Case 049Forecasting and modellingCore

Two-hour written test: build a first view on an unfamiliar industry from a data pack on Kestrova Cold Chain. Industry capacity is 40 million tonnes growing 12%, Kestrova's share is 3% and its utilisation 65%. Write five questions you would ask and a first forecast.

The situation

You are given a two-hour written test and a data pack on cold storage, an industry you have not covered. Industry capacity is 40 million tonnes, growing about 12% a year, and industry utilisation is about 70%. Kestrova Cold Chain owns 3% of that capacity, 1.2 million tonnes, and runs at 65% utilisation. It earns about Rs 30,000 a year for each tonne of space actually used, so revenue is about Rs 2,340 crore.

Kestrova has announced 0.15 million tonnes of new capacity a year for the next three years. The pack also lists customer segments, power costs and a few recent contract prices, but no guidance.

Your task

In a timed test, how do you structure the first view, which five questions do you write down, and what is your first three-year revenue forecast?

Quick check

You have two hours and a thick data pack. What do you do in the first fifteen minutes?

Worked solution

Try it on paper, then open one step at a time.

30-second answerThe answer to give first

Set up the question tree before reading the pack: revenue is capacity times utilisation times price, so every question should attack one of those three. The two that matter most are whether demand keeps pace with 12% capacity growth, and why Kestrova runs at 65% when the industry runs at 70%. A first forecast with 0.15 million tonnes added a year, utilisation rising to 70% and rents up 4% a year takes revenue from about Rs 2,340 crore to Rs 3,898 crore in three years.

Step 1Why set up the questions before hunting for answers?

A student with two hours to write about a city she has never visited does better to list what she needs, where people live, how they travel, what they earn, than to read the guidebook from page one. In a timed test, the structure is the answer's skeleton: it tells you which pages matter, and it shows the examiner you know what drives the business before you know the business. For a warehouse business the skeleton is simple. Revenue is space owned, times how much of it is used, times the rent per tonne used.

Set up the questions before hunting for answersWhat will Kestrova earn in three years?Industry40 mt of capacity, growing 12%industry utilisation 70%Kestrova1.2 mt, 3% share, 65% utilisedRs 30,000 a utilised tonne a yearQ1Is demand growingas fast as capacity?Q2Why 65% when theindustry runs 70%?Q3Rents per tonne:rising or falling?Q4Who are the clients:pharma, dairy, farm?Q5Cost of a new tonne,and what it earns?Highlighted: the two questions that move the forecast most. Answer them first, then fill in the rest.
The research plan splits Kestrova's three-year earnings into industry and company drivers and hangs five open questions beneath them; the first two, demand against capacity and Kestrova's utilisation gap, move the forecast most.
Step 2Which five questions would you write down?
QuestionWhy it mattersWhere to look in the pack
1. Is demand growing as fast as capacity?If capacity grows 12% and demand 8%, utilisation and rents fall across the industryIndustry utilisation trend, not the capacity table
2. Why is Kestrova at 65% against 70%?New sites still filling is temporary; poor locations are permanentUtilisation by site and site age
3. Are rents per tonne rising or falling?Price is a third of the revenue equationRecent contract prices against older ones
4. Who are the customers?Pharma pays more and is steadier than seasonal produceCustomer segment split
5. What does a new tonne cost, and what does it earn?Growth only creates value if new capacity earns above its cost of capitalCapex per tonne, power cost per tonne
Kestrova's five open questions each attack one driver of revenue or value, and each points to the page of the data pack that would answer it.
Step 3What is the first forecast?

Make each assumption explicit and tie it to a question. Capacity rises by the announced 0.15 million tonnes a year to 1.65 million, which keeps Kestrova's share near 3%. Utilisation climbs from 65% to 70% as new sites fill, which assumes question 2 has a temporary answer. Rents rise 4% a year, roughly with inflation. Revenue goes from Rs 2,340 crore to about Rs 2,822, 3,358 and 3,898 crore, about 19% a year, faster than the industry because utilisation is catching up.

A first forecast from the data pack, Rs crore2,340Today1.20 mt at 65%2,822Year 11.35 mt at 67%3,358Year 21.50 mt at 69%3,898Year 31.65 mt at 70%Capacity +0.15 mt a year, utilisation 65% to 70%, rent per tonne +4% a year: each assumption is one of the five questions.
Adding 0.15 million tonnes a year, lifting utilisation from 65% to 70% and raising rents 4% a year takes Kestrova's revenue from Rs 2,340 crore to Rs 3,898 crore in three years, about 19% a year.

Finish the written answer with what would change the forecast most. If question 1 comes back badly, with demand growing only 8%, industry utilisation slides and Kestrova's catch-up to 70% becomes unlikely. That single sentence tells the reader which assumption carries the forecast, which is what an examiner marks in a first view.

Where candidates lose it

The common loss is spending ninety minutes summarising the data pack and fifteen on a view. The test is marked on the questions and the forecast, not on how much of the pack you copied out.

The second is forecasting from the 12% capacity growth alone. Capacity is supply; without asking whether demand keeps up, the forecast assumes the one thing most likely to be wrong in a fast-building industry.

What the interviewer asks next

  • The pack shows power is 35% of operating cost. How would a 10% power tariff rise change your view?
  • How would you check whether 12% capacity growth is creating oversupply?
  • What would a comparable listed company need to look like for its multiple to be useful here?

Company names and figures are illustrative.