Skip to content

alirezarezvani/claude-skills / finance/skills/stock-analysis

stock-analysis

Unverified

Analyze, research or value a listed stock or company (US, global, NSE or BSE) with verified data, ratios, scores and valuation. Use whenever asked if a stock is a buy, what it is worth, or how a company looks.

FileKindSize
SKILL.mdskill md32.7 KB
evals/evals.jsonconfig13.2 KB
evals/fixtures/synthetic-ar-excerpt.mdmarkdown5.7 KB
evals/fixtures/synthetic-drhp-excerpt.mdmarkdown3.4 KB
examples/README.mdmarkdown1.9 KB
examples/forensic-analysis-example.mdmarkdown8.8 KB
examples/standard-analysis-example.mdmarkdown15.6 KB
references/01-data-sourcing.mdmarkdown38.5 KB
references/02-core-factors.mdmarkdown49.3 KB
references/03-earnings-quality.mdmarkdown40.1 KB
references/04-balance-sheet-and-cashflow.mdmarkdown56.6 KB
references/05-returns-and-dupont.mdmarkdown49.8 KB
references/06-valuation.mdmarkdown53.7 KB
references/07-forensic-red-flags.mdmarkdown57.5 KB
references/08-governance.mdmarkdown67.0 KB
references/09-risk-and-macro.mdmarkdown52.7 KB
references/10-peer-set.mdmarkdown45.5 KB
references/11-scoring-rubric.mdmarkdown36.8 KB
references/12-report-template.mdmarkdown26.6 KB
references/13-situations.mdmarkdown60.6 KB
references/14-accounting-comparability.mdmarkdown48.6 KB
references/15-document-diligence.mdmarkdown76.1 KB
references/16-market-mechanics-and-tax.mdmarkdown48.6 KB
references/17-process-and-epistemics.mdmarkdown59.6 KB
references/18-forensic-mode.mdmarkdown16.5 KB
references/19-ipo-mode.mdmarkdown15.6 KB
references/20-challenge-pass.mdmarkdown14.1 KB
references/21-data-integrity-tools.mdmarkdown7.2 KB
references/sectors/_index.mdmarkdown67.8 KB
references/sectors/auto.mdmarkdown47.5 KB
references/sectors/aviation-hotels.mdmarkdown49.3 KB
references/sectors/banks.mdmarkdown38.6 KB
references/sectors/biotech-clinical.mdmarkdown47.8 KB
references/sectors/chemicals-cement.mdmarkdown46.2 KB
references/sectors/exchanges-payments.mdmarkdown41.2 KB
references/sectors/fmcg-consumer.mdmarkdown35.7 KB
references/sectors/holdco-assetmgr.mdmarkdown41.4 KB
references/sectors/infra-capitalgoods.mdmarkdown43.0 KB
references/sectors/insurance-brokers-services.mdmarkdown49.9 KB
references/sectors/insurance.mdmarkdown38.0 KB
references/sectors/it-saas.mdmarkdown44.3 KB
references/sectors/metals-mining.mdmarkdown49.4 KB
references/sectors/mortgage-reit-specialty-finance.mdmarkdown43.5 KB
references/sectors/nbfc.mdmarkdown40.3 KB
references/sectors/oil-gas.mdmarkdown43.5 KB
references/sectors/people-businesses.mdmarkdown46.3 KB
references/sectors/pharma-healthcare.mdmarkdown40.6 KB
references/sectors/rail-freight.mdmarkdown66.2 KB
references/sectors/realestate-reit.mdmarkdown44.3 KB
references/sectors/retail-ecommerce.mdmarkdown48.3 KB
references/sectors/semiconductors.mdmarkdown43.2 KB
references/sectors/shipping-logistics.mdmarkdown47.2 KB
references/sectors/telecom-media.mdmarkdown43.1 KB
references/sectors/utilities-power.mdmarkdown44.9 KB
references/sectors/waste-environmental.mdmarkdown46.5 KB
scripts/benchmarks.jsonconfig200.5 KB
scripts/lint_report.pyscript47.8 KB
scripts/ratios.pyscript127.4 KB
scripts/score.pyscript96.8 KB
scripts/valuation.pyscript37.9 KB
scripts/verify_data.pyscript72.5 KB

references/sectors/people-businesses.md

View on GitHub
# Staffing, consulting, advertising and professional services — sector playbook

Use this when: the company sells human time, judgement or attention rather than a product — temporary and permanent recruitment, flexi-staffing and payrolling, IT and management consulting, engineering and design services, advertising and media agencies, market research, and outsourced business services such as facilities management, security manpower, testing/inspection/certification and contact-centre operations.

The defining fact is that the productive asset walks out of the building every evening and can resign in the morning. Capital employed is trivial or negative, so ROCE and ROIC print at 40–300% and rank nothing; the binding constraints are headcount, utilisation and bill rate, and the entire margin question reduces to whether the spread between what the client pays for an hour and what the worker is paid for that hour is widening or narrowing. Reported revenue is also unreliable as a scale measure here in a way it is not in most sectors: in staffing it includes pass-through wage cost, and in media it can include client money that never belonged to the agency. Work on **gross profit / net revenue** as the top line, and on **conversion of gross profit into operating profit** as the margin.

If the company owns delivery outcomes with offshore leverage, fixed-price transformation contracts, licensed IP or subscription revenue — Indian tier-1 IT, global SI, SaaS, technology-led BPM — use `references/sectors/it-saas.md`. The two look adjacent and are not: an IT staffing firm bills a markup on a head with no delivery responsibility, while an IT services firm owns the outcome. Applying the offshore-mix, backlog and NRR framework to a staffing company, or EV/gross profit to a services exporter, produces confident nonsense in both directions.

## Contents

- [Why the generic ratio set fails here](#why-the-generic-ratio-set-fails-here)
- [The metrics that actually matter](#the-metrics-that-actually-matter)
- [How to value companies in this sector](#how-to-value-companies-in-this-sector)
- [Peer set construction](#peer-set-construction)
- [Sector-specific red flags](#sector-specific-red-flags)
- [Cycle and structural context](#cycle-and-structural-context)
- [India vs global notes](#india-vs-global-notes)
- [Checklist](#checklist)

## Why the generic ratio set fails here

**ROCE / ROIC / return on assets / asset turnover — spectacular, and empty.** A staffing branch needs desks, phones and a receivables facility; a consulting practice needs laptops. Capital employed is a rounding error against gross profit, so the ratio prints 50–300% for a mediocre operator and can be *negative or undefined* where negative working capital exceeds the fixed asset base (common in agencies holding media payables). Worse, the denominator is then dominated by whatever is unrelated to operations: acquisition goodwill at roll-ups (which mechanically *depresses* ROCE for the firm that paid cash for a good business), net cash piles, and IFRS 16 / Ind-AS 116 right-of-use office assets. Ranking two people businesses on ROCE ranks their acquisition history and lease footprint, nothing else. If capital returns must be discussed at all, discuss **gross profit per employee** and **cash conversion**, which are the economically equivalent questions in a model where the "capital" is payroll.

**OPM / EBITDA margin on reported revenue — mis-signed, and mostly a mix statistic.** In staffing, revenue includes the contractor's own pay, statutory on-costs and any pass-through. A blue-collar temp business at 12% gross margin cannot mathematically print more than ~4% operating margin; a permanent-placement business at ~100% gross margin prints 20%+. Neither number says which is better run. Margin on revenue therefore *falls* when temp volumes grow faster than perm — a mix change, often a share gain, read by a generic screen as deterioration. The sector's own margin metric is the **conversion ratio: operating profit (EBITA) ÷ gross profit**, and it must be used in place of OPM throughout.

**Revenue itself is not comparable across companies.** Advertising agencies do not earn billings; billings are the client's media money passing through. Principal-versus-agent determination under Ind-AS 115 / IFRS 15 / ASC 606 decides whether pass-through media, print production, talent fees and third-party costs are grossed into revenue or netted out — and the answer changes reported "revenue" by a multiple with zero change in economics. US holding companies and Indian agencies report on different conventions, and a firm that shifts media buying from agency to **principal** (inventory bought on its own book and resold) grosses up revenue overnight. Never build a growth series, an EV/Sales multiple or a market-share estimate on the top line until you have confirmed the basis.

**Free cash flow — inverted in staffing, seasonal and window-dressable in agencies.** Staffing pays contractors weekly or fortnightly and collects from clients in 45–75 days, so growth consumes cash and contraction releases it. A staffing firm in a violent downturn posts its best-ever FCF as the receivable book unwinds; the same firm in a strong recovery posts weak FCF while creating value. Agencies run structurally *negative* working capital because media payables exceed receivables, which makes year-end net cash a function of payment timing rather than earnings — always use **average net debt** (many disclose it; if not, model intra-year peaks) and treat a large year-end cash balance as unavailable to shareholders. In both cases, FCF is a working-capital signal first and a profitability signal second.

**D/E, net debt and current ratio — distorted at both ends.** Balance sheets are small, so modest absolute debt produces alarming-looking ratios, while receivables and payables gross-ups from pass-through costs swamp the current ratio in agencies and outsourcers. More importantly, the real obligations sit off the standard leverage screen: **deferred consideration and earn-outs** on acquired agencies and boutiques, **put options held by minority shareholders** of subsidiary agencies (a contractual cash obligation, frequently excluded from screened net debt), pension deficits at old-line professional services firms, and IFRS 16 lease liabilities on large office estates that are a genuine operating cost, not financing.

**EV/EBITDA post-IFRS 16 — overstated and non-comparable.** Office rent is a real, recurring, unavoidable cost in a business whose only other cost is payroll. Capitalising it moves 3–8% of gross profit out of EBITDA. Compare on EBITDA-after-leases (EBITDAaL) or on EBITA, and confirm every peer is on the same basis and the same standard-adoption date.

**P/E on reported EPS — the classic cyclical trap, in both directions.** Because the cost base is semi-fixed (recruiters and consultants cannot be hired and fired at the speed of demand), **drop-through** of a gross-profit decline to operating profit runs 50–70%. A 20% fall in gross profit can halve or eliminate EPS. P/E is therefore lowest at the peak — when perm fees and utilisation are maxed — and infinite or negative at the trough, which is often the correct time to buy. Normalise to mid-cycle before applying any multiple.

**P/B and book value — no information.** Book equity is acquisition goodwill plus receivables less buybacks; serial acquirers of people businesses have large intangible balances backed by staff who can leave, and mature agencies have driven tangible book negative. Impairment tests here are a lagging confirmation of a client loss you should already have found in the organic growth bridge.

**Inventory, fixed-asset turnover, capex/sales — not applicable.** No inventory; maintenance capex is typically 0.5–2% of revenue. Depreciation ex-leases is trivial, so EBITDA ≈ EBITA ≈ EBIT and EBITDA adds nothing except the lease distortion.

**Headline growth as an unqualified positive — inverted twice.** First, acquired growth is bought, not earned, and this sector consolidates constantly. Second, revenue growth achieved by taking on payrolling, MSP pass-through or low-markup volume *dilutes gross profit margin* and can raise revenue while gross profit is flat. Growth must be stated as **organic, constant-currency, gross-profit (or net-revenue) growth** or it is not a fact.

## The metrics that actually matter

Ranges are **indicative only**. They vary by sub-sector, geography, wage regime, cycle phase and reporting period, and any of them can be wrong for a specific company in a specific year. Comparison against a tight peer set and against the company's own 5–10 year record — including its last full downturn — overrides every absolute band below.

| Metric | Definition / how to compute | Indicative healthy range | Why it matters |
|---|---|---|---|
| **Gross profit / net fee income (the real top line)** | Revenue less the direct cost of the people billed (contractor pay, statutory on-costs) and less pass-through third-party costs. UK/European staffing calls it **net fee income (NFI)**; agencies call it **net revenue** or revenue-less-pass-through. Rebuild it consistently for every peer. | Staffing GP margin: blue-collar/industrial temp 10–16%, professional/IT temp 18–28%, blended 15–25%; perm placement ~100%. Agencies: net revenue typically 15–25% of billings, but the ratio is convention-dependent. | This is the only number that measures the company's own economic activity. Everything above it belongs to the worker or the media owner. Growth, margin, per-head productivity and valuation must all be computed on this line. A company that talks about revenue growth while gross profit is flat is growing its pass-through, not its business. |
| **Conversion ratio (EBITA ÷ gross profit)** | Operating profit before amortisation of acquired intangibles and before exceptionals ÷ gross profit. Compute on a lease-consistent basis across peers. | Staffing 10–18% mid-cycle, >20–25% at a cyclical peak with high perm mix, 0–5% or negative at the trough. Consulting 12–20%. Agencies (operating margin on net revenue) 13–18% for large groups. Judge against the company's own last two cycles. | The sector's true margin, and immune to the temp/perm and principal/agent mix distortions that destroy OPM. It answers the only operating question that exists here: how much of the spread the firm captures survives its own SG&A. A conversion ratio at an all-time high late in an upcycle is peak earnings, not a re-rated business. |
| **Utilisation (billable hours ÷ available hours)** | For delivery staff only, stated on a defined available-hours base, and disclosed both including and excluding trainees/bench. Track alongside average billable headcount. | Strategy/management consulting 60–75%; IT and implementation consulting 75–85%; engineering/testing services 80–88%. Above ~88% means no bench and imminent delivery and attrition risk. | Utilisation is the volume lever and the earliest demand signal in consulting — it falls a quarter or two before revenue, because staff are hired to a forecast. Roughly 100 bps of utilisation converts to a meaningful chunk of EBIT margin in a business where payroll is 60–70% of net revenue. A margin beat produced by pushing utilisation to a record is borrowed from next year's attrition. |
| **Realisation rate (billed ÷ standard rate)** | Fees actually invoiced and collected ÷ fees at standard rate card for hours worked. Equivalently, the discount and write-off leakage between timesheet and invoice. | 85–95% healthy; a persistent slide below ~80% signals pricing indiscipline, scope overruns being absorbed, or a rate card that no longer clears the market. | Utilisation can be held up by billing work cheaply — realisation is where that shows. Rising utilisation with falling realisation is a demand problem being disguised as a capacity success. Also the first place fixed-price overruns appear, since unbilled effort is written off here before it reaches the margin line. |
| **Pyramid ratio / leverage** | Billable staff per partner or per senior manager; share of headcount by grade; average cost per billable head; in Indian delivery models, fresher share of headcount. | Depends entirely on the model: strategy houses run flat pyramids, implementation and engineering practices run steep ones. Track the *direction* — a flattening pyramid at constant pricing is structural margin loss. | The pyramid is the profit engine: juniors are billed at a multiple of cost, seniors are not. Margin can be manufactured for two or three years by hiring cheaper and pushing leverage, and that is a different quality of earnings from a rate rise. A pyramid that flattens because juniors left, or because clients refuse to pay for inexperienced staff, is the mechanism behind most consulting margin decay. |
| **Bench cost and time-to-deploy** | Non-billable delivery payroll ÷ net revenue; average days between a consultant rolling off a project and rolling on to the next; share of headcount on bench beyond 30/60 days. | Bench cost low single-digit % of net revenue in a normal market; deployment gaps rising sharply is a leading indicator of a utilisation and then margin fall. | Bench is committed cost against uncertain revenue — the closest thing to inventory in this sector, and it spoils. Companies frequently hide it by reclassifying benched staff to "internal initiatives", "investment projects" or capitalised platform work; check whether internal-project headcount grew when utilisation fell. |
| **Attrition (voluntary, LTM) and replacement cost** | Voluntary attrition of billable staff, split regretted vs unregretted and by grade. Replacement cost = recruitment fee + onboarding + lost productivity ramp, conventionally estimated at 50–150% of annual salary for professional staff. | Consulting 12–20% is typical; >25% is a delivery-quality and cost crisis. Very *low* attrition often signals a weak external job market, i.e. weak demand — read it with hiring. | This is the depreciation schedule of the only asset. High attrition means backfilling at market wages (spread compression), more subcontracting, project escalations and lost client relationships. Senior attrition matters disproportionately: partners and practice heads leave with the client and often with the team. |
| **Bill rate, pay rate and the pay-bill spread** | Average client bill rate per hour/day and average worker pay rate, tracked separately, with the spread in both absolute currency and percentage terms. Never rely on the percentage alone. | Spread stable-to-widening in real terms is the whole thesis. A stable % spread on a rising pay rate is *absolute* margin growth; a stable % on a falling pay rate is decline. | Wage inflation is passed on as a higher bill rate at a constant percentage markup, which mechanically raises gross profit per head — flattering "growth" that is pure inflation. Splitting the two separates volume, price and inflation. It is also where client procurement pressure first becomes visible, months before it reaches the margin. |
| **Contractors on assignment / billable headcount** | Number of temps or contractors out on assignment at period-end and average during the period (staffing); average billable delivery headcount (consulting); average full-time equivalents on contract (outsourcing). Track weekly or monthly where disclosed. | Direction and inflection matter, not the level. Compare growth in heads on assignment against growth in gross profit to isolate mix and rate. | The physical unit of production. Gross profit = heads × hours × spread; decomposing it tells you whether growth came from volume, hours (a real risk in a slowdown, as clients cut overtime before cutting heads) or rate. Heads-on-assignment inflects before revenue and before any accounting number. |
| **Permanent placement mix and perm fee** | Perm placement revenue as % of gross profit; average fee per placement (typically 15–25% of first-year salary); placements made per consultant per month. | Perm is often 15–35% of a diversified staffing firm's gross profit, and materially more of its profit. | Perm is the highest-margin, most cyclical and most information-rich line in the sector: it is ~100% gross margin, so its decline drops through to EBIT almost one-for-one, and it is the first thing corporate clients stop buying and the first thing they resume. A staffing firm's profit cycle is largely the perm cycle amplified by operating leverage. |
| **Fill rate and order-to-fill time** | Orders filled ÷ orders received (or vacancies taken); median days from requisition to placement/start; interview-to-placement conversion. | Fill rate 55–80% depending on skill scarcity and whether the firm accepts speculative orders; order-to-fill trending down. | The operational quality of the recruiting engine, and the leading indicator inside a leading indicator. Rising orders taken with a falling fill rate means either a labour shortage (bill rates about to rise) or an eroding candidate database. Falling orders taken is the cleanest early read on client demand and turns before placements. |
| **Client concentration, retention and contract tenure** | Top-5 / top-10 clients as % of gross profit (not revenue); annual client retention %; contract renewal/rebid win rate; average remaining contract tenure for outsourced services; average client relationship age for agencies. | Top-10 below ~20–25% of gross profit for a diversified firm; any single client >8–10% is a material risk. FM/security/outsourcing renewal rates 90–95%+; agency top-20 client tenure often measured in decades, which cuts both ways. | Revenue here is a relationship, not a contract — often terminable on 30–90 days' notice in advertising, and rebid every 3–5 years in outsourcing. A single account loss can remove several points of growth in a quarter with no warning in the financials. Compute concentration on gross profit, since a low-margin pass-through account overstates its own importance on revenue. |
| **Organic, constant-currency net revenue / gross profit growth (with the full bridge)** | Reported growth = organic + acquisitions/disposals + FX + (for agencies) pass-through and principal-accounting effects. Demand the bridge; rebuild it if not given. | Agencies: 0–4% organic in a normal market, >5% strong, negative organic with positive reported growth is the standard warning configuration. Staffing: swings from +20% to −25% across a cycle — the level is meaningless without cycle position. | This is where managements hide. Roll-ups mask organic decline with acquisitions; agencies have flexible definitions of "organic" that sometimes absorb pass-through and same-store adjustments. Also check whether "organic" includes revenue from businesses acquired 13 months ago at their inflated pre-deal run rate. |
| **Backlog, bookings and book-to-bill** | Signed but undelivered fees; new bookings/TCV in the period; book-to-bill = bookings ÷ revenue; backlog expressed as months of revenue. Split project vs retainer/managed-service, and fixed-price vs time-and-materials. | Consulting book-to-bill >1.0x, 1.1–1.3x in an upcycle; backlog cover of 3–9 months typical. Retainer/annuity share above ~50% materially reduces cyclicality. | The only forward-looking number in a sector with no order book in the industrial sense. Bookings turn 2–3 quarters before revenue. The fixed-price share carries the estimation risk: percentage-of-completion accounting on fixed-price work is where optimistic cost-to-complete assumptions live, and where a delivery problem becomes a sudden write-off rather than a gradual margin slide. |
| **Wage-inflation pass-through coverage** | Share of contracted revenue with explicit indexation (CPI, wage-index or statutory-minimum-wage pass-through clauses); the reset frequency and lag in months; historical realised pass-through vs actual wage inflation. | Contracted outsourcing: aim for majority coverage with ≤6–12 month lag. Staffing: markup-based pricing passes wages through mechanically but only on *new* assignments — legacy assignments reprice at renewal. | The central margin question in the entire sector. Where pricing is a percentage markup, wage inflation is neutral-to-positive; where pricing is a fixed rate per head or per site on a multi-year contract, wage inflation is a direct hit to the spread and cannot be recovered until rebid. Statutory minimum-wage rises are the acute case: cost rises on a legislated date whether or not the contract allows recovery. |
| **Gross profit per employee, and cash conversion** | Gross profit ÷ average *internal* headcount (recruiters, consultants, support) — not contractors on assignment. Cash conversion = operating cash flow ÷ EBITA, and DSO including unbilled/contract assets. | GP per internal head should rise in real terms across a cycle. Staffing DSO 45–75 days; consulting DSO including unbilled 60–90 days. Cash conversion >90% of EBITA through a full cycle. | Productivity per internal head is the operating-leverage measure that ROCE cannot provide, and it exposes whether technology and AI investment is doing anything. Cash conversion is the honesty check on revenue recognition — unbilled/contract assets growing faster than revenue in a consulting firm is the same warning it is in IT services: work recognised but not yet accepted or invoiced. |

### Family-specific additions

**Staffing.** Split gross profit into temp and perm, and temp further into professional and industrial. Track heads on assignment weekly where disclosed, average assignment length, temp-to-perm conversion income, and the share of gross profit from MSP/RPO/VMS programmes (large, sticky, low-margin) versus direct branch recruitment (small, high-margin, cyclical). Watch employment-tax and worker-classification exposure (UK IR35, US federal/state independent-contractor tests, EU Agency Workers Directive equal-treatment after a qualifying period). Note the sub-sectors with their own cycles: healthcare/nurse staffing (driven by clinical shortages and rate spikes, not GDP), education, and government/defence contracting.

**IT and management consulting.** Beyond utilisation, realisation and pyramid: partner/practice-head tenure, revenue per partner, share of revenue from the top practice and top industry vertical, subcontractor cost as % of net revenue (spikes when demand outruns hiring), and the fixed-price share of backlog with the associated cost-to-complete provisioning. Distinguish transformation projects (discretionary, cut first, high margin) from run/managed services (annuity, resilient, lower margin). A book that has drifted from project to run is more defensive and worth a lower multiple, not a higher one.

**Advertising and marketing agencies.** State plainly in any report that **billings are not revenue**. Then: organic net revenue growth, staff cost ÷ net revenue (typically 60–68%; the single biggest lever), new business wins and losses net, media versus creative versus data/technology/CX mix, share of media bought as principal, and the payables/receivables float. Principal media buying converts an agency fee into a trading spread — higher headline revenue, undisclosed margin, and a genuine inventory and transparency risk; the industry-wide advertiser scrutiny of media rebates and non-transparent principal trading is the reference case, and it never fully went away. Check whether client contracts are 30–90 day terminable and how many of the top-20 relationships are up for statutory review.

**Other professional and business services (security, facilities management, TIC, outsourced customer operations).** Contract tenure and the maturity ladder of the contract book, renewal and rebid win rates, revenue under contract for the next 12/24 months, site or seat counts, revenue per site/seat/FTE, wage-indexation clause coverage, statutory minimum-wage and social-security exposure by geography, and mobilisation cost on contract wins (upfront cash out, recovered over the term — a real drag on FCF during a growth phase). TIC businesses are the exception in this family: they carry laboratory capex and accreditation moats, so asset-based returns are *not* meaningless there, and margins are structurally higher (15–25% EBIT).

## How to value companies in this sector

Value the **enterprise on normalised mid-cycle profit**, and never capitalise the profit the company is earning today unless you have established where in the cycle today sits.

**Primary — EV/EBITA on normalised earnings.** Use EBITA, not EBITDA: amortisation of acquired intangibles should be excluded (it is an artefact of deal accounting), but depreciation and lease costs should not be, because office estates are a real recurring cost. Confirm every peer is on the same lease standard and the same treatment. Normalisation means resetting the conversion ratio to the company's own mid-cycle level, not the current one — a staffing firm at a 22% conversion ratio in a hot labour market and one at 4% in a recession may be identical businesses. Indicative: staffing 7–11x mid-cycle EV/EBITA, consulting 9–14x, agencies 7–11x on net revenue-based EBITA, contracted outsourcing 8–12x. All cycle- and market-dependent.

**Staffing — EV/gross profit is the sector-standard cross-check, and the only comparable revenue multiple.** Because revenue includes contractor pay, EV/Sales compares temp/perm mix rather than value. EV/gross profit removes that entirely. Indicative 0.8–2.0x, with the position inside that range set almost mechanically by the mid-cycle conversion ratio and the perm share — a firm converting 18% of gross profit at the mid-cycle deserves roughly double the EV/GP of one converting 9%. Sense-check the implied EV/EBITA against it.

**Agencies — EV/net revenue and EV/EBITDA on net revenue.** Indicative EV/net revenue 1.2–2.5x, EV/EBITDA 6–10x for the large groups, higher for genuinely growing digital/data-led independents. Two adjustments are non-negotiable: add **deferred consideration, earn-outs and the fair value of minority put options** to enterprise value, and use **average** rather than year-end net debt, because the media float makes the balance-sheet date unrepresentative.

**P/E on normalised cycle earnings.** Legitimate here, unlike in banks, because these are genuinely simple P&Ls — but only after mid-cycle normalisation. The practical method is to compute mid-cycle gross profit (trend line through the last two full cycles), apply the mid-cycle conversion ratio, tax it, and apply a mid-cycle multiple. Reported trough P/E of 40x and peak P/E of 8x can describe the same fair value.

**Free cash flow yield and DCF — usable, with two corrections.** Capex is trivial and cash conversion is high, so DCF is more tractable here than in most sectors. But (i) model working capital explicitly as a function of gross profit growth — it is the difference between a cash-generative and a cash-consumptive year in staffing, and it flips sign across the cycle — and (ii) build the forecast off mid-cycle margins, since terminal value dominates and a terminal margin set at the current peak is the standard way DCFs in this sector produce absurd answers. For contracted outsourcing with long tenure and high renewal rates, a contract-by-contract run-off plus renewal-probability model is materially better than a growth-rate DCF.

**Cross-checks.** EV per billable head or per consultant against transaction comparables; the implied mid-cycle conversion ratio and implied organic growth solved backwards out of today's price (then ask whether the company has ever achieved them); and dividend/buyback sustainability against average, not year-end, net debt.

**Do not use:** EV/Sales or P/S for staffing and for agencies reporting on a principal or grossed-up basis; P/B and any book-value-anchored method; ROCE- or ROIC-based screens and quality scores; EV/EBITDA compared across firms on different lease standards; PEG on reported EPS at either end of the cycle; and replacement-cost or asset-based methods, which have no meaning when the assets are employed under a resignation clause.

## Peer set construction

A valid comparable shares the **revenue model, the top-line convention, the geography's labour regime and the cycle position** — not the label "human capital" or "professional services".

**Splits that must never be mixed:**

- **Staffing vs consulting vs agencies vs outsourced services.** Four different economic models. Staffing sells heads at a markup; consulting sells outcomes at a pyramid margin; agencies sell attention and take a fee on someone else's money; outsourcing sells a contracted service at a fixed price with wage risk retained. Their margins, cycles, working capital and multiples do not overlap.
- **Within staffing: perm-heavy vs temp-heavy, and professional vs industrial.** A firm with 35% of gross profit from perm has roughly double the earnings cyclicality of a temp-only industrial firm at the same size. Also separate branch-led recruitment from MSP/RPO/VMS programme businesses — different margins, different stickiness, different capital intensity.
- **Within consulting: strategy vs implementation vs engineering/R&D vs regulated-adjacent (audit/tax/risk).** Different pyramid shapes, different discretionary exposure, different pricing power.
- **Within agencies: holding companies vs independents vs performance/digital specialists vs martech.** And within holdcos, never compare across different principal-vs-agent revenue conventions without restating.
- **Contracted outsourcing vs discretionary services.** Revenue visibility of 12–36 months versus 30–90 days is the single biggest driver of the multiple in this family.
- **Geography, because the labour regime changes the economics.** US at-will employment gives high margin volatility and fast cost adjustment; Continental European employment protection dampens both the upswing and the downswing and changes the trough conversion ratio structurally; Japan has its own dispatch-labour rules; the UK carries IR35 and agency-worker equal-treatment. India's flexi-staffing is a different business again (see below).
- **Cycle position.** Comparing a firm in a market at the top of its labour cycle with one at the bottom, on current-year multiples, is the most common error made in this sector. Compare mid-cycle to mid-cycle.

**Also align:** the top-line basis (gross profit / net fee income / net revenue, restated identically for all peers); lease standard and adoption date; fiscal year end; whether amortisation of acquired intangibles is in or out of the "adjusted" figure; and organic-growth definitions. Aim for 5–8 peers, state the basis explicitly, and benchmark every metric twice — against peers and against the company's own record through its last full downturn.

## Sector-specific red flags

- **Revenue growing while gross profit is flat.** Volume taken at negligible markup — payrolling, MSP pass-through, principal media, low-margin master-vendor deals. It buys scale headlines and no profit, and it permanently dilutes the reported margin, which management then explains away as "mix" every year.
- **Conversion ratio at an all-time high, presented as structural improvement.** Late-cycle perm fees and peak utilisation are not a re-rating; they are the thing that reverses. Ask what the conversion ratio was in the last two troughs.
- **Cost taken out of the front line to defend margin while gross profit falls.** Cutting recruiters, business development or junior consultants protects this year's EBIT and mortgages the recovery — the firm cannot fill orders when demand returns. Track sales/recruiter headcount against gross profit; a firm cutting producers faster than gross profit is falling is managing to a number.
- **Serial "adjusted" earnings.** Restructuring charges every single year, acquisition and integration costs treated as exceptional at a company whose growth strategy *is* acquisition, and amortisation of acquired intangibles added back at a firm that must keep buying teams to replace ones that left. Compute a 5-year total of "exceptionals" against 5-year cumulative reported EBITA.
- **Acquisitions masking organic decline.** Demand the organic bridge, check the definition, and check whether acquired businesses are lapped at their pre-deal run rate. Also watch earn-out structures: an earn-out expiring is a well-documented trigger for founder departure and revenue decline in acquired agencies and boutiques.
- **Unbilled revenue / contract assets growing faster than revenue in consulting.** Fixed-price percentage-of-completion recognition with optimistic cost-to-complete estimates. A step-change in unbilled, or a lengthening gap between recognition and invoicing, precedes write-offs.
- **Bench disguised as investment.** A jump in "internal projects", "platform development", capitalised software or "training investment" headcount at exactly the moment utilisation falls. Reconcile total delivery headcount to billable headcount every period.
- **Withdrawal of a previously given operating disclosure.** Companies here voluntarily publish heads on assignment, perm placement counts, utilisation, organic growth by discipline or net new business. A firm that quietly stops disclosing one it used to give is telling you the number turned.
- **Agency-specific:** a shift toward principal media buying without margin disclosure; client-money and float being described as balance-sheet strength; "net new business" reported as wins without netting losses; a top client entering statutory media review; late or vague disclosure of an account loss; and organic growth definitions that flex between years.
- **Key-person and team-lift risk.** Departure of a practice head, creative leader or desk manager, typically followed by a team and then the clients. Check non-compete enforceability by jurisdiction (unenforceable or narrowly enforced in several US states), partner/senior tenure, and how much of gross profit sits with the top 10 producers.
- **Worker-classification and employment-tax exposure.** Contractors reclassified as employees creates retrospective tax, social-security and benefit liabilities that dwarf annual profit. Read the contingent liabilities note for statutory wage, provident fund, social-security and misclassification claims — this is the sector's equivalent of a credit event.
- **Wage inflation without pass-through.** Multi-year fixed-rate contracts signed in a low-inflation period, statutory minimum-wage increases, or a rebid won on price that assumed flat wages. The margin damage arrives on a legislated date and lasts until renewal.
- **Client concentration measured on revenue rather than gross profit**, and a single client, sector or government framework agreement carrying a disproportionate share of profit.
- **Receivables quality.** Rising DSO, factoring or receivables-financing programmes introduced quietly (which flatter reported cash conversion and net debt), concentration of receivables in one distressed client, or extended payment terms conceded to hold an account — a price cut recorded as a working-capital movement.
- **Goodwill that never impairs** at a roll-up whose acquired businesses are visibly shrinking, alongside an accounting policy of long or indefinite intangible lives on customer relationships that churn.

## Cycle and structural context

**Staffing is a leading macro indicator, and knowing that is half of using it.** Temporary employment is one of the cleanest early cyclical series available — clients add and cut flexible labour before they touch permanent headcount. Within the sector, **permanent placement turns first at both ends**: perm fees are pure discretionary spend and collapse two to three quarters ahead of the broader economy, and resume ahead of it. The sequence to watch is: orders/vacancies taken → fill rates and perm placements → contractors on assignment → temp gross profit → reported revenue. Because perm is ~100% gross margin and the cost base is semi-fixed, drop-through of a gross-profit decline to EBIT of 50–70% is normal, which is why peak-to-trough EPS declines of 50%+ are routine and why the reported P/E is at its most misleading at both extremes.

**Consulting lags staffing modestly and asymmetrically.** Backlog cushions the first two or three quarters of a downturn, then discretionary transformation work is cancelled or deferred while run/managed services persist. On the recovery, bookings turn before revenue by a similar lag. Cost-reduction and restructuring practices are counter-cyclical and partially offset — check the practice mix before assuming a uniform cycle.

**Advertising tracks nominal GDP with a beta above one.** Ad spend is a discretionary corporate budget cut early and restored late, and agencies sit one step further from the money than media owners. Event-driven boosts (elections, major sporting cycles) accrue mostly to media owners, not agencies. Currency matters disproportionately, since the large groups earn across dozens of currencies with costs in the same places.

**Contracted outsourcing, security and facilities management are the defensive corner** — low cyclicality, high renewal, but low margin and permanent exposure to wage regulation. The risk here is not demand; it is a minimum-wage increase, a social-security rate change or an aggressive rebid.

**Structural pressures to score explicitly, not to wave at:**

- **Generative AI on the entry-level pyramid.** The economics of consulting, research, creative production and outsourced customer operations depend on billing junior hours at a multiple of junior cost. Anything that automates first-draft research, deck production, code scaffolding, creative variants or tier-1 support attacks the widest part of the pyramid. Test the claim empirically: if a firm says AI is making it more productive, gross profit per internal head must rise. Flat for five years means the story is marketing.
- **In-housing.** Advertisers have moved media planning, creative production and data operations in-house; corporates have built internal consulting and internal recruitment functions. This removes fee pools permanently rather than cyclically.
- **Platform self-serve.** Automated buying on the large ad platforms and retail media networks compresses the media planning and buying fee pool; the agency response has been data, technology, commerce and CX services, which carry different margins and different competitors (consultancies, SIs, martech vendors).
- **Procurement professionalisation in staffing.** MSP/VMS programmes, direct sourcing, talent marketplaces and freelancer platforms all compress markups and disintermediate the branch. Large, stable, low-margin programme revenue displacing small, cyclical, high-margin direct revenue is a genuine change in business quality that shows up first as gross margin dilution.
- **Regulatory drift toward reclassifying flexible labour as employment** across most major markets. This is the sector's tail risk.

## India vs global notes

| Dimension | India | US / global |
|---|---|---|
| Sector composition | Dominated by **general (blue-collar) flexi-staffing** and **IT staffing**, plus facilities management, security manpower and a fast-growing GCC-support ecosystem. Listed advertising and management consulting pure-plays are few; agency businesses are often subsidiaries of global groups | Deep listed universe across staffing, consulting, agency holding companies, TIC and outsourced services, with long multi-cycle histories |
| Top-line convention | Ind-AS 115 principal/agent determination; most Indian staffing firms report gross billings as revenue with associate salary cost below it. **Gross margin is far thinner than Western peers** — general staffing often mid-single-digit % of revenue, IT staffing low-to-high teens — so net margins of 1–3% are structurally normal, not distress | UK/European staffing headline **net fee income**; US staffing reports revenue and gross profit prominently; agency holdcos report revenue with pass-through costs disclosed, conventions changed materially post-ASC 606 |
| Labour regulation (India-specific) | Contract Labour (Regulation and Abolition) Act 1970 and state rules; the four Labour Codes (Wages, Industrial Relations, Social Security, OSH) and their staged implementation; EPFO and ESIC contributions; state-wise minimum wages revised on a fixed cycle; Shops and Establishments Acts state by state; principal-employer liability for statutory dues, which makes compliance itself a competitive moat for large organised players | US: federal/state independent-contractor tests (including California's ABC test), joint-employer doctrine, at-will employment. UK: IR35 off-payroll rules and Agency Workers Regulations equal treatment after a qualifying period. EU: Agency Work Directive and strong dismissal protection |
| Tax and working capital (India-specific) | GST at the standard rate on the **full billing including wages passed through**, creating a large input-credit and working-capital drag; TDS under s.194C/194J; PF/ESI deposit timing. Formalisation and GST have been a structural share gain for organised staffing over the unorganised market | Sales tax generally not applied to staffing wages in the same grossed-up way; payroll taxes are the equivalent cash-timing item |
| IT staffing vs IT services (India-specific) | Distinct businesses that screens routinely merge. Staffing bills a markup on a deployed head, carries no delivery risk, has no offshore pyramid, no fixed-price exposure and no backlog in the services sense. **Do not** apply utilisation-and-offshore-mix, NRR, TCV or FCF/PAT frameworks from `it-saas.md`; **do not** apply EV/gross profit to an IT services exporter. The GCC (global capability centre) build-out is a demand driver for IT staffing, contract-to-hire and facilities services simultaneously | The equivalent split is contract IT staffing vs systems integration; the same warning applies but the gross margin gap is narrower |
| Disclosure practice | Quarterly results under SEBI LODR; associates/headcount deployed, client counts and segment splits are **voluntary** and disclosure quality varies widely — read the investor presentation and concall, which are often the only source of operating KPIs. ₹ crore/lakh; April–March fiscal year; promoter holding and pledge disclosure matter; related-party note for group manpower contracts | 10-K/10-Q on EDGAR with segment and often weekly/monthly operating data; UK/European annual reports disclose NFI by discipline and geography; earnings calls give perm/temp splits and heads on assignment |
| Regulator / oversight | SEBI for disclosure; labour ministries (central and state) for compliance; EPFO/ESIC; the Advertising Standards Council of India (self-regulatory) for advertising content | SEC; Department of Labor and state agencies; FTC for advertising practices; industry bodies for media audit and transparency standards |
| Valuation convention | Multiples usually quoted on P/E and EV/EBITDA of reported (grossed-up) numbers — recompute EV/gross profit before comparing an Indian staffing firm with a global one, or the Indian firm will look absurdly expensive on EV/Sales-adjacent measures | EV/gross profit and EV/EBITA are the working staffing conventions; EV/net revenue for agencies |

## Checklist

- [ ] Confirm the family: staffing, consulting, agency, or contracted business services — and confirm it is not an IT services/BPM company that belongs in `it-saas.md`.
- [ ] Delete ROCE, ROIC, ROA, asset turnover, P/B and EV/Sales from the analysis, and say in the report why they are inapplicable.
- [ ] Rebuild the top line as gross profit / net fee income / net revenue, on an identical basis for every peer, and do all growth and margin work on that line.
- [ ] Replace OPM with the conversion ratio (EBITA ÷ gross profit) and compare it to the company's own last two cycle peaks and troughs.
- [ ] Decompose gross profit into heads × hours × spread; state which of the three produced the growth.
- [ ] Split bill rate and pay rate separately — establish how much of "growth" is simply wage inflation passed through at a constant markup.
- [ ] Staffing: split temp vs perm gross profit, get perm as % of gross profit, and estimate the drop-through if perm falls 30%.
- [ ] Staffing: check contractors on assignment, orders taken, fill rate and order-to-fill trend — the earliest demand signals available.
- [ ] Consulting: check utilisation (ex-trainees), realisation, pyramid shape, bench cost, subcontractor cost and the fixed-price share of backlog.
- [ ] Consulting: check unbilled revenue / contract assets against revenue growth, and DSO including unbilled.
- [ ] Agencies: state explicitly that billings are not revenue; get organic constant-currency net revenue growth with the full bridge; check staff cost ÷ net revenue.
- [ ] Agencies: check principal vs agency media buying, use average (not year-end) net debt, and add earn-outs, deferred consideration and minority put options to enterprise value.
- [ ] Outsourced services: check contract tenure ladder, renewal/rebid win rates, revenue under contract for 12/24 months, mobilisation cash cost, and wage-indexation clause coverage with its lag.
- [ ] Quantify wage-inflation pass-through: what share of the book reprices automatically, what share only at rebid, and what statutory wage changes are scheduled.
- [ ] Compute attrition (voluntary, by grade), estimate replacement cost, and identify key-person and team-lift risk in the top producers.
- [ ] Compute client concentration on **gross profit**, not revenue, and check contract termination notice periods.
- [ ] Build the organic growth bridge: organic + M&A + FX + pass-through/principal effects; interrogate the company's definition of "organic".
- [ ] Total five years of "exceptional" and restructuring charges against five years of reported EBITA.
- [ ] Normalise to mid-cycle gross profit and mid-cycle conversion ratio *before* applying any multiple; state explicitly where in the labour/ad cycle this sits.
- [ ] Value on EV/EBITA (mid-cycle) and EV/gross profit (staffing) or EV/net revenue (agencies); model working capital as a function of gross-profit growth in any DCF.
- [ ] Check whether reported cash conversion is helped by factoring or receivables financing, and whether DSO is drifting.
- [ ] Score the structural threats explicitly — AI on the junior pyramid, in-housing, platform self-serve, MSP/VMS markup compression — and test the productivity claim against gross profit per internal head over five years.
- [ ] India: check GST working-capital drag, EPF/ESI and minimum-wage exposure, Labour Code implementation status, contract-labour compliance and the contingent liabilities note for statutory claims.
- [ ] Peer set: same family, same sub-model, same top-line convention, same labour regime, same lease standard, same cycle position — stated explicitly.
add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall