Market Sizing

From experience (and from interviewing candidates), I'll say the following points:

• Top Down is usually easier, and works well for the majority of market sizing questions (though not for all, I admit):
  • Yes, your assumptions are a little harder to justify, but just go with them until you have a result you can sense check down the line, then go back and revise your assumptions that seem least plausible or which affect your result the most
  • It requires fewer assumptions (in general) to get to your result
  • It is easier to anchor on big numbers that are less objectionable (e.g., US population) than to smaller numbers based on experience / guesswork (e.g., avg footfall in a retail store in a shopping mall)
• For the times where Top Down is not easier / not helpful, then of course go Bottom Up, but be careful:
  • You will need way more assumptions than for Top Down usually (e.g., you may need different assumptions for # of sessions per day in a cinema depending on the weekend or a weekday), even if they may be easier to justify, because they're more based on lived experiences
  • I have seen often that most people end up over estimating their target number when going bottom up, that's usually due to mistakenly taking an uncommon occurence (e.g., a full movie theatre or restaurant on a Saturday night) for the average, so be careful to take appropriate haircuts where useful

Either way you choose to go, know that you should sense check (1) your approach / structure, (2) your assumptions, and (3) your results.

Best of luck!