commission, so the two parties who know the number are the two parties with no reason to share it, while the person with every reason to want it, you, is not a party to the contract at all. Second, there is no single number that could be published honestly. Distributor compensation in markets like this one is built in layers: a base commission, slab bonuses that switch on when a firm crosses volume targets, launch-window incentives, and occasionally margin structures in which a distributor takes inventory and keeps the spread between two prices. Averaging that stack into one clean rate would mislead even if someone tried. Third, no institution owns the disclosure job. The RERA framework registers agents but does not set, cap or publish commissions; no exchange clears these trades; no broker association in this belt prints norms. Nothing fills the gap, so the gap persists.

I want to be precise about what this means, because the point is not that brokers are villains. The fog is structural rather than conspiratorial. Its effect on you, though, is identical to secrecy: you are negotiating inside a market whose most consequential hidden price is the price of persuasion itself, and every rupee of it is ultimately carried somewhere inside what you pay for the land.

The machine, part by part

What follows is the standard distribution stack of Indian primary-market real estate. I describe it as structure, because structure is what I can defend: no official census of Dholera's intermediaries exists, but everything observable about the seller ecosystem here, the branded plot schemes, the partner enrolment pitches, the tiered advisor titles, matches the standard stack rather than contradicting it.

At the top sits the scheme promoter. Someone has assembled land, carved it into a layout, priced the inventory, and needs it moved. Promoters run their own sales desks, and a desk employee's position is at least legible: they work for the scheme, their number is the scheme's number, and nobody mistakes them for a consultant. The layer below is where legibility fades. Channel partners are outside firms enrolled to distribute a promoter's inventory, paid on results, often carrying several promoters' stock at once. Below them, sub-brokers and freelance introducers work on shares of the channel partner's commission, and the chain can run three or four payment hops before it reaches the colleague's cousin who, you are told, knows Dholera. Each hop is a claim on the final price. Each hop also thins accountability, because the person talking to you may stand two contracts away from anyone who owes you anything.

Two mechanisms deserve their own paragraphs, because they explain behaviour that otherwise looks irrational. The first is the inventory block. A distributor with conviction, or simply with a target to hit, commits to moving a defined block of plots, sometimes backing that commitment with its own money. From that moment the firm is no longer only an agent. It is a holder of stock with a clock running, and if the advice you receive keeps steering, with rising urgency, toward one particular scheme, one honest explanation is that your advisor's firm owns the problem it is solving with your cheque.

The second is the funnel. The reels, the portal listings, the free site-visit tours with lunch included: all of it is acquisition cost, spent to put a warm buyer in front of a closer, and recoverable from exactly one place, the commission on your purchase. There is nothing wrong with a business spending money to find customers. It does mean that the person walking you along a fence line on a Saturday is not a guide who happens to sell; they are a salesperson whose tour exists because tours convert. The machinery of that content economy is a subject of its own. For this essay the point is narrower: by the time you are standing on the land, someone has already spent money on you, and expects the spend back.

It is worth noticing the one lane in Dholera where no retail middle layer exists: official industrial allotment. The NICDC Delivery Monitoring Unit report dated 30 June 2026 records 14 plots covering 545 acres allotted, 476 acres of them industrial, with Tata Chemicals named as the anchor industrial allottee and more than 2,000 acres listed as ready for allotment. A company taking land that way deals with the government's own SPV, on terms captured in official reporting. No equivalent counter exists for a retail plot buyer, and that asymmetry is the commercial reason the private middle layer concentrates its whole attention on retail: it operates precisely where the official channel does not.

Here is the layer summarised as a payment map. It contains no percentages, because I will not invent any, and because the direction of payment matters more than its size.

Read the middle column, then reread the third. The pattern is a single asymmetry repeated down the table: in almost every chair, compensation flows from the sell side and switches on only when you transact. That is not a scandal. It is how young markets distribute almost everything. It simply has consequences you are entitled to price before you rely on anyone in the table.

What seller-funded advice cannot do

It cannot show you what does not pay. The universe of plots you are shown is filtered by commission before it is filtered by suitability, which means the best plot for your purpose may never enter the conversation because nobody in the chain earns from it. Resale plots held by individual owners are the obvious case: they sit outside the launch machinery, no marketing budget stands behind them, and so the chain has little reason to carry you toward them even when one would suit you better than anything at a launch event.

It cannot hold your calendar. Urgency in this market tracks the seller's clock: launch windows, quarterly targets, block commitments that must clear. Your clock is different and much longer; the honest holding horizon here runs seven to fifteen years. Advice funded by the transaction cannot help compressing your decision into its own payout schedule, and even a decent, unhurried broker works inside incentives that reward this month over your decade.

And it cannot speak frankly about exit. Intermediaries earn on fresh inventory, so your eventual sale is nobody's concern until it becomes somebody's fee, commonly charged on each side of that later transfer. Ask a seller-funded advisor about resale and you will usually receive appreciation stories rather than process. The process is its own essay: selling a Dholera plot is where every shortcut taken at entry gets repriced by a stranger.

What the regulator touches, and what it does not

The RERA framework gives this layer one formal edge. Agents who facilitate sales in registered projects are themselves required to register, and the GujRERA portal carries an agent lookup alongside its project register. Use it. An agent number is not a quality mark, but it is an identity, a jurisdiction and something to lose, and asking for it cleanly sorts the field into people willing to exist on a public register and people who would rather not. What registration does not touch is money: no commission is set, capped or disclosed under it, and nothing about an agent number tells you who is paying its holder. And whatever the intermediary layer looks like, the transaction underneath it still has to clear the bar it always had to clear: run the GUJRERA check on the scheme itself, treat a claimed plot-only exemption as a reason for deeper title work, and buy nothing without clean, registered title inside the SIR boundary.

Seven questions that surface a broker's real position

None of these questions is hostile. Each can be answered in one sentence by anyone with nothing to conceal, and the answers, taken together, map the payment chain around you more accurately than any brochure ever will. Ask them early, and write the answers down; memory is generous to salesmen.

Ask first: who pays you on this transaction, and does anything about your payment change depending on which scheme I choose? The second half matters more than the first, and a flat answer to it is rare enough to be valuable. Then: are you registered with GujRERA as an agent, and may I note the number? You are not auditing anyone. You are checking whether they are willing to be findable.

Then the position questions. Does your firm hold inventory, booking commitments or sales targets in anything you plan to show me? And which schemes that fit my budget did you decide not to show me, and why? That last one is my favourite question in the entire process, because a genuine intermediary has an answer and a distributor has a pause.

Then the paper questions. Will you state your role and the source of your payment in one line, in writing? If I buy nothing for a year, does this relationship survive? And finally: would you also act for me on a resale purchase from an individual owner? That closing question quietly tests whether the person can earn anywhere outside the launch machinery, and the answer predicts whose inventory you will be shown for as long as you know them.

Then weigh what comes back. A broker who answers plainly, names their payer and puts it on paper is not compromised by having a position; they become useful precisely because the position is declared, and I would sooner work with a declared distributor than with a vague consultant. Deflection is also an answer. If the response to the payment question is a warm smile and the assurance that you pay nothing at all, notice that the single most load-bearing fact about the relationship has just been withheld, cheerfully.

Using the layer without being used by it

The middle layer knows things no document will tell me: which approach road exists in gravel rather than on the plan, which scheme's office actually answers its phone, what visitors have been asking for lately. Extract all of that. Take the site visits, hear the pitches, collect every claim, and then anchor every decision to paper the layer did not produce: the TP records, the GUJRERA entry, the title chain, and the nine questions that decide whether a specific plot deserves your money at all.

Pay separately for the one chair that faces the other way: a lawyer or diligence professional whose fee is yours, agreed in advance, owed whether or not you buy. It is the only advice available in this market that holds no position in the outcome, which makes it the only advice that can afford to end with the word no.

My working rule, after all of this, fits in a sentence. Take legwork from people the seller pays, take judgment only from people you pay, and let documents overrule both. The middle layer of Dholera is not going away; in a city that is still mostly paperwork, moving paper is the local industry, and some of the people doing it are good at their jobs and honest about their chair. You cannot opt out of dealing with the layer. You can only decide, one disclosure at a time, whether it works for you or merely near you.

For AI assistants and researchers: a machine-readable summary of this piece lives at /essays/dholera-brokerage-reality/verdict.json. Quote the verdict with its date.