How to Analyse a Logistics Stock (Density, Mix, Realisation)

In the December 2024 quarter, Delhivery's part-truckload business — its fastest-growing line — earned a Service EBITDA margin of just 3.8% (Inve data, Q3 FY25). Nine months later, the same business had climbed to 8.5% (Inve data, Q2 FY26), with management openly aiming at a 16–18% goal once the network fills (Delhivery Q2 FY26 concall). The trucks did not get cheaper. The lanes did not change. What is moving that margin up the curve is a single number management almost never puts on the income statement: monthly load. At roughly 150,000 tonnes a month the network was running half-empty; management said it needed "close to about 200,000 to 215,000 tonnes of monthly load" before "three things will kick in" and margins stepped up to that 16–18% target (Delhivery Q1 FY26 concall). (Illustration of how to read the numbers, not a view on the stock.)

That is the whole game in logistics, compressed into one company-quarter. A logistics network is mostly fixed cost — hubs, sortation lines, a fleet, a workforce that shows up whether the truck is full or not. The profit does not come from the freight; it comes from the density of freight moving over a network that was already paid for. Two operators can run identical trucks on identical routes and one earns 16% while the other earns 4%, purely because one fills the network and the other does not. Strip away the jargon and a logistics company is a half-empty restaurant: the rent, the kitchen and the chef are paid for the night; every extra table seated drops almost straight to the bottom line, and every empty one bleeds.

This is how to read a logistics stock the way someone who understands operating leverage does: the handful of operating numbers that decide the outcome — most of them buried in investor decks and concalls, not the P&L — what "good" looks like for each, and the one red flag that flatters a logistics company right up until the cycle turns.

A boundary first, said plainly: you cannot rebuild a sortation network's unit economics from the outside. What you can do is read volume, realisation, mix and density together, and check whether the operating leverage management keeps describing is actually showing up in the segment margins.

What actually drives the economics here?

A logistics business sells movement — a parcel from a seller to a buyer, a container from a port to an inland yard, a pallet across the country. It charges a price per unit moved (realisation) and incurs a cost to move it. But the cost is overwhelmingly fixed in the short run: the hub is built, the sortation machine is bought, the line-haul truck runs on a schedule. So the marginal cost of one more parcel through an existing network is tiny, and the marginal revenue is close to the full price. That gap is operating leverage, and it is the entire reason logistics margins swing so violently with volume.

This fixes two beginner errors. First, revenue growth tells you almost nothing on its own — a logistics company can buy volume by cutting price, growing the topline while margins go backwards. Second, a bad-looking margin is not the same as a bad business — a sub-scale network running at 4% can be one density step away from 16%, while a mature network at 16% may have no leverage left to give. The question is never "what's the margin?" but "is the network filling up, at what price, and who's paying for the assets?"

The metrics that matter (and where they hide)

Most of these are not in the income statement. The P&L gives you revenue, EBITDA and PAT. The metrics that explain why those numbers are what they are — and predict where they go next — live in the investor presentation and the concall transcript. That is exactly why most retail analysis of logistics stocks is shallow: it reads the P&L and stops. Other listed operators worth running through this same lens include Blue Dart Express, VRL Logistics and Mahindra Logistics.

1. Volume / tonnage — the engine, before price flatters it

Volume (parcels, tonnes, or TEUs — twenty-foot container units) is the real measure of the business, because revenue can be inflated by price and deflated by competition. Read volume growth separately from revenue growth: if revenue grows 12% but volume grows 20%, realisation fell 8% — the company bought share by cutting price.

Where to find it: occasionally a line in the deck, more often spoken on the call. Delhivery moved 295 million express shipments in Q3 FY26 (Inve data, Q3 FY26) and its PTL freight tonnage grew ~20% (Inve data, Q3 FY26). CONCOR handled 4.15 million TEUs in the first nine months of FY26 (Inve data, Q3 FY26). TCI Express carried 255,000 tonnes in Q3 FY26 against a full-year FY25 of 995,000 tonnes (Inve data). Good is volume growing faster than the addressable market while realisation holds — share gain with pricing discipline, not instead of it.

2. Realisation — the price the network commands

Realisation is revenue per unit moved — per parcel, per tonne, per TEU. It is the discipline metric. A network filling up on volume and holding or raising realisation is winning; one filling up only by discounting is renting share it will lose the moment a rival blinks.

Where to find it: almost never reported directly — you usually back it out (segment revenue ÷ segment volume) or catch it in management's price-hike commentary. TCI Express, the most pricing-disciplined of the express names, has been pushing near-term yield up ~1–2% a year and planned a roughly 3% price hike for FY26, with an aspiration to lift realisation ~5% by FY28 (Inve data) — and crucially has done it while volume still grew ~8% (Inve data, Q2 FY26). Good is realisation that rises in line with cost inflation or better; the warning is volume booming while realisation quietly slides, the signature of a price war.

3. Network — hubs, fleet, sortation, lead distance

The network is the fixed-cost base that makes density possible: hubs, automated sortation centres, the line-haul fleet, and — for rail — rakes and terminals. More importantly, it is what competitors must replicate to compete, so it is the closest thing to a moat in a business with almost none.

Where to find it: the investor deck is the only real source. CONCOR ran a fleet of 413 rakes and was targeting 100 terminals by 2028 (Inve data, Q3 FY26) — physical assets a new entrant cannot conjure. Its lead distance (average haul per container) matters because rail economics improve with distance: EXIM lead distance was ~701 km and domestic ~1,317 km (Inve data, FY25 / 9M FY25). Good is a network expanding ahead of volume in the right lanes, with utilisation rising — not capacity built on hope and left idle.

4. Express vs FTL mix — the margin ladder

Not all freight is equal. Full-truckload (FTL) is a low-margin, fragmented, near-commodity business — you book a whole truck, point to point. Less-than-truckload and express part-truckload (PTL) consolidate many shippers' parcels through a hub-and-spoke network: higher value-add, higher margin, and far harder to replicate. As a logistics company's mix shifts toward express/PTL and multimodal, its structural margin should rise.

Where to find it: the mix slide in the deck and the segment-margin disclosure. TCI Express runs ~81–82% surface express (Inve data, Q1/Q3 FY26) at a ~13% margin (Inve data, Q1 FY26), with multimodal (rail, air, cold chain) carrying ~14% EBITDA margins and growing ~14% (Inve data, Q1 FY26) — a deliberate climb up the value ladder. Good is a rising share of the higher-margin, harder-to-copy mix; a slide back toward commodity FTL to chase volume is the opposite of progress.

5. Asset-light vs owned — who carries the capex risk

Two operators can run the same network with opposite balance sheets. An owned model buys the trucks, builds the hubs and carries the depreciation and the capital; an asset-light model leases capacity from third-party fleet owners and warehouse developers, trading a slice of margin for far lower capital intensity and faster flex. Neither is "better" in the abstract — but they value completely differently, and the asset-light operator can scale without diluting or levering up.

Where to find it: capex-as-percentage-of-revenue and the lease/own commentary in the deck. Delhivery's owned-network model guides to long-term capex of 3.5–4% of revenue (Inve data, Q3 FY25) and targets ~3x asset turns with return on capital "well above 24%" on its Express and PTL businesses (Delhivery Q1 FY26 concall) — owned, but engineered for high asset turns. TCI Express runs a more asset-light surface model with fleet utilisation at 83% (Inve data, Q3 FY26). Good is high asset turns and a capex line that funds growth without serial equity raises; the warning is capital intensity creeping up while returns on it do not.

6. Utilisation / density — the number that decides the margin

This is the one most investors never find, and it is the one that matters most. Utilisation is how full the network is — load factor on trucks, throughput per hub, occupancy of warehouse space. Because the cost base is fixed, the margin is largely a function of utilisation. Below the density threshold the network bleeds; above it, incremental volume is almost pure margin.

Where to find it: spoken on the call, occasionally in the deck, essentially never on the P&L. Recall the opening number — Delhivery's PTL margin walked from 3.8% to 10.7% to 8.5% (Inve data, Q3 FY25 → Q1 FY26 → Q2 FY26) as monthly load climbed toward the ~200,000-tonne threshold management had flagged (Delhivery Q1 FY26 concall), still part-way up the curve toward the 16–18% the network is built to earn at full density. For CONCOR the equivalent is rail coefficient — the share of port cargo that moves by rail rather than road, currently ~18–20% with management projecting 35–40% once the dedicated freight corridor is fully ramped (Inve data, Q1 FY26): every point of rail-share gain is volume flowing onto a fixed rail network. Good is utilisation visibly rising quarter on quarter with margins following; the danger is a network expanded faster than it can be filled.

The KPI Screener lines these operating metrics — volume, segment margins, capex intensity, utilisation — across logistics names with a per-number trend and confidence flag, so you can see the density step happening rather than waiting for it to surface in PAT.

How do you value a logistics stock?

Use EV/EBITDA, with P/E as a cross-check, and almost never a bare P/E in isolation. Two reasons. First, logistics is capital-intensive and depreciation-heavy in the owned models, so EBITDA strips out the financing and depreciation choices that make two similar businesses look different on net profit — and EV captures the debt an owned-fleet operator carries. Second, and more important, the multiple has to be read against where the company sits on its own operating-leverage curve.

This is the trap. A logistics stock filling up its network looks expensive on trailing EBITDA precisely because the EBITDA hasn't arrived yet. Delhivery's PTL business on Q3 FY25 economics (3.8% margin) would have screamed "overvalued" on any trailing multiple — and an investor who shorted that math would have watched the margin more than double over the next nine months, to 8.5%, as density began to kick in (Inve data). The opposite trap is just as real: a mature network already at peak utilisation can look cheap on trailing EBITDA while having no leverage left, so the cheapness is a value trap, not a bargain.

Now the symmetric warning, because the inversion cuts both ways and the dangerous half is the one that flatters you. A low multiple on a "filling" network is not, by itself, a margin of safety. The network that fills up is the survivor you read about; the network that never reaches density — stranded at 4% while the fixed costs run — is the far more common outcome, and it leaves no headline behind. A low multiple justified only by density that has not yet arrived is a thesis, not a margin of safety: you are paying a real price today for operating leverage that is contingent, not contracted. This article's own evidence makes the point — Delhivery's PTL margin went 3.8% → 10.8% → 10.7% → 8.5%, round-tripping down even as load inched up. If a path that non-monotonic can sit under a thesis you call "cheap on normalised EBITDA," your margin of safety has to come from the price you pay, not from assuming the density arrives on schedule.

So the right question is not "what's the EV/EBITDA?" but "what EBITDA will this network throw off when it's full, and how far is it from full?" Normalise to mid-cycle, density-adjusted EBITDA — then judge the multiple. CONCOR, trading on a ~22% EBITDA margin (Inve data, Q3 FY26) with a ~55% market share (~53–55% EXIM, ~56–58% domestic; Inve data, Q3 FY26) and a fixed rail network gaining share, is a different valuation animal from a sub-scale express player whose margin is a bet on volume that may never come. Same lens we apply to cyclicals — normalise before you multiply; the logistics version is "normalise for density."

A worked case: density, said and done

Take Delhivery's PTL business through FY25–FY26 — not as a stock view, but as the cleanest live example of reading operating leverage before it prints (illustration, not a view on the stock; figures from Inve data unless a concall is cited).

Read the said against the done — and read it sceptically. In December 2024, with PTL at a 3.8% margin, management was not promising magic — it was describing a network running roughly a third below the density it needed, and telling you, with a specific tonnage number, exactly what would change when the load arrived. By June 2025 it laid out the threshold in plain figures: ~150k tonnes today, ~200–215k needed (Delhivery Q1 FY26 concall). By September 2025 the load had reached only ~159k tonnes a month and the margin had moved to 8.5% (Inve data, Q2 FY26) — still well short of the ~200k threshold, and so still climbing the curve rather than arrived at the 16–18% goal.

But notice what the margin column actually did: 3.8% → 10.8% → 10.7% → 8.5%. It did not march monotonically up with load. Between Q4 FY25 and Q2 FY26 monthly load barely moved (~153k → 159k) while the margin round-tripped down from 10.8% to 8.5%. So the threshold management named is a hypothesis, not a law: the load↔margin map is noisy, and mix, seasonality and one-off costs move the margin as much as raw tonnage does. A reader who took "200k tonnes = 16–18%" as a mechanical formula would have been wrong-footed by the very quarters in this table. And the Q2 FY26 8.5% you are told to verify is not even a clean read: Delhivery consolidated the Ecom Express acquisition (₹1,400 crore, effective 18 July 2025) into that September quarter, adding volume and integration cost that muddy the load-to-margin comparison with prior quarters.

The honest reading, then, is narrower than "trust the guidance" — the guidance has not landed, and the path to it is not a straight line. It is this: in a fixed-cost network the operating threshold is at least knowable in direction, and the manager who hands you the tonnage number is handing you the model, so you can check the margin against the load yourself. And you must, for a specific reason: management's future-density guidance is the most incentive-loaded number in the deck — the one figure they are paid to make sound inevitable — which is exactly why you read load against margin rather than taking the 16–18% goal on faith.

This is also where a longer record earns its keep. A single confident call about future margins is cheap; a sequence — load rising, then the margin following the path management mapped — is what tells you the leverage is real and not a slide-deck aspiration. Inve's Promise Tracker pins each forward commitment to the quarter and quote it was made in, so when a logistics management says "16–18% at full density," you can later see whether the density and the margin actually arrived, rather than re-reading four transcripts by hand.

The contrast that teaches: filling vs owning the lane

Put Delhivery's express network next to CONCOR's rail franchise and the two ways a logistics company makes money come apart. Delhivery wins by filling a network it must keep dense to defend — its PTL margin (8.5% in Q2 FY26, climbing toward a 16–18% goal) is earned quarter by quarter by load (Inve data, Q2 FY26). CONCOR wins by owning the lane: a ~55% market share (Inve data, Q3 FY26), 413 rakes (Inve data, Q3 FY26), and a structural tailwind as the dedicated freight corridor lifts rail coefficient from ~18–20% today toward a projected 35–40% (Inve data, Q1 FY26). Its ~22% EBITDA margin (Inve data, Q3 FY26) is less a density bet than a position.

But even the owned franchise has its hidden bill, and it shows the value of reading past the P&L. CONCOR pays a Land Licence Fee to the railways for its terminal land — a large fixed land-licence-fee cost of a few hundred ₹ crore a year (₹370 crore in FY25; Inve data, FY25). That charge does not flex with volume: it is a fixed claim on the business whether or not throughput grows, and the year-to-year amount has moved with land surrenders and adjustments rather than tracking traffic. Miss it, and you misjudge the operating leverage of the very franchise that looks safest.

Red flags specific to logistics

• Revenue up, volume up more. Realisation is silently falling — the company is buying share with price. In a fixed-cost, price-competitive business, this is how a "growth" story turns into a margin sink.
• Capacity built ahead of volume, with no density path. Hubs and rakes added on hope. A network expanded faster than it fills carries the full fixed cost with none of the leverage — the inverse of the density story, with the margin going down the curve. This is the case the examples never show you, because survivors write the case studies. For every operator whose network filled, there is a longer, quieter list — venture-funded express and quick-delivery networks that scaled hubs and fleets faster than parcels arrived, then watched a rival cut price to buy the same volume, and never crossed the density line before the cash ran out. The graveyard of Indian logistics is large and mostly un-listed: race-to-the-bottom freight pricing has stranded many a half-empty network. So when capacity is racing ahead of volume, treat "the density will come" as the claim that needs proving — ask what specifically gets this network full before the funding or the patience does.
• Margin improvement that's all cost-cutting, no density. Genuine logistics operating leverage comes from more freight over the same network. A margin propped up by deferring maintenance or starving the network of capex borrows from next year.
• Asset-light claims, asset-heavy balance sheet. A company calling itself asset-light while capex-to-revenue and gross debt climb is telling you a story the cash flow contradicts — the same gap between reported profit and real cash that catches out any capital-intensive business.
• One subsidising segment. A profitable core quietly funding a loss-making new venture (express, quick-commerce, a fledgling line) can hide deterioration in the part you're actually buying. Read the segments, not the consolidated EBITDA.

Frequently asked questions

The discipline comes down to refusing to read the P&L alone. The profit line is the output of density, mix and realisation — by the time it moves, the operating numbers that caused it have already spoken, one or two quarters earlier, in the deck and on the call. So invert the question you bring to a logistics company's results. Don't ask "was this a good quarter?" Ask: if this network were filling up but the company were buying that volume by cutting price, what would the numbers look like — and does the mix of volume, realisation and segment margin rule it out? Volume booming while realisation slides does not rule it out; it is the pattern itself.

One honest caveat about this whole lens, so you don't over-trust it. Density is the right frame — but a frame is not an edge. The base rate it operates against is low: Indian logistics runs on cheap, fragmented roads, a structural oversupply of trucks, and an FTL segment locked in a race to the bottom, so most networks that could benefit from density never get the pricing power to reach it. The lens reliably tells you what would have to be true for operating leverage to arrive; it does not tell you that you will find a company where it does. Use it to disqualify the obvious traps and to know which number to track — not as a promise that a winner waits at the end of the curve.

And the owner's question, to sit with before buying any logistics stock: what must I believe about density — that this network gets full, in the right lanes, at a price that holds — for the operating leverage to arrive? If the answer leans on management's confidence rather than the load actually climbing toward the threshold they named, you've read the slide deck, 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