NerdWorld Report

The Tire Question
Interviewing Young Engineers

Published September 20, 2025

SUMMARY: Gain immediate and deep insight into the problem-solving capability of young engineers with a simple interview question. Can they logically solve problems? Can they make realistic estimates quickly? Are they sloppy thinkers?

Background: In the early 2000's, an engineer who worked for me was interviewing a candidate for one of our engineering intern positions and asked this question. I thought the question was brilliant, and I've been using it ever since.

The Question:

How many passenger car tires were sold in the United States in the previous calendar year?

I provide the candidate with a calculator, pencil, and paper. I tell them that I understand that this is outside their knowledge base, but that I need a numerical answer within the next 5 minutes.

There are 3 classes of answer:

They make up a number off the top of their heads, thinking I'm an idiot for asking the question.
They say they don't know and refuse to give an answer. They explain that they don't want to give a false answer, or provide information that they are not certain of.
They make some type of estimate of the number, based on general knowledge. [e.g.: 4 tires per car; 350 million Americans, 25% of whom own a car; tires last about 4 years; etc; etc]

The first answer is the worst. When under pressure, a good engineer will NEVER give unreliable information without first undergoing some intellectual process to validate the answer. This candidate is likely to be unreliable in problem solving.

The second answer is better, but not optimal. We appreciate that "no information" is better than "bad information", but engineering is the art of estimating from incomplete data. This candidate will probably be a "plodder", which is excellent if absolute accuracy and certainty is required, but is less useful in a dynamic environment.

The third answer is the best. A good engineer should be able to estimate almost anything, based on their current knowledge. However, they must also be cognizant of the limits of their own estimate. So I would expect this answer to include some weasel words such as "this is just a very rough estimate" and "more research is required to validate this answer".

Postscript:

The best answer I ever got was when the candidate took a moment to estimate a numerical answer during the interview. The next day, the candidate then sent me a thank-you eMail that included the actual number, with a link to an internet article with the information. I hired that guy.
I love interesting interview questions. Another one of my favorites includes "Are you a 'words' person or a 'numbers' person?"