The sector playbook

Part Eight has argued one thing in many forms: the financial statements are a record of the past, and for every business there is an earlier, public, upstream signal that moves before them. This closing module asks the question that turns that argument into a tool you can actually pick up on a Monday morning - for the sector in front of me, which signal, where do I find it, how far ahead does it fire, and what is the one way it lies?

That question has no single answer, and the absence of a single answer is the point. There is no universal leading indicator. The signal that leads a cement maker's result by three years is useless for an IT firm, whose lead time is a single quarter; the number that warns you about a bank is silent for a developer, whose own reported revenue is the least useful line in its accounts. Foresight is not one rule applied everywhere. It is a small shelf of playbooks, one per sector, each naming a different leading indicator that the corresponding lagging statements will only confirm quarters later.

This module hands you that shelf - a one-page playbook for the major Indian sectors - and, more importantly, the routine for building the pages you will still need for the sectors it does not cover.

The temptation, after a part full of forward signals, is to compress them into a single maxim - "watch volumes", "watch capacity", "watch cash". Every such maxim is right in one sector and actively misleading in the next. Watch capacity, and you read cement and capital goods well and telecom badly. Watch collections, and you nail real estate and misjudge a software firm that has no collections problem to speak of. The unit of foresight is the sector, because the thing that leads a result is a fact about how that specific business turns effort into revenue - how long its cash-to-result cycle is, where its future is committed before it is earned, and which of its numbers management controls versus which an outside party publishes.

There is a deeper reason this belongs at the end of Part Eight rather than the start. The reported result, when it finally lands, is the noise the crowd reacts to; the upstream signal that led it by quarters is the signal a reader could have acted on. A per-sector playbook is simply the organised habit of watching the second and not the first - and because the routine is repeatable and the same each quarter, it is a process you can run regardless of how any single quarter turns out, not a bet that happened to work once.

Strip the playbook to its skeleton and every sector page has the same four fields:

• The signal - the one or two most-forward, most-public metrics that move before this sector's result. Not every operating number; the earliest one with real lead.
• The source - exactly where it is published, and by whom. A regulator, an exchange, a government portal, an industry body, or the company's own disclosure - ranked, always, by distance from the company's control.
• The lead - the rough gap between the signal moving and the result confirming it. This is what you are buying: the window in which the future is legible and the headline is still blank.
• The trap - the one way this particular signal misleads, so that reading it mechanically does not simply swap one error for another.

The reason the fields are worth tabulating is that the lead differs enormously by sector, and the difference is the whole value of the exercise. A developer's pre-sales and collections lead its completion-based revenue by years; an automaker's dispatch-versus-registration gap leads its result by barely a quarter. Knowing which you are dealing with tells you how early you can act and how long you must be willing to wait to be proved right.

The ladder is deliberately uneven. It is not a claim that longer leads are better - a two-quarter warning you actually act on beats a three-year one you sit through and rationalise away. It is a map of how much patience each sector demands of a reader who wants to be early, and how quickly each will punish one who is late.

Here is the shelf. Each card is one sector's page - its forward signals, where they live, the rough lead, and the trap. Read across them and the inversion that runs through the whole book is unmistakable: there is no metric that means the same thing twice. Capacity announcements are a warning in cement and a promise in a growth industry; a full order book is comfort for a contractor and irrelevant to a developer; the reported revenue line is the answer in IT and the least useful number in real estate.

The cards are compressed on purpose; each unpacks into the full sector module behind it. But even at this density the routine is visible. Pick the sector, read its one or two forward signals from the source furthest from the company's control, note how far ahead they fire, and hold the trap in mind so you do not read the signal mechanically. That is the entire method, and it is the same shape whether the sector is on this page or one you have to build yourself.

The one-page summary makes the four fields line up at a glance:

Take a composite mid-cap automaker, Suraksha Motors illustrative, reporting a strong quarter: dispatches up 14%, revenue up 12%, margins holding. [illustrative] The reflex is to read the result and move on. The routine says: this is an autos name, so open the sector page. The forward signal is dispatches read against retail sell-through, and the source furthest from the company's control is the government's VAHAN registration portal - a dataset Suraksha does not write. VAHAN shows registrations of its models up just 2% over the same period. The 12-point gap is the tell: the dispatch growth is stock being pushed onto dealers, not vehicles reaching buyers, and it will reverse next quarter as dealers stop taking inventory. The reported result is fine; the forward signal, read from the independent source, says the next one will not be.

Now change the sector and watch the page change with it. A composite cement maker, Deccan Cement illustrative, reports the same kind of strong quarter - realisation and margin at multi-year highs. [illustrative] The autos routine is useless here; there is no VAHAN for cement. The cement page tells you to read industry capacity against demand, and the disclosures show three regional players commissioning large new lines over the next two to three years into a region already running near full utilisation. The peak margin you are looking at is not the durable state; it is the top of a cycle whose supply side has just turned. Same-looking result, opposite forward read, because the signal, the source and the lead all changed with the sector. That is the shelf doing its job - and it is exactly the pass Module 119 will teach you to run in two hours on a company you have never seen before.

The playbook sharpens the question; it does not answer it. Three limits are worth holding as firmly as the method itself.

It cannot give you timing or magnitude. A forward signal tells you the direction the result is likely to move, not when the market will agree or by how much it will reprice. A capacity glut can take years to bite; a slippage trend can be papered over for several quarters. Being early and being wrong feel identical while you wait, and the signal does not carry a date.

It cannot survive a structural break in its own lead time. Every entry on the shelf rests on a historical relationship - dispatches leading demand, order books leading revenue, approvals leading launches. A change in the industry's plumbing can void that relationship: a shift to direct-to-consumer selling weakens the dispatch signal, a regulatory overhaul rewrites what a tariff order means. The playbook is a set of empirical regularities, not laws, and part of the routine is asking whether the regularity still holds.

And a single forward signal is a question, not a verdict. One soft print of registrations, one quarter of order-book weakness, is a prompt to look harder, not a conclusion. The confident read is a cluster - the forward signal, the coincident numbers and the balance sheet pointing the same way - which is why the sector pages list two or three signals, not one, and why the routine reconciles across sources rather than trusting any single feed.

The first trap is using the wrong sector's page. Reach for the metric that worked last time - volume, capacity, collections - and apply it to a business whose future is committed through a different channel, and you will stare intently at a link that was never going to fail while the real one snaps. The playbook fails silently when you open it to the wrong page.

The second is reading the signal from the company's own mouth. Many forward signals exist in two versions: the flattering one the company narrates and the harder one an outside party publishes. Dispatches versus registrations, guided order pipeline versus filed tender awards, "robust demand" versus VAHAN - the value is in the independent source, and a reader who takes the company's version has kept the label and lost the signal.

The third is reading the signal mechanically and skipping the trap. Every card has a trap because every signal lies in a specific way: a full order book won at loss-making bids, a busy approval pipeline losing to price erosion, a low credit cost sitting on rising slippages. Read the signal without its trap and you have simply swapped a lagging error for a leading one.

The fourth is mistaking public for priced-in, and giving up. Because the sources are public, it is tempting to conclude there is no edge. But public data is read inconsistently, acted on slowly, and abandoned by most participants long before its lead time plays out. The edge was never secret information; it is the discipline of running the same routine every quarter and the patience to sit with a signal the crowd has not yet respected.