Shakti Polytarp Limited

Why:

• Governance: the year the whole growth story is measured from was signed off by an auditor who held no quality-review certificate, nobody has re-audited it, and it is the only one of the three years that had to be corrected. Revenue went from ₹62 crore in the year to March 2024 to ₹216 crore in the year to March 2026, 3.48 times, and every growth number in this offer is struck off that first year (T1). The firm that audited it, a one-man practice in Indore, signed the accounts on 15 July 2024 and resigned 19 days later, giving "preoccupation in other assignments" as the reason (T1, RHP p.78). The Institute of Chartered Accountants publishes the list of firms that hold a peer review certificate, which is the certificate a company going public needs its auditor to have, and this firm is not on it, not today and not in May 2024, two months before it signed (T2, two vintages of the institute's own list). The firm that replaced it states in this prospectus that it carried out no audit of that year and has relied on its predecessor's report (T1, examination report para 5). When the accounts were restated for the offer, the corrections fell entirely in that one year: a staff gratuity liability that had never been provided for, and tax booked in the wrong years, ₹10.80 lakh in all, about an eighth of that year's audited profit, with the two later years carrying nothing of the kind (T1, Note 34). The balance has to be stated, because it is real. The change was an upgrade, not a downgrade: the incoming firm has three partners, its own quality-review certificate dated 12 September 2024, and nothing against it on any public register (T2). The opinions are clean on both sides. The corrections made the old year look slightly better, not worse. And the old firm broke no rule by not holding a certificate, because a small firm auditing a private company did not need one then. There is an innocent explanation for the whole sequence, that an IPO needs an auditor with that certificate and the old one did not have it, and the filing never gives it, carries no risk factor about the change among fifty-nine risk factors, and prints the new auditor's certificate number while saying nothing about the old auditor having none. Separately, and unresolved: two weeks after the resignation, three people sharing the resigning auditor's surname subscribed to 5,72,000 shares at ₹35 in a rights issue whose entitlement both promoters renounced in full, ₹200.20 lakh in all, and a 1-for-1 bonus four months later doubled that to 11,44,000 shares, 9.10% of the company, making them the third, fourth and fifth largest shareholders (T1, RHP p.89). The resigning firm's sole proprietor is a chartered accountant of the same name, at the same Indore building the filing prints for the firm (T2, institute register). Whether he is the shareholder cannot be established from any public record, because the institute's member index is unreachable, company registry documents that would settle it are paid, and no other route exists. The filing never addresses the connection. We are not saying they are the same person and we cannot say they are not.
• Edge: the margin that makes this business look good is a bulk discount the refineries hand out, they decide it month by month, they can stop it, and part of it is already leaking back to customers as price. The factory side's operating margin went from 6.92% to 15.98% in two years (T1, derived from the filing's segment tables), and the filing attributes the cost improvement in both halves of the business to volume-based discounts and credit notes from Reliance Industries and Hindustan Petroleum earned on the quantity of granules bought (T1, RHP pp.244, 248, 254 to 255). A blended discount rising from nothing to roughly 5% to 6% of granule purchases accounts for the whole of the gain, and 6.7 of the 8.6 points arrived in a single year (T1, derived). To qualify, the company buys far more granules than it needs and resells the surplus: ₹104 crore of granule resale last year at 1.39% net, 48.2% of revenue for 6.7% of gross contribution (T1, derived). The filing's own risk factor says these schemes are "subject to the commercial policies of suppliers, minimum offtake commitments, market conditions, and continued eligibility", and that any modification, discontinuation or non-renewal may reduce the cost advantage and limit the ability to trade granules profitably (T1, RHP p.23). The refiners publish these schemes and declare the rate month by month at their own discretion, so there is no dealership to lose and equally nothing owned: any buyer big enough to fund the tonnage can join (T3). It is not a windfall from cheap resin, which is the one alternative explanation that would have been worse: Indian granule prices moved within about 5% across the year the gain arrived, with polypropylene up, and the nearest listed peer gained margin in the opposite order of years (T3). And it is already being handed back. A line called "Discount on Manufacturing Sales" appears for the first time last year at ₹102.22 lakh, 4.8% of manufacturing gross profit, beside a management commentary that explains a 43.59% fall in receivables by discounts offered to get paid faster (T1, Note 26 and RHP p.251).
• Price: the IPO buys capacity at 1.55 to 3.05 times per tonne what this company itself paid three months earlier, and three-quarters of it goes into a product whose only measured demand channel is an order of magnitude smaller than the capacity. The single object is ₹20.88 crore of plant and machinery (T1, RHP p.99). ₹7.86 crore of that has already been spent, on machines an independent chartered engineer certifies were installed and producing by March 2026, and it bought 3,000 tonnes a year of capacity, about ₹0.26 lakh for each annual tonne (T1, RHP pp.103, 179). The forward spend, with the ₹4.90 crore printing machine that adds no tonnage stripped out, buys 2,000 tonnes a year at between ₹0.41 lakh and ₹0.80 lakh a tonne, the range being the gap between the two readings of the filing's own deployment schedule, which does not reconcile with itself. That is 1.55 to 3.05 times its own cost of one quarter earlier, for the same product family at the same site, and the quotation is 13.7% above what it paid for the identical machines (T1, derived from RHP pp.99 to 107 and 179). Outside benchmarks put the gap wider, 3.1 to 5.1 times, against a spread of only 1.6 times between the benchmarks themselves, although they cover raffia and tarpaulin rather than shade net and no per-tonne shade-net benchmark appears to exist publicly (T3). Of the machinery money, 76.5% goes to shade net, which will be 43.6% of capacity after the issue and has never been reported as revenue anywhere in this filing; the one revenue line that contains it has fallen from ₹13.64 crore to ₹11.81 crore in two years (T1, RHP p.23). The only demand driver the filing's own logic leans on, government subsidised shade-net houses, works out on the National Horticulture Board's published cost norm to roughly 600 to 2,300 tonnes of netting a year for the whole of India, against 6,500 tonnes of shade-net capacity at this one company after the issue, off a build rate that is slowing (T1 for the norm, T2 for the area series). We are not saying the price is inflated, and we could not find the benchmark that would settle it. We are saying the filing never explains the gap.

Valuation at the band

At the cap the offer asks about 10 times last year's profit, against 43.93 times for Commercial Syn Bags of Indore, the one genuinely comparable listed name in the filing's own peer table, and 24.26 times for the other peer it chose (T1, RHP p.115); the same chapter's separate industry price-to-earnings box prints 84.67 to 110.22 and ties to neither of them. The price does not move the verdict: the six ratings judge the business and the shape of the offer, not what is being asked for it.

The story

A buyer here is buying a single-plant tarpaulin maker in Madhya Pradesh wrapped in a plastic-granule dealership. The factory is the part that grew, up 68.8% last year on half again as much certified output at better margins, but it is only 51.8% of revenue; the other 48.2% is buying granules from the refineries and reselling them at 1.39%, a book that exists to hit the purchase volumes that earn the discount which is itself the factory's whole margin story.

What this business is

Shakti Polytarp melts plastic granules, draws them into tape, weaves the tape into fabric and laminates, cuts, stitches and prints it into tarpaulin sheets. Tarpaulin is the waterproof sheeting used to cover trucks, grain heaps, building sites and ponds, and it is 89.4% of what the factory sells (T1, RHP p.23). Everything is made at one leased plant at Nirmani in Khargone district, Madhya Pradesh, and 91.77% of sales are billed inside that state (T1, RHP p.178). The products are sold under the name Dinotarp, mostly to businesses on purchase orders rather than contracts. The top five customers are 69.54% of revenue (T1, RHP p.180).

Half the revenue line is not manufacturing at all. The company buys granules from Reliance, Hindustan Petroleum, Indian Oil and Mangalore Refinery under their bulk purchase schemes, uses part of it in its own factory and resells the rest in the open market. That resale was ₹10.94 crore three years ago and ₹104.01 crore last year, 48.2% of revenue, and it earns about 1.39% after its own cost of funding (T1, Note 17, and derived). The company is explicit about why: buying in bulk under these schemes is what gets it the discount, and the discount is what makes the factory's costs low.

So the money is made in two linked places. The factory converts resin into sheet at a spread, and the spread is wide mainly because the resin is cheap, and the resin is cheap because the company buys a great deal more of it than it converts. Remove the discount schemes and both halves change at once.

Easy or difficult business? Run-of-the-mill. Extruding tape and laminating tarpaulin is standard equipment and standard practice, the machines are bought off the shelf from China and India, there is no approval a competitor cannot get, and the filing holds one unnamed ISO certificate against listed peers that list five standards each. The company's own answer to why customers choose it is that it helps them pick the right thickness, weight and finish and makes each order to size, which is a fair answer for this kind of business but is not measured anywhere. The two real advantages are freight, because bulky cheap sheet is expensive to truck and there is no organised competitor in its home state, and the balance sheet needed to buy resin by the crore. Neither is owned.

Key numbers

Profit grew twice as fast as revenue last year, and the cause is in the cost stack rather than in pricing. Gross margin rose from 13.8% to 16.1% while revenue rose 29.7%, because the mix swung back towards the factory (manufacturing revenue up 68.8%, granule resale up 3.9%) and because the granule discount was still feeding through (T1). Below that, other expenses rose only 11.7% and employee cost is under 0.5% of revenue, so almost all of the gross-margin gain reached EBITDA, and EBITDA rose 80.5% (T1). About a fifth of the two-year margin improvement is not operations: finished goods and work in progress swung from a ₹2.77 crore drawdown in FY2024 to a ₹10.39 crore build in FY2026, and on a value-of-production basis the factory's margin trajectory is 9.76% to 17.37% rather than the reported 9.23% to 19.00% (T1, derived). Finished-goods days rose from 69 to 85 while production rose 131%, so the incremental output of the two machines commissioned in the final quarter has not yet been shown to sell.

Segment revenue is the filing's own note (T1, Note 17). The two margin rows are derived from the filing's cost tables with all overhead charged to manufacturing, which flatters the resale line rather than the factory (T1, derived). Certified production went 2,715 to 4,151.60 to 6,277 tonnes, so the factory growth is tonnes and not price (T1, chartered engineer certificate).

Scorecard

Watch out for

• The 2024 accounts are treated here as unassured, and only those. The firm that signed them held no peer review certificate on the signing date, no peer-reviewed auditor has ever examined that year, and the incoming firm says in the prospectus that it audited none of it (T1 and T2). The 2025 and 2026 numbers are taken at face value in the normal way. Where a 2024 figure matters, and the growth multiple everyone quotes is the main one, it is worth cross-checking rather than accepting.
• The identity question behind the auditor's resignation cannot be closed, and we say so rather than leaving it implied. The auditor who resigned in August 2024 is a one-man firm in Indore. Two weeks after the resignation, three people with that same surname bought 9.1% of the company at ₹35 a share. The firm's owner is a chartered accountant of the same name, at the same address the filing prints. Whether he is the shareholder cannot be established from any public record, and the filing never addresses the connection. We are not saying they are the same person and we cannot say they are not.
• The largest customer, at 41.16% of revenue, is almost certainly a buyer of resold granules rather than tarpaulin, which cuts both ways. It bought ₹88.76 crore last year, flat in rupees against the year before at plus 0.80%, so the fall from 52.97% to 41.16% is entirely the rest of the business growing (T1, RHP p.180). The good news is that the factory business does not hang on one purchase order. The rest is not good: there is no long-term contract, the customer is not named, and we established from outside that the name is not publicly knowable, because no credit rating exists at any agency, counterparty tax data is not public and the company has no exports and so no customs trail. It is not a related party: the largest family firm's entire turnover is 18% of what this customer bought (T1 and T2).
• The objects chapter tells the reader the IPO is buying machines, and does not tell them straight that it is partly paying the company back for machines already on site. The chapter names 2 machines as already acquired and repayable out of proceeds; a footnote 72 printed pages later says 6 more knitting machines are too (T1, RHP pp.107 and 179). On the filing's own wording the single object buys about 4 new knitting machines and reimburses 6, which is of the order of ₹6.94 crore of reimbursement the chapter never identifies, against the ₹7.86 crore and 37.66% it does. The schedule itself adds ₹7.86 crore of past spending to ₹20.88 crore of future spending and totals ₹20.88 crore (T1). Separately, the state may repay up to 40% of the machinery cost as a capital subsidy in seven instalments, and because the object is sized on the full cost, that money would come back to the company with no named use; there is no monitoring agency over the proceeds because the raise is small, only half-yearly audit committee certification (T1).
• The balance sheet is the most levered in the filing's own peer set and next year's costs step up with no debt repayment in the offer. Borrowings are ₹72.51 crore against net worth of ₹27.86 crore, a debt to equity of 2.60 times, where the filing's own commissioned report puts peers at 0.13 to 0.76 times; the current ratio is 0.86 against peers at 1.27 to 3.63, and working capital is negative ₹6.68 crore (T1 and T3, RHP pp.117 and 167). Moratoria are still running on ₹19.24 crore of facilities and several loans were drawn late in the year, so last year's interest charge understates a full year at the contract rates printed facility by facility, up to 12.50% on one invoice discounting line. A full year of interest plus a full year of depreciation on an asset base that doubled, with promoter pay normalised, adds roughly ₹3.1 crore to ₹4.8 crore of cost, 23% to 35% of last year's pre-tax profit (T1, derived). None of the ₹20.88 crore object repays debt, and there is no working capital object either.
• Family firms in the same trade sit outside the company, and the filing says there are fewer of them than its own table shows. Risk factor 41 states that no group company other than two named ones is in a similar business; the filing's own group entity table describes a third, Shri Balaji Plastopack, as trading the same plastic products and polymers, and that third one is the company's largest related-party customer at ₹4.18 crore of the ₹4.36 crore total. The February 2025 non-compete agreements follow the statement rather than the table, so the largest of them is not covered. Same-trade revenue outside the listed entity was ₹22.54 crore in FY2025, 13.56% of the company's revenue that year (T1). Related-party dealings inside the company are small and falling, 2.79% of revenue last year from 3.53% and 9.13%.
• Two smaller disclosure failures worth knowing, because the pattern is the point. The company says only 5 of its 114 employees need to be in the provident fund because the other 109 earn above the wage ceiling; its own total staff and labour cost is 15.93% too small for that to be arithmetically possible, and it states its state insurance coverage as 81 employees on one page and 99 on another (T1). The money at stake is small, roughly ₹35 lakh to ₹70 lakh with interest and penalties, under 2.5% of net worth. It is the explanation that does not hold, not the amount. And no customer, supplier or advance balance was confirmed with the other side in any of the three years, while the company admits "weaknesses and lags" in its internal controls on one page and denies any control failure on another (T1, Notes 42 and 44(A), RHP pp.34 and 47).
• What is clean, stated so the page is not one-sided. Almost no litigation, ₹1.25 lakh of tax matters at family firms against a materiality threshold of about ₹10 lakh. No contingent liabilities in any year. No auditor qualification, emphasis of matter or going-concern doubt in any of the three years, on either side of the auditor change. No shares pledged, stated four times. No offer for sale and no selling shareholder. Nothing adverse on any public register against the company, the promoters, the directors or either auditor, including the accounting regulator's full disciplinary list, and the company's borrowings as filed match the government charge register (T2). The one loan the proceeds repay is the only facility the promoters have not personally guaranteed, and no capital goods were bought from any family entity in three years.

The offer

• Raising about ₹26 crore at the floor and ₹27 crore at the cap, all of it a fresh issue; there is no offer for sale and no selling shareholder, so every rupee goes into the company (T1).
• For ₹20.88 crore of plant and machinery at the existing Khargone plant, three-quarters of it shade-net knitting and blown film, of which ₹7.86 crore reimburses machines already bought and installed, with the balance to general corporate purposes capped at the lower of 15% of the raise or ₹10 crore (T1, RHP pp.99 to 100). No debt repayment and no working capital.
• Implied valuation: the table above, ₹101 crore at the cap.
• Promoters hold 88.51% before the offer and 64.92% after. They sell nothing into the offer and no promoter share is pledged (T1).

Assessment from the Red Herring Prospectus dated 8 September 2026, with no outside verification beyond the price band and offer dates, which come from the BSE public record, and the checks this page names: the Institute of Chartered Accountants' published registers of firms and peer review certificates, the accounting and securities regulators' disciplinary records, the government charge register, official trade data and published industry studies. Numbers carry source tiers: (T1) the filing's audited and certified sections, (T2) exchange, regulator and institute records, (T3) press, aggregators and commissioned or third-party studies, UNVERIFIED where nothing supports them. Not a recommendation. The valuation section states what the announced band implies and is not a view on whether that price is right.