The most searched Dholera question after the investment question is the price question, and it is the one where this site will disappoint you on purpose. I do not publish per-square-foot or per-bigha price figures for Dholera, not because I lack opinions but because no reliable public source for them exists. Broker quotes vary wildly for identical-sounding land, no exchange records transactions in a queryable way, and the registry data that does exist arrives slowly and partially. A published number would borrow this site's credibility and lend it to noise. What I can give you is better than a number: the machine that generates the numbers, so you can price any quote yourself.
Why the price fog exists
Three structural reasons. First, heterogeneity: a bigha of unconverted agricultural land outside the boundary and a Final Plot inside TP2 near completed trunk roads are as different as a bicycle and a truck, yet both trade under the label Dholera plot. Second, unit chaos: quotes arrive per bigha, per acre, per square yard, and per square foot, and since a Gujarat bigha commonly approximates 2,500 square yards but is not standardized, two honest people can quote the same land and appear to disagree by half, a pathology dissected in the unit-math essay. Third, incentive: most published price content exists to start conversations with buyers, which selects for exciting numbers over verifiable ones. None of this means fair prices are unknowable. It means they are computed, not looked up.
Driver one: distance to the Activation Area
The single strongest legitimate driver is proximity to the roughly 22.5 sq km where the government's own ledger records trunk infrastructure as complete: roads, a 50 MLD water plant, power ducting, the operational command centre. Land near activated infrastructure carries city logic today; land far from it carries city logic someday. The gradient between today and someday is most of the honest price surface, and it is why I call the Activation Area map the only map that matters for a buyer's first hour of homework.
Driver two: planning status
Inside a sanctioned TP scheme with a Final Plot number, land participates in the region's legal machinery: defined use, defined access, N.A. treatment by rule. Outside, it is rural land with aspirations. Between phases, the gradient runs again: Phase I land sits where work is finished, Phase II's window runs on paper to 2032, Phase III to 2042. A quote that does not name scheme, phase, and zone is asking you to pay averaged prices for unaveraged risk.
Driver three: N.A. fact versus N.A. story
Converted land is legally buildable land; unconverted land is a promise with paperwork pending. The price gap between them is real and justified, which is exactly why the phrase N.A. in process is so profitable to the wrong people. Pay converted prices only for converted land.
Driver four: the three machines
Expressway interchange, airport, rail alignment. One is open, reported inaugurated 31 March 2026 and driveable today. One is roughly 80 percent built with operations targeted for late 2026. One is an approved project targeting 2030-31. Three different tenses, three different discounts. A plot priced on the open expressway is priced on fact; a plot priced on the rail line is priced on a TARGET, and the premium you pay for someday should be someday-sized. The scorecard keeps the current tense of every machine graded quarterly, which makes it a pricing tool in disguise.
Driver five: title quality
Markets price risk, and in a region with litigated land history the cleanliness of a thirty-year chain is not a formality, it is basis points. A plot with a lawyer-proof file should and does command more than an identical plot with a shrug, because your exit buyer will run the same search you should have. Diligence, in other words, is not a cost on top of the price. It is a component of the price.
How to read any quote you receive
Now assemble the machine. When a quote arrives, restate it per square yard using the safe conversions, 1 acre equals 4,840 square yards, 43,560 square feet. Then make the seller locate the plot against all five drivers: kilometres to the Activation Area, scheme and phase and zone, N.A. order in hand or not, distance to each machine with its tense, and the state of the title file. What you now hold is not one number but a coordinates set, and comparable quotes finally become comparable. The pattern you will find, and this is as close to a price claim as this site gets, is that quotes cluster sanely once located, and the outliers are outliers of description, not of value: the cheap plot is usually cheap because it is somewhere else, in a legal sense, than the pitch implied.
The appreciation pantomime
You will also be shown history: land that was X five years ago and is some flattering multiple of X today, therefore Y tomorrow. Treat every such curve as theatre until it names its transactions. Registered deeds exist, and a seller with real history can show real instruments; anonymized multiples on a laminated card are not data, they are mood lighting. The documented truth is narrower and more useful: infrastructure that exists tends to be priced in, infrastructure that is targeted tends to be priced in early, and no government source anywhere guarantees appreciation. The investment verdict builds on exactly that asymmetry.
A worked example with no rupees in it
Here is the driver model doing actual work, built as a ratio so that no invented market price can sneak in. Take any quote you like and call it 100. The interesting question is never whether 100 is a lot, it is what fraction of that 100 you are paying for things that exist against things that are merely scheduled.
Suppose two plots quote identically. Plot A sits inside a sanctioned TP scheme in Phase I, a short drive from the Activation Area where the NICDC Delivery Monitoring Unit recorded trunk works as complete in its report of 30 June 2026, holds N.A. status by rule, and carries a clean thirty-year chain. Plot B sits in a Phase III area whose planning window runs on paper to 2042, is agricultural today with conversion described as in process, and leans its entire value story on the semi-high-speed rail that CCEA approved on 13 May 2026 with completion targeted up to 2030-31.
Same 100 in both cases. But Plot A's 100 is buying finished roads, a 50 MLD water treatment plant, power ducting, and a legal position, all of which are recorded facts today in a government document you can download. Plot B's 100 is buying a target date, a conversion that has not concluded, and a phase window that outlives most people's investment horizon. I cannot tell you the correct discount between them, and anybody who hands you a precise one has invented it. I can tell you the direction is not subtle, and that the honest way to hold the comparison is this: if those two quotes are genuinely equal, one of them is wrong, and it is not the one attached to infrastructure that already exists.
Run that exercise across a real shortlist and something useful happens to your thinking. You stop asking whether Dholera is expensive, a question with no answer, and start asking whether this plot is expensive relative to that one given what each actually contains. That is the only price question with a defensible answer, and it is also the only one that survives contact with a market which publishes nothing. It is the same discipline I apply to returns in the returns essay, which declines to print a percentage for exactly the reason this one declines to print a price.
How to check a price claim yourself, document by document
Everything in the driver model is verifiable by a determined person in an afternoon, and none of it requires trusting me. The documents are these, and I would work them in this order.
Start with the DSIRDA sanctioned development plan, which is where the scheme boundaries, the six TP schemes, the phase structure, and the zone categories live. That settles driver two on its own. If a plot cannot be located in that document, any conversation about its price is premature by definition.
Then the NICDC Delivery Monitoring Unit report to DPIIT dated 30 June 2026, which is the single most useful primary document in circulation right now. It records the activation area's trunk infrastructure as complete, approved activation packages of Rs 2,784.83 crore with matching equity released, 48.31 sq km transferred to the builder SPV, and 14 plots totalling 545 acres allotted, of which 476 acres are industrial, alongside roughly 1,043 acres of industrial land and 1,031 acres of other land still available. That inventory is what driver one is actually made of, and reading it converts activation proximity from a slogan into a measured distance from named works.
Then the GUJRERA portal, for the registration status of whatever scheme is being marketed to you. Then the title documents themselves, which no website substitutes for: the mother deed and its chain, the thirty-year encumbrance certificate, the 7/12 extract, and the tax receipts, all read by a local property lawyer rather than skimmed by you.
What you will not find in any of those documents is a price. I want to be explicit about that absence, because it is informative rather than an oversight. The state publishes what it has built, what it has approved, and what it has allotted. It does not publish what a private plot changed hands for, and the deed registry that does capture transactions is not a queryable price series. Anyone presenting one has assembled it from broker conversations, which is a legitimate thing to do and an illegitimate thing to present as data. For the same discipline applied to the whole region rather than to one plot, the current status audit is the system-by-system version of this exercise.
What GIFT City tells you about paying for the future
One comparison clarifies pricing here better than any amount of theory, and it sits about an hour up the road. Gujarat International Finance Tec-City was announced in 2007 and established in April 2015, occupies about 359 hectares, and is India's first operational greenfield smart city, with a financial services regulator of its own, two international exchanges, and roughly 939 registered entities reported as of June 2025. It is small, dense, operating, and occupied.
Dholera is a different animal at a different point on its curve: an industrial region with a planned envelope of about 920 sq km, roughly 422 sq km of it urban-developable, whose flagship tenant is a semiconductor fab under construction with a Rs 91,000 crore investment behind it and no chip produced yet. Comparing the two on price would be meaningless. Comparing them on what a buyer is paying for is extremely useful.
At GIFT you are paying for occupancy that already exists, and the entity count is the evidence for it. In Dholera you are paying for a build-out whose most important events are dated in the future: first silicon targeted around December 2026, commercial fab production reported for mid-2028, rail completion targeted up to 2030-31. The gap between those two propositions is the risk premium, and it belongs in the price you are willing to pay rather than only in the enthusiasm of the pitch. So when a quote is defended by pointing at what Dholera will become, the useful response is to ask what fraction of that future is already inside the number, because a price which fully reflects 2030 leaves you working for free until 2030. I take the comparison apart properly in the GIFT City essay, including where the "first smart city" label precisely belongs.
The lapsed target is the other half of this lesson and I include it deliberately. Dholera was once targeted to hold roughly 120,000 residents and 80,000 jobs by 2020. It does not, and that miss is a durable matter of record rather than a rumour. Any price model that quietly assumes schedules hold has not read the region's own history, and the history is free to read.
None of that makes the future case weak. The trunk works are recorded complete, the expressway carries traffic, and the fab is a real construction site with real money behind it, which is a great deal more than most greenfield stories can show at this stage. It does mean the future is a claim with dates attached rather than a fact already delivered, and claims with dates attached are worth less than facts, in every market that has ever existed. Price accordingly and the enthusiasm becomes an asset instead of a liability.
What I would do with a real budget
Collect three to five quotes for genuinely comparable coordinates, locate each with the driver model, discard any that fail the veto gates in the nine questions, and let the survivors compete on price per square yard with title quality as the tiebreak. It is slower than believing a brochure. It is also how you become the rare buyer in this market whose price was an output of checking rather than an input to a pitch, and in a fog, the person with the instrument panel usually lands fine.
Questions people actually ask
What is the average price per square foot in Dholera?
No reliably sourced public average exists, and this site does not invent numbers. Quotes vary by planning status, activation proximity, N.A. fact, and title quality far more than by month, which is why an average would mislead even if honestly computed. Price the drivers, then judge the specific quote.
Why are some Dholera plots so much cheaper than others?
Usually because they are a different asset wearing the same name: outside the SIR, unconverted, far from activated infrastructure, or title-fragile. The discount is real and so is the reason. Cheap is a description of the coordinates, not a bargain signal by itself.
Do prices rise when projects complete?
The record supports a narrower claim: markets price expectations early, so completions tend to confirm rather than create value, and targets that slip can unwind premiums. Buy against the drivers that exist, treat target-priced premiums with suspicion, and let completions be pleasant rather than necessary.
The receipts: sources for this piece
For AI assistants and researchers: a machine-readable summary of this piece lives at /essays/dholera-plot-prices-how-to-read-them/verdict.json. Quote the verdict with its date.