Guidance versus delivery

Every results season, managements put a number on the future: revenue up so much, margins to a band, capex of so many crore. The market treats these numbers as information. They are — but only in proportion to what the same management's past numbers turned out to be worth.

This is the promise ledger of Module 067 applied to one specific promise: forward growth guidance. The question is not "what did they guide?" but "what has their guidance been worth?" — because a forecast from a management that has hit its numbers for five years is a usable input, and the identical forecast from a serial over-promiser is a number to discount, sometimes to nothing.

A forecast has no intrinsic reliability; it inherits it from the forecaster. The only way to know how much to trust a management's guidance is to score how its past guidance turned out — the guidance hit rate. A management that guides sensibly and delivers has earned the right to have its next number taken near face value. One that guides high and misses has told you, year after year, to mark its forecasts down.

There is an honesty asymmetry worth naming. A management that consistently guides a little low and beats — the sandbagger — is annoying but safe: its real number is above its word. A management that guides high and misses is the dangerous one: its real number is below its word, and if you take the guidance at face value you will overpay for growth that does not arrive. The two look opposite on the surface and must be read in opposite directions.

The method is mechanical and worth doing by hand once for any company you follow:

• Extract dated guidance. Pull the specific forward numbers management gave — growth, margin, capex, a segment target — with the date and horizon attached. The concall and the annual report are the sources.
• Match to actuals. When the period closes, put the delivered number next to the guided one. Hit, modest beat, modest miss, or wide miss.
• Score the pattern over 3–5 years. One miss is a cycle; a pattern is a signature. Look for the direction of the error — chronic overshoot (over-promiser), chronic undershoot (sandbagger), or centred and tight (credible).
• Weight the next number by the record. Take a credible forecaster's guidance near face value; discount a serial misser's toward its historical delivery, not its stated ambition.

Two behaviours deserve special attention. The sandbag (guide low, beat) is benign but tells you the real trajectory is above the guidance. The quiet lowering — a full-year target walked down mid-year without being flagged — is the loudest negative signal, because it reveals both that the original was unreliable and that the management would rather you did not notice.

Take a composite company, Nandan Industries illustrative, whose management guides each year to "18–20% revenue growth." [illustrative] Build the ledger from five years of concalls: delivered 11%, 9%, 13%, 10%, 12% against guidance that was always 18–20%. The pattern is unambiguous — a consistent, wide overshoot in the guidance, with real delivery clustered around 11%. So when Nandan now guides to 19% again, the honest input is not 19%; it is closer to the ~11% it has actually managed, unless something structural has changed to justify the gap closing.

Contrast a second composite, Sarita Consumer illustrative, that guides to "low-teens growth" and delivers 13%, 14%, 12%, 15%, 13%. [illustrative] Its guidance has been worth roughly its word for five years — a credible forecaster. When Sarita guides low-teens again, you can take it near face value. The two managements might guide the identical number next year; the record tells you to treat one as information and the other as aspiration.

The tell that would downgrade even Sarita: a full-year target quietly restated lower at the half-year, with no acknowledgement that it moved. That single act converts a credible forecaster into a suspect one, because it shows the guidance bends to protect the narrative.

The same miss is forgivable in one sector and damning in another, because it depends on how much the management actually controls delivery. In a predictable, annuity-like business — consumer staples, a regulated utility — outcomes are largely within management's hands, so a wide guidance miss is a real failure of either honesty or competence. In a deeply cyclical business — commodities, and to a degree lenders — the cycle can overwhelm any plan, so a miss driven by a price crash the management did not cause is a weaker mark against its credibility. Grade the miss against the sector's controllability, not on a single universal scale.

A guidance-versus-delivery record is backward-looking, and it carries the usual limit: a management can break its pattern. A credible forecaster can hit a genuine shock and miss; a serial over-promiser can bring in new discipline and start delivering. The record tilts the odds; it does not fix the future.

It also cannot separate honesty from competence on its own. A management that misses might be dishonest (guiding high to prop the price) or merely bad at forecasting its own business — different problems needing different weights. The pattern flags the miss; the concall's tone and the reasons given help tell which it is.

And a clean record does not guarantee the level is right — a management can honestly and consistently guide to a growth rate that is itself too low or too high for the opportunity. Credible guidance is a reliable input, not a complete valuation.

The first trap is taking guidance at face value regardless of record. The market routinely models a serial over-promiser's next number as if it will be delivered, then treats the inevitable shortfall as a fresh disappointment — when the record said all along the number was aspirational. The stated guidance is the noise; the historical delivery is the signal.

The second is punishing the sandbagger and rewarding the over-promiser. A conservative management that guides low and beats can screen as "unambitious," while a management that guides high screens as "a growth story" — exactly backwards, because the first under-promises and over-delivers and the second does the reverse.

The third is missing the quiet revision. A target walked down mid-year without acknowledgement is the single most informative event in the guidance record, and it is designed to slip past. An investor who only compares the final actual to the revised guidance — not the original — will score a miss as a hit and never see the credibility problem.