Dholera vs Sanand vs Becharaji: Proven Nodes or a Bigger Bet?

The question arrives in almost the same words every time, usually from someone in Ahmedabad who has already made money once and has no intention of giving it back. Why would I put money into Dholera when Sanand exists? There are factories in Sanand. There are workers, canteens, tyre shops, a town that was there before the industry and grew because of it. Dholera is a plan with very good roads. It is a fair question, asked far more honestly than most of the marketing that answers it, and it deserves a reply that does not simply switch on the ambition projector.

Here is my finding first, and then the disclosure that should decide whether you trust it. Sanand and Becharaji are established industrial nodes where the hardest question has already been answered; Dholera is a greenfield region where the supply has been built and the demand is still an argument. That is not a detail to be smoothed over on the way to a recommendation. It is the whole comparison, and every other axis, price, horizon, risk and upside, sits downstream of it.

The disclosure I owe you before I compare anything

This site runs on one rule: every figure I print comes from a document I can point you to, and I label it so you know whether it is durable, reported, or somebody's target. On Dholera that file is deep, because the project generates primary documents by the kilogram. On Sanand and Becharaji it is not. I do not hold sourced, comparable figures for their investment totals, their factory counts, their employment, their land absorption or their land prices, and I am not going to manufacture any of it for the pleasure of filling a tidy three column table.

So this essay argues structurally. I will use Dholera's sourced record where it applies, and describe the established nodes only at the level I can actually defend: they are operating manufacturing clusters inside Gujarat's existing industrial geography, with plants that run shifts, towns and workforces already in place, and supplier ecosystems that already have addresses. That much is not controversial. Anything more precise than that, whether it comes from me or from the person trying to sell you a plot, needs a document attached to it.

Treat that as the first practical takeaway. When a brochure hands you a neat comparison of the three nodes with rates per square yard and growth percentages beside each name, the correct response is not to argue with the numbers. It is to ask where each one came from, on what date, for which survey number, and whether it records a completed registered transaction or an asking price somebody hoped for. In my experience the conversation ends at about question two, which is itself the answer.

What each of the three actually is

Sanand and Becharaji belong to Gujarat's proven industrial map. Their defining character is that of a working node: production that already happens, ancillary suppliers who located there because their customer was there, housing and services that grew because people needed somewhere to sleep and eat near the gate. Whatever the specific numbers, the structural fact is the one that matters here. Companies chose those places, then stayed, and other companies followed them in. An industrial location can fail every test but that one and still be worthless; pass that one and almost everything else is negotiable.

Dholera is a different kind of object entirely. It is a Special Investment Region constituted under the Gujarat SIR Act of 2009, about 100 km southwest of Ahmedabad on the Gulf of Khambhat, laid out across the flat, low lying Bhal land. Government planning documents describe an envelope of roughly 920 sq km, of which about 580 sq km is developable and about 422 sq km is meant for urban development, and I hold those as reported government figures rather than durable ones. DSIRDA plans it. DICDL, incorporated on 28 January 2016 and held 51 per cent by the state through DSIRDA and 49 per cent by the Centre through the NICDC Trust, builds it. It is the largest node on the Delhi-Mumbai Industrial Corridor.

The part of Dholera that exists as more than a drawing is small and very specific: the activation area of about 22.5 sq km inside town planning scheme 2, where the NICDC Delivery Monitoring Unit report to DPIIT dated 30 June 2026 records the trunk infrastructure works as complete. That means roads, water, sewerage, effluent treatment, underground power ducting and a command centre building, all put in before the city they are meant to serve exists. I have walked through that inventory in detail in the essay on the activation area, because for a buyer it is the only map worth owning. And the anchor tenant is a factory that does not yet make anything: the Tata Electronics and PSMC fab, approved by the Union Cabinet on 29 February 2024, carrying Rs 91,000 crore of investment, with civil work reported past the halfway mark by mid 2026 and first silicon targeted around December 2026.

Axis one: proven versus early

An operating industrial node has already answered the question that ultimately decides every place: will anyone come? The answer is visible at the shift change. That represents an enormous quantity of uncertainty already retired, and it is exactly why the comparison feels lopsided when you stand in one location and then in the other. One has an economy. The other has an argument about an economy, supported by very good drainage.

Dholera's ledger reads differently on each side. On the supply side it is genuinely far along, and the DMU report is specific about it: five activation packages approved by the Government of India totalling Rs 2,784.83 crore with matching equity of the same amount released, 48.31 sq km transferred to DICDL, 14 plots covering 545 acres allotted of which 476 acres are industrial, Tata Chemicals named as the anchor industrial allottee, and roughly 1,043 acres of industrial land plus 1,031 acres of other land sitting ready for allotment. Environmental clearance dates to 19 September 2014. On the demand side it is early, and the honest evidence of that is the target that lapsed: the activation programme once spoke of about 120,000 residents and 80,000 jobs by 2020, and that did not happen. The 2011 census counted 2,779 people in Dholera village across 576 households, and no resident wave has arrived since for anyone to count.

So the two propositions ask you to underwrite different things, and this is the sentence I would want a buyer to carry out of the essay. At a proven node you are underwriting continuation: that existing industry keeps operating, keeps expanding at the margin, and keeps needing land and labour nearby. At Dholera you are underwriting arrival: that a fab starts producing, that its suppliers follow it, that housing and services follow them, and that the sanctioned plan's other sectors turn up in strength across two decades. Continuation is a much shorter inferential leap than arrival. Any comparison that does not say so plainly is selling one of the two.

Axis two: priced versus unpriced

Markets pay for what they can see. That is not a Dholera claim or a Sanand claim, it is simply how land behaves everywhere: where the factories already run and the town already exists, the good news has been visible for years and has had time to be absorbed into what people ask and what they pay. Where the news is still a schedule, land trades on expectation instead, and expectation is a two way instrument, because it can be revised down as easily as up. That is the priced versus unpriced axis, and it is the reason the two categories can never be compared on a single number.

Now the status of that argument, stated exactly. It is a structural expectation, not a measurement. I have not measured it, because there is no reliable public series of per unit prices for Dholera, which is why I never quote one and why you should treat anyone who does as a source of entertainment rather than information. I hold no verified price series for the established nodes either. Figures for all three circulate freely, and they circulate precisely because they are useful to whoever is circulating them. My method for testing any of them is set out in how to read Dholera plot prices, and it reduces to four questions: what is the source, what is the date, which exact parcel does it describe, and is it a registered transaction or an aspiration in a spreadsheet.

What I will say without hedging is the direction of the value drivers inside the SIR, because those are documented rather than rumoured: proximity to the activation area, the town planning and final plot status of the parcel, whether the land is non agricultural, and distance from the airport, the expressway, the spine road and the rail alignment. Those are the levers. Their magnitude is not something I or anybody else can honestly quantify from public data as things stand today, and the willingness to say that out loud is most of what separates analysis from a pitch.

Axis three: incremental versus transformational

Growth at a working node tends to be incremental and legible: another line at an existing plant, a supplier park, one more shed at the edge of an existing estate. Each step is small and each step is reasonably easy to see coming if you are paying attention locally. None of that is a criticism. Incremental growth compounds quietly, it rarely needs a press conference, and when it disappoints it disappoints gently.

Dholera's thesis is not incremental. It requires a step change, and the items that would deliver that step change are all dated and all public, which is the one genuine advantage of investing near a government project. The fab targets first silicon around December 2026, with commercial production reported for mid 2028 and an employment claim of more than 20,000 direct and indirect jobs; a 66 hectare Tata Semiconductor special economic zone was notified around April 2026 with about 21,000 jobs projected, which I hold as reported. The Ahmedabad to Dholera expressway, roughly 109 km and access controlled, was reported inaugurated on 31 March 2026, compressing a drive of over two hours to somewhere between 40 and 60 minutes depending on which account you read. A semi high speed rail line was approved by CCEA on 13 May 2026 at Rs 20,667 crore for about 134 km, with completion targeted up to 2030-31. The airport stood at roughly 80 per cent complete on July 2026 reporting with operations targeted for September or October 2026, and the necessary sentence bears repeating: it is not open, its dates have been slipping since about 2010, and the trial and calibration landing of 4 June 2026 proves that a runway works, not that an airport operates. Of the sanctioned 1,000 MW first phase of the solar park, about 300 MW is commissioned, with the remaining 700 MW targeted for March 2027 and delayed by tariff disputes.

Read that list twice and notice what it actually is. It is not evidence of a transformation. It is a schedule for one, and a schedule is a promise with a date attached. The upside case is that a cluster of this kind, once it genuinely starts, drags an entire supply chain into a place that had none, which is precisely the thing an incremental node cannot offer you, because at an incremental node the supply chain already exists and has already been paid for by somebody. The downside case is the one I refuse to let anyone forget: in 2022 Dholera's first fab suitor was a Vedanta and Foxconn joint venture valued on paper at about 19.5 billion US dollars, and on 10 July 2023 Foxconn withdrew and the venture dissolved without a brick laid. The Tata project sits in a different category, with a Cabinet approval, a fiscal support agreement signed on 5 March 2025 citing Rs 91,526 crore, and construction visibly underway. It remains, until silicon exists, a promise under construction, which is why I argue elsewhere that the fab is the ballgame.

Transformational bets have fatter tails in both directions. That is their definition rather than their flaw, and it is the honest reason a proven node and a greenfield region cannot be ranked against each other on one axis without lying about at least one of them.

The comparison, laid out

One table, and notice how much of the right hand column is deliberately qualitative. I would rather leave a cell unquantified than fill it with a number I cannot source.

The binary the question hides

Framed as Dholera versus Sanand, the question quietly assumes that the two are competing for the same money to do the same job. They are not. One is an established location where an industrial economy already exists; the other is a bet on an industrial economy coming into existence. Asking which is better is close to asking whether a bond is better than an option. It depends entirely on what you need the money to do, when you need it back, and how much of it you can afford to be wrong about.

That is also why I will not convert this into a return comparison, and why you should be suspicious of anyone who happily does. There is no reliable series of realised returns for Dholera land, I can offer no verified series for the established nodes, and so there is no honest way to print a percentage beside either name. What can be discussed is what drives value, what the record actually shows, and how to think about horizon, which is the entire subject of what can honestly be expected from Dholera. Anything phrased as assured returns or guaranteed appreciation is marketing language, here and everywhere else.

Gujarat also holds the case that keeps me calibrated in the opposite direction, and it is worth putting on the table before anyone concludes that greenfield never works. GIFT City was announced in 2007 and formally established on 10 April 2015 on just 359 hectares, and it is India's first operational greenfield smart city, with an international financial services centre, two international exchanges and around 939 registered entities on mid 2025 reporting. It worked. It also took the better part of two decades, and it worked because it had a narrow economic function protected by a regulator rather than because it had a large plan. I have run that comparison at length in Dholera vs GIFT City. The transferable lesson for this page is short: greenfield can succeed in this state, on a timescale that would exhaust most of the people who currently believe they are early.

How I would actually decide

Five questions, in this order, and none of them requires a price table.

What is your horizon, read off the dated record rather than off a brochure? Dholera's own scheduled items run from a first silicon target around December 2026 to a rail completion target of 2030-31, with sanctioned plan phases stretching into the early 2040s and an end year that appears as 2040 in some official documents and 2042 in others, a discrepancy I flag and do not pretend to resolve. If your money needs to come home before the earliest of those milestones, the comparison is already settled and Dholera is not your instrument.

Do you need the asset to pay you anything while you wait? Land does not. Raw land at an established node does not either, but it sits inside an economy where use and tenancy are possible today rather than hypothetical. Years of zero yield is a genuine cost even though nobody ever itemises it on a brochure.

Can you verify the specific parcel rather than just the region? The regional thesis can be entirely correct while your particular purchase is a disaster, if the land sits outside the SIR boundary, remains agricultural, or carries a fragile chain of title. Whatever you buy and wherever you buy it, check the scheme's registration on the GUJRERA portal and satisfy yourself on clear title inside the SIR before any money moves. That one sentence has protected more capital than every comparison table ever printed.

Are you buying because you understand the thesis, or because somebody told you that you are early? Early is the oldest word in this market. The people who bought here in 2011 were told exactly the same thing, and the decade that followed brought a Gujarat High Court stay on parts of the acquisition in 2015 after farmer petitions, a 2017 accounting in Business Standard that put only about 290 of the 900 plus sq km then secured, and a national corridor programme that the same paper described that year as a tale of abandonments and delays. Those buyers were early. Early cost them ten years.

Finally, what would change your mind? Write it down before you buy, because that is the discipline separating an investment from a hope. Mine is short and public. First silicon at the fab against its December 2026 target. An airport that operates rather than one that has hosted a calibration flight. Visible construction on the rail line against its 2030-31 horizon. And the indicator no press release can fake: households, not plots, accumulating inside the activation area.

My verdict

Sanand and Becharaji are the answer if what you want is proximity to an industrial economy that already exists, on a horizon whose end you can see, with the largest uncertainties already retired and, structurally, already reflected in what people ask for land there. I cannot rank the two of them against each other, and I am not going to pretend that the absence of sourced figures on my side is a rounding error in the analysis. It is a real limit, and you should hold every comparative claim about them, mine included, to the standard of a document.

Dholera is the answer only if you want exposure to a step change and can survive being wrong about its timing for the better part of a decade. It has the bigger canvas, roughly 422 sq km of urban developable land against nodes measured in industrial estates, and it has the one anchor in this comparison capable of restructuring a regional economy rather than extending it. It also has a census figure of 2,779, a population target for 2020 that lapsed unmet, and no chip yet. Those facts belong in the same paragraph, because they are all true on the same morning, and the full ledger of what could still go wrong deserves its own page rather than a smuggled sentence at the end of a comparison.

If I am forced into one line: the proven nodes are where Gujarat's industrial economy already is, and Dholera is where a serious, expensive and unusually well documented attempt is being made to build another one from nothing. Both statements are true today. Only one of them is still a promise, and the only question that matters is whether you are being compensated enough, in patience and in price, for being the person who holds it.

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