How to Analyse a Staffing / Business Services Stock

Quess Corp employs roughly 478,594 people. It made an EBITDA margin of 2.2% (Quess Q4 FY26 investor presentation). Put those two numbers next to each other and let them sit. Nearly half a million people on the rolls of India's largest domestic staffing player, and the business keeps about two paise of operating profit on every rupee that runs through it. TeamLease, the other listed pure-play staffing house, runs even thinner: an operating margin of 1–2% across the last eight quarters (Inve data, Q3 FY26). (Illustration of how to read the economics, not a view on the stock.)

A first-time reader concludes the business is broken. It isn't. It is a pass-through. A staffing company collects, say, ₹100 from a client to place a worker, pays the worker and their statutory dues maybe ₹90, keeps a markup, and runs its recruiters and compliance machinery out of that thin slice. The "revenue" is mostly other people's salaries flowing through the company's bank account. So almost every ratio you'd reflexively reach for — net margin, return on sales — misleads you, because the denominator is inflated by money that was never the company's to keep.

This is how to read a staffing or business-services company the way the people who run one do: the handful of metrics that decide the outcome, the ones buried in the investor deck rather than the income statement, and the single risk that turns a boring compounder into a statutory landmine. A boundary first — you will not audit a company's PF and ESI compliance from outside. What you can do is read whether the markup is holding, whether the cash is coming in, and whether management's account of its margins survives the concall.

Why is a staffing company a markup-on-payroll business, not a margin business?

Strip the model down and a general-staffing company is a labour-arbitrage utility. It hires a worker the client doesn't want on its own books — to keep its headcount flexible, or to outsource the compliance headache — and bills the client the worker's cost plus a markup. The markup, often called PAPM (profit/revenue per associate per month), is the entire economic engine. Everything else is scale.

That reframes two beginner errors. First, headline revenue growth tells you almost nothing — a staffing firm can grow revenue 15% simply because minimum wages rose and it passed the increase through, with not one extra rupee of its own profit. The revenue went up; the business didn't. Second, the thin margin is the feature, not the bug — this is a volume-times-tiny-spread game, like a toll booth that earns a few rupees per car and makes its money on traffic. The questions that matter: is the per-associate markup holding, is the company collecting what it bills, and is it growing the number of associates profitably?

One fork inside the sector changes everything: general (or "flexi") staffing versus specialised (or "professional") staffing. General staffing places high-volume, lower-skilled workers — retail, logistics, BFSI back-office — at razor-thin markups. Specialised staffing places IT, engineering, and GCC (global capability centre) talent at a far richer one. The gap is not subtle. At Quess in Q3 FY26, General Staffing ran a 1.3% EBITDA margin while Professional Staffing ran 12.5% (Quess Q3 FY26 concall). Same company, same "staffing" label, a near ten-fold difference. If you don't know a company's general-versus-specialised mix, you don't know the business.

The metrics that matter (and where they hide)

The income statement gives you the pass-through. The metrics that decide the investment live mostly in the investor presentation and the concall. Here are the five that earn their place.

Associates / headcount on rolls — the only volume that matters

This is the unit of production. Revenue per associate is roughly fixed by the wage; profit comes from more associates at a stable markup. So track net associate additions, not revenue. Quess carried 478,594 associates at Q4 FY26 (Quess Q4 FY26 investor presentation); TeamLease's general-staffing book alone stood at 2.82 lakh associates at Q3 FY26 (TeamLease Q3 FY26 concall). Where it hides: the deck's operational slide, never the P&L. And read it net of churn — TeamLease lost roughly 27,000 headcount in one quarter when a single BFSI client moved its workers in-house (TeamLease Q3 FY26 concall). A growth story can be one client decision away from reversing.

EBITDA / PAPM per associate — the markup, in rupees

Because percentage margins are distorted by the wage pass-through, the cleaner gauge is profit per associate per month in absolute rupees. TeamLease reports it directly: general-staffing PAPM was ₹665 at Q4 FY25, and — this is the tell — it noted a "QoQ decline of ₹5 and YoY dilution of ₹14" (TeamLease Q4 FY25 concall). Fourteen rupees a month per worker doesn't sound like much until you multiply by 2.8 lakh workers and twelve months. Where it hides: management quotes it on the call; you rarely see it tabulated. A falling PAPM while headcount climbs means the company is buying growth by cutting its own markup — running faster to stand still. In specialised staffing the unit is bigger: Quess reported average revenue PAPM of ₹1.2 lakh per associate in Professional Staffing (Quess Q3 FY26 concall) — two orders of magnitude above a general-staffing worker, which is the whole reason the segment mix matters.

Gross margin and the markup — thin by design, watch the trend

Gross margin here is essentially the markup before overheads, and it is tiny: TeamLease's core staffing (CSR) gross margin sat around 10% (Inve data, Q3 FY25). Don't judge the level — judge whether it's holding. A gross margin sliding while the company wins "marquee" logos is a company buying revenue with price. Where it hides: sometimes only as a derived number; compute it from segment revenue and direct cost in the deck if it isn't stated.

DSO and collect-and-pay — the working-capital game that breaks companies

This is the one that actually kills staffing firms, and it's the most under-watched. The company pays its associates every month whether or not the client has paid. If clients stretch payment to 90 days and the company pays workers in 30, it is financing two months of half-a-million people's salaries out of its own balance sheet. So days sales outstanding (DSO) is not a back-office metric here — it is the survival metric.

The spread between sub-sectors is enormous, and it's the cleanest illustration in the whole sector. Quess and TeamLease (general/specialised staffing, billing large corporates on tight terms) run DSO of 7 days for TeamLease's staffing business and 24 days for Quess general staffing (TeamLease Q2 FY26 concall; Quess Q3 FY26 concall). SIS, a security and facility-management company billing a long tail of smaller, slower-paying sites, runs DSO of 67 days (SIS Q3 FY26 concall) — nearly ten times TeamLease's staffing figure. Same broad sector, wildly different cash dynamics. The related metric Quess discloses is collect-and-pay: the ratio of cash collected to cash paid out, which it deliberately keeps at 75–80% (Quess Q1 FY26 concall) so that it is never paying out faster than it collects. Where it hides: DSO appears on the call and occasionally the deck; collect-and-pay you'll only find if you read the transcript. A widening DSO with reassuring commentary about "lumpy collections" is the single most reliable early warning in this sector.

Cash conversion — does the EBITDA become cash?

Thin-margin businesses live or die on whether reported profit turns into bank balance. Watch operating-cash-flow-to-EBITDA conversion. SIS converted OCF to EBITDA at 163–175% across FY25 (SIS Q3 FY25 and Q4 FY25 concalls) — above 100% because it was releasing working capital as DSO tightened. Quess guides to a sustainable 80–90% and printed 92% in Q3 FY26 (Quess Q3 FY26 concall). Where it hides: the cash-flow slide, or a direct quote on the call. A company whose EBITDA grows but whose cash conversion sags is one whose profit is sitting in receivables — exactly the staffing failure mode.

(Figures from the cited concalls/decks and Inve data; illustration, not a view on any of these stocks. Segments differ — TeamLease and Quess are staffing-led, SIS is security/FM-led — so read down each column, not across rows.)

What valuation lens fits a pass-through business?

Because reported net margins are distorted by the wage pass-through and the share counts are clean, the market values these on price-to-earnings, not EV/EBITDA or price-to-sales — a price-to-sales multiple on a business that's 95% other people's salaries is close to meaningless. But P/E alone misses the point. The right way to think about value here is: what does each associate earn the company, how many associates can it add, and how reliably does that profit turn to cash?

A worked example of why the headline P/E lies. SIS posted a net loss of ₹138 crore in Q3 FY26 and ₹223 crore in Q4 FY25 (Inve data) — yet its operating profit in those same quarters was positive (₹189 crore and ₹165 crore). The losses came from a goodwill-impairment charge, not the operating business. A reader anchoring on the loss-making P/E would have mispriced a business whose underlying security and FM operations were quietly improving their margins. In a serial-acquirer business-services roll-up, you must separate operating performance from acquisition accounting before any multiple means anything. The owner's question is not "what's the P/E" but "what is each of these associates worth to me, and is that number growing?"

A worked case: when "above 6%" became "close to 5%"

The most instructive said-versus-did in this sector is SIS on its core margin, because management is unusually candid about it and the numbers let you grade them.

For years SIS framed its India businesses around a 6% EBITDA margin. On the Q3 FY25 call its management guided that the India Security business should run "at and above 6%" and the FM business likewise (SIS Q3 FY25 concall). In Inve's Promise Tracker, that 6% guidance is now flagged at risk — and the reported numbers show why. India Security's operating EBITDA margin came in at 5.5–5.6% through FY25 and 5.2–5.5% into FY26 (SIS Q4 FY25 and Q3 FY26 concalls); Facility Management improved but only to 5.4% by Q3 FY26 (SIS Q3 FY26 concall). Close, improving — but not yet the number that was guided.

What makes this honest rather than damning is how management explained it, in their own words on the Q3 FY26 call: "we were a 6% margin business in Security, a 6% margin business in Facility management... We slid all the way back to 4%... close to 5% EBITDA margin business" (SIS Q3 FY26 concall). They named the slide, owned it, and explained the climb back. That is a different thing from a company that quietly drops a number and hopes you forget it. TeamLease offers the contrasting shape: its "7.3–7.4% sustainable" specialised-staffing margin guidance was later revised down (Inve data) — a guide that moved, marked against the quarter it was made.

The lesson isn't "SIS missed." It's that in a thin-margin business, a 1-point margin guide that comes in at 5.2% instead of 6% is not a rounding error — it's a 13% miss on the entire profit pool. Half a point of margin is the difference between a good year and a flat one. Watch the margin guidance the way a credit analyst watches credit cost: not the level, but whether the company keeps making, then quietly walking back, the same number.

The non-obvious moat: who actually benefits from labour reform?

Here is the insight that separates this sector from "boring low-margin services," and it came straight out of the SIS Q3 FY26 call. An analyst asked whether labour reforms — which force compliance with PF, ESI, and minimum wage — are an opportunity. Management's answer reframes the whole sector.

Today, they explained, the largest organised players "barely control 5%, 6% market share... the top 10 will not even comprise 15% market share" (SIS Q3 FY26 concall). Why so fragmented? Because of compliance arbitrage. A small unorganised operator competes by simply not paying full statutory dues — and with compliance historically spread across "29 different acts" and "937 different forms," there was "no mechanism for the government to cross-check" (SIS Q3 FY26 concall). The cheat was undetectable, so it was rampant, and it capped the addressable market for the honest, fully-compliant players.

The new labour codes — consolidating that mess into a single licence and a single statutory deposit uploaded to a government portal — make the cheat visible. When compliance becomes enforceable, the unorganised operator's cost advantage evaporates, and the formal share of a vast market shifts to the players who were already paying in full. The same statutory burden that crushes a compliant company's reported margin is its moat — but only if enforcement arrives. That's the bet inside a formalisation thesis, and it's why the security/FM sub-sector, with the longest tail of unorganised competition, has the most to gain. (Illustration of the argument management makes, not a forecast that it will play out.)

The red flags specific to this sector

• Rising DSO with soothing commentary. A staffing company that lets receivables stretch while paying associates monthly is financing its clients' payroll with its own equity. "Collections are lumpy this quarter" three quarters running is the tell.
• Headline revenue growth that's just wage pass-through. If revenue grew 14% and associate count grew 3%, most of the "growth" was minimum-wage indexation, not the business compounding.
• Falling PAPM / markup while headcount climbs. Buying volume by cutting the per-worker spread is running to stand still — and it's invisible if you only watch revenue.
• Inability to pass on minimum-wage increases. When statutory wages rise and a company can't re-price its contracts fast, the increase eats the markup directly. An analyst asked SIS exactly this (SIS Q3 FY26 concall) — it's the right question for every name in the sector.
• Goodwill on the balance sheet from acquisition sprees. Business-services roll-ups grow by buying. SIS's net loss quarters came from goodwill impairment (Inve data). A serial acquirer's reported earnings can swing on accounting, not operations — and a large goodwill block is impairment risk waiting for a bad year.
• Statutory-compliance liabilities. The nightmare scenario: a PF/ESI shortfall surfaces across hundreds of thousands of associates, with penalties and back-dues. This is the one risk that can dwarf the entire thin profit pool, and it never shows up in a ratio.

Frequently asked questions

Where this lens can be wrong. The strongest case against everything above is that the formalisation thesis has been "imminent" for a decade and the labour codes have been notified-but-not-fully-enforced for years. If enforcement keeps slipping, the unorganised operators keep their compliance arbitrage, the addressable market stays capped, and the organised players stay stuck financing thin margins and long receivables — a worse business than the moat story implies. A reader who buys the formalisation narrative without watching whether enforcement actually arrives is buying a thesis, not a business. And the per-associate metrics, however clean, can't see a single large client deciding to in-source its workforce overnight, the way TeamLease's 27,000-headcount loss showed. The honest claim is narrow: these metrics tell you whether the markup is holding and the cash is coming in now — not whether the regulatory tide will lift the sector.

To invert it the way a skeptic should: don't ask "is this a good growth story?" Ask — if this company were quietly buying its headcount growth by cutting the per-worker markup and stretching its receivables to flatter the top line, what would the numbers look like? A falling PAPM, a rising DSO, and revenue growing faster than associate count. If you see all three while management talks about "marquee logos," you've found the failure mode the headline hides.

Compare these against the sibling guides for how to analyse an NBFC — another business where the headline profit lies and the balance sheet is the real story — or how to analyse a steel company, where the per-tonne economics play the role PAPM does here. Inve's KPI Screener lines up associate counts, DSO, and segment margins across business-services names so the first read takes minutes.

And the owner's question to sit with before buying any staffing stock: what must I believe about the next five years of enforcement and pricing power — not this quarter's headcount add — for a business that keeps two paise on the rupee to compound my capital? If the answer leans on a formalisation tide that may or may not come in, you've read the story, not the business.

Related reading

• How to Analyse a Bank Stock (NIM, CASA, GNPA, ROA)
• How to Analyse a Building Products Company (Pipes, Tiles)
• How to Analyse a Cables and Wires Stock