How to Approach Guesstimates

A semi-guide to solving these engaging questions.

Guesstimates have now become an important part of several interview processes and B-School competitions. They are also a lot of fun!

When I first encountered guesstimates and looked them up on the internet, it wasn’t easy to find a lot of solutions. For someone like me who loves to learn by observing, this isn’t very helpful. In the upcoming weeks, I’ll upload some of the approaches taken by me in competitions while solving these guesstimate questions.

The process of solving a guesstimate problem is pretty simple: Look at the possible parameters that may affect the final quantity, and estimate its numbers. Confusing? Let’s look at an example —

Q. Guesstimate the quantity of chocolate (in all its forms) used in Delhi in a day.

Step 1 : The Umbrella

There are multiple approaches to this problem:

1. Estimate the consumption side: end customer, grams per person etc.

2. Estimate the production side: bakeries, chocolate factories in Delhi or the estimated consumption of chocolates in the country and extrapolate the consumption of Delhi by allocating it an appropriating share based on the population, income etc with respect to the rest of the country.

Which umbrella do you choose? The answer should depend upon how easy you think it is to guess/obtain the amount for the parameter by general observation. The goal is NOT to reach the exact answer. The goal is to develop the approach to such problems. However, there exists a dilemma here:

1. Use assumed parameters? (eg : consumption per person)
2. Use realistic parameters that are possible to obtain? (eg : consumption of chocolate in the country)

From a competition perspective, the first approach would be more advisable. A three-level classification is usually considered good enough and this approach gives you plenty of room to do just the same. The second approach makes more sense in case you’re doing it for research purposes. 

Personally, I find the first approach more fun and engaging.

Step 2 : The Assumptions

So far, so good. The next step is simply to divide these categories further and allocate them appropriate numbers. We can break down the ‘End Customer’ in the following ways:

1. Age Group
2. Heavy/Medium/Light consumers
3. Diabetic and Non-diabetic and so on.

Remember that each of the above would have further classifications which could be any of the others.

Step 3 : The Numbers

After you’ve chosen your Umbrella and your Assumptions, it is time to start the guesswork. Apply your general awareness of the environment you’re dealing with to come up with numbers. The golden rule is to use beautiful numbers. Sure, you may get the exact population of Delhi as 18,686,902 with a simple Google search but it sure doesn’t sound as pretty as 20,000,000. How many people to assign to different age groups? Work in percentages. We’re a young country, so the maximum weightage goes the 15–40 year group. Pretty simple, right?

Over the course of next few weeks, I’ll be uploading some solutions to different guesstimate questions such as “Guesstimate the number of Whatsapp messages sent in NCR everyday”.

Liked this semi-guide to guesstimates? Could be better? Got a guesstimate question you want solved? Let me know in the comments!

Read more on Medium.