IPO Analysis · Mainboard · 20 August 2026

₹650 crore raised. ₹18 crore builds anything.

This is a genuinely good niche industrial business. It is also an offer where, of the ₹650 crore being raised, roughly ₹18 crore actually builds new capacity. Everything else goes to existing shareholders or to paying down debt.

What the company does

Tempsens makes the instruments that measure and control heat in heavy industry — temperature sensors, electrical heating elements, and specialised cables. Its customers are petrochemical plants, metal producers, power stations, defence, nuclear, glass, automotive and pharma.

It has been going since 1990, is run from Udaipur, and operates 15 manufacturing units across India, the UAE, South Korea, Indonesia, Germany and Poland, selling into more than 80 countries.

The headline claim is real: it says it is the largest maker of contact and non-contact temperature sensors in India, with about 10.5% of the sensor segment, and the only Indian manufacturer of non-contact sensors at about 21.3% share.

One qualification worth holding onto. That leadership claim covers the temperature-sensing business — which is now under half of revenue:

The fastest-growing segment — heating, from 4% to 17% of revenue — grew mainly because the company bought a heating business. More on that shortly.

Where the ₹650 crore goes

Read that first row again. 85% of this offer is existing shareholders selling shares. That money does not reach the business at all.

An offer for sale is entirely legitimate — early backers are entitled to exit, and a company with low debt and good margins does not need to raise much. But it changes what you are buying. You are mostly buying shares from someone who is leaving, not funding a company that is expanding.

The plan that shrank between draft and final

Now compare the draft with the final offer:

Between the draft and the final offer, the money earmarked for building things was cut by about half, while the number of shares being sold by existing holders went up.

That is a fact, not an accusation. Companies revise plans for perfectly ordinary reasons — a project gets deferred, costs come in lower, internal cash covers part of it. But it is the single most useful thing to know about this offer, and it is not in any of the summaries circulating today.

The numbers

Revenue has roughly doubled in three years and profit has more than doubled. The margins are good and improving. On the face of it, this is a strong record.

Two things underneath it are worth understanding.

Cash has never kept up with profit. Across FY23 to FY25 the company reported ₹136.70 crore of profit and generated ₹115.38 crore of operating cash — about 84 paise of cash for every rupee of profit, and that is before spending anything on new plant. The gap is working capital: money tied up in stock and in invoices customers have not yet paid.

Receivables are growing faster than sales. Money owed by customers rose from ₹64.24 crore to ₹85.74 crore in FY26 — up 33% — while revenue grew 17.5%. Working capital days went from 152 to 193.

Neither is evidence of anything wrong. In project engineering, bigger orders come with longer payment terms. But it is exactly what to watch in the first two sets of results after listing.

Three years, three different companies

That is why FY25 revenue jumped 37.7%. Part of that growth was bought, not earned.

And it did not stop there:

• FY25 — Marathon Heater merged in
• FY26 — acquired Tempsens Instruments GmbH (Germany) and Tempsens Polska, consolidated from 16 January 2026
• April 2026 — took a majority stake in another measurement business

The prospectus itself says the years are not directly comparable. So the neat rising bar chart everyone is sharing is a chart of three differently-shaped companies. That is ordinary corporate activity, not wrongdoing — but it means the growth rate tells you much less about the future than it appears to.

The company reports two return figures

Why the gap? The merger created two large accounting entries:

Goodwill is the premium paid above the value of what was actually acquired. It sits on the balance sheet as an asset. Together with the intangibles, roughly ₹156 crore — about a third of FY25 net worth — is acquisition accounting rather than accumulated profit.

The adjusted figure removes that goodwill from the equity base and adds back the amortisation. Doing so is disclosed, permitted, and arguably shows the operating business more clearly.

But it is the central question of this IPO. At ₹300 you pay roughly 4.9 times book value. If the true return on that equity is 13.5%, that is expensive. If it is 23%, it is defensible. The reported accounts say the first number. The price assumes the second.

One more thing about goodwill: it is tested for impairment every year. If the acquired business underperforms, book value can fall without a single rupee leaving the company.

The anchor book — the strongest signal here

This is the part of the offer I rate most highly, and it deserves to be said plainly.

Anchor investors are the large institutions that buy a day before an IPO opens. When domestic mutual funds take well over half the anchor book, professional long-term money has examined the price and accepted it.

The largest allocations, each at 8.55%: Aranda Investments (Singapore), SBI Resurgent India Opportunities, Nippon India Small Cap, Goldman Sachs India Equity, Ashoka WhiteOak Emerging Markets. Then Aditya Birla Sun Life Small Cap, Kotak Infrastructure & Economic Reform, HDFC across three schemes, and Axis Max Life and Kotak Life on the insurance side.

For contrast: the Sunshine Pictures IPO I wrote about yesterday had nil allocation to mutual funds and nil to insurers. Tempsens has 57.39% and 7.20%. These are not remotely the same kind of book, and anyone treating all IPOs as interchangeable momentum trades is missing the difference.

Two honest counterweights. A long-short fund and a “recently listed IPO” fund also appear in the book — those are not conviction holders, they are mandated participants in listing cycles. And anchor shares unlock in two tranches, roughly 30 and 90 days after allotment, which means supply arriving around late September and late November.

GMP, and why the spread matters

On 19 August, four trackers quoted four different premiums for the same share on the same day:

A 37% spread between the highest and lowest quote, on the same day, for the same share. That dispersion is itself the argument for not treating GMP as information.

It has also moved violently: around ₹85 on 15 August, ₹65 on the 16th, then ₹154, ₹172 and ₹220 as the price band and anchor news landed. Roughly tripled in three days.

And here is the part worth sitting with. At a listing price of ₹460 to ₹520, this company would be valued at roughly ₹3,850 to ₹4,360 crore — about 54 to 61 times earnings, for a business earning 13.5% on its equity. Whatever you think of the company, that is a demanding price.

What is genuinely good

• Real niche leadership. Largest contact and non-contact temperature sensor maker in India; the only Indian manufacturer of non-contact sensors.
• Backward integration. The Udaipur operation runs alloy melting, wire drawing, fabrication, assembly and calibration in-house — genuine control over cost and lead time.
• Certifications are the real moat. ATEX, IECEx, UL, BIS, CE, PESO and a NABL-accredited calibration centre. In process industries these take years to obtain and are what keeps competitors out.
• Research with output. 83 R&D staff, 12 Indian patents, 39 overseas trademarks; fibre-optic sensors, aerospace-grade cables, catalyst bed heaters for space applications.
• Customer concentration is falling — the top ten customers dropped from 24.74% of revenue in FY24 to 18.59% in FY26, across 3,800+ customers.
• A third of revenue is repeat maintenance work (32.45% in FY26), which is steadier than one-off project orders.
• Low debt. Debt to equity of about 0.15 and debt to EBITDA of 0.74.
• Exports compounding — from 21.6% of revenue in FY23 to 26.5% in FY25, with EU manufacturing now owned.

What worries me

1. 85% of the offer is shareholders selling, and only about 2.8% builds new capacity.
2. The capex plan was cut roughly in half between the draft and the final offer, while the shares on sale went up.
3. Three consecutive non-comparable years. The prospectus says so itself.
4. Return on equity has halved, from 21.2% to 13.5%, and the flattering alternative figure is the adjusted one.
5. ₹156 crore of net worth is acquisition accounting, subject to annual impairment testing.
6. Cash has never once matched profit, and receivables are growing at roughly twice the rate of revenue.
7. No comparable listed company. The prospectus states it plainly — so there is no peer multiple to sanity-check ₹300 against.
8. Cyclical exposure. Metals and petrochemicals are about 41% of revenue; project and OEM work about 68%. Both follow customer capital spending.
9. Concentration of production. Ten of the fifteen manufacturing units are in Udaipur.

Common questions

Verification ledger

What to watch after listing

1. Receivable and inventory days in the first two quarterly results. If working capital keeps stretching, the cash gap widens.
2. Cash from operations against profit. It has never exceeded 0.91x. Crossing 1.0x would settle the biggest open question.
3. Any goodwill impairment note. That would tell you the acquisition is not delivering.
4. The anchor unlock dates, roughly late September and late November.

My position — a disclosure, not a recommendation

What persuaded me was the anchor book. When 57% of it goes to domestic mutual funds across 14 schemes, and insurers take another 7%, professional long-term money has done its own work on this price. That is a genuinely different situation from an issue carried by retail momentum alone, and I do not think it should be waved away.

Underneath that, the business is real: a defensible niche, certifications that take years to earn, backward integration, falling customer concentration and a third of revenue from repeat maintenance work.

But I want to be straight about what I am accepting to hold it. I am paying about 4.9 times book for a business whose reported return on equity has halved to 13.5%. I am accepting that 85% of my money goes to a departing shareholder rather than into the company. I am accepting a growth record across three years that the prospectus itself says are not comparable. And I am accepting that cash has never once caught up with profit.

Those are real objections, they are all on this page, and none of them is resolved by my having applied.

What would tell me I was wrong: receivable days continuing to stretch in the first two results after listing; cash conversion staying below profit; a goodwill impairment; or the reported return on equity failing to recover towards the adjusted figure the company points to.

I am not registered with SEBI as an Investment Adviser or a Research Analyst. This is research and education published in public, not investment advice and not a recommendation to buy or sell. Where I say what I am doing myself, that is a disclosure of my own position, not a suggestion about yours. Read the prospectus and speak to a SEBI-registered adviser before investing.

More IPO analysis

• Sunshine Pictures IPO — a 54% profit margin in a business that doesn’t have them
• Horizon Industrial Parks IPO — Blackstone priced it at ₹59.81 in December
• Lalithaa Jewellery IPO — profit tripled, and it sold 0.74% more gold
• Behari Lal Engineering IPO — profit up 22%, volume down 1%, cash down 55%
• Shiprocket IPO — a profitable business funding a loss-making one
• Dhoot Transmission IPO — a 41% market share, priced at a 35% discount

Not SEBI registered. This is not investment advice and is not a recommendation to buy or sell any security. Figures are drawn from the Draft Red Herring Prospectus of Tempsens Instruments (India) Limited dated 29 September 2025 and the anchor intimation letter dated 19 August 2026, together with final offer figures from exchange filings and market sources which are identified above as not verified here. Price band, lot size, grey market premium and FY26 figures are from market sources. Please read the prospectus and do your own research before investing.

A listing scorecard for this call will be published after listing, whatever the outcome.