Market sizing case study example

Question: How much milk does Starbucks use on a typical working day in the UK?


You (interviewee):

  • For the sake of the analysis, I will assume that there are approximately 1,000 Starbucks stores across the UK. I am aware that these stores vary in size, but I will perform calculations based on a sample average-sized store. I am assuming that this sample store opens at 7am and closes at 7pm on the average working day (meaning there are 12 hours of trading per working day). 
  • I will proceed on the basis that there are typically two cash registers operating at any one time and each serves, on average, one customer per minute. This average is based upon the assumption that more than 1 customer per minute is served during busy periods (for instance the mornings) and fewer at other times. We therefore have 2 customers per minute (remember, we have assumed that there are 2 cash registers operating simultaneously) x 60 minutes x 12 hours, giving us 1,440 customers per day. 
  • I am aware that not every customer will buy drinks that include milk (for instance soft drinks and black coffee). However, some customers will also make multiple purchases (for instance, if they are buying drinks for other people). I will therefore proceed under the assumption that 1,440 drinks requiring milk are sold on average in each Starbucks store every day. I will also assume that the average sized drink is medium (and medium drinks are approximately 480 millilitres in size), and that milk accounts for 25% of the volume (i.e. 120ml per drink).
  • Finally, we can multiply the number of drinks requiring milk that are sold on average per day (1,440) by the average amount of milk contained within each of these drinks (120ml), then multiply the resulting figure by the estimated number of Starbucks stores in the UK (1,000). 

 

Interviewer: Do you think 172,800 litres is a realistic estimate?

You (interviewee): 

  • As the analysis is comprised of a wide range of assumptions, there is a realistic chance that the estimate may differ from the actual figure. 
  • The assumption that 1,000 stores exist around the UK is based solely on intuition, as are the assumptions relating to the average number of drinks requiring milk that are sold and the average amount of milk that goes into each drink. 
  • Opening hours will vary, as will the number of cash registers. During peak times, the queues in stores may force staff to work faster and more efficiently (and there may be some staff that work particularly slowly, regardless of the setting). 
  • Furthermore, the average number of drinks sold may change in correlation with seasons (i.e. more hot, milky drinks may be sold in the winter) or with the release of new products or promotions (e.g. a new brand of espresso or the launch of the Starbucks loyalty card). 

As part of your analysis, you could have further segmented Starbucks stores. For instance, you could have given estimates for small stores in big cities (for instance in train stations), large stores in big cities, small regional stores (e.g. in regional towns) and large regional stores (e.g. in motorway service stations). 

You could then make estimations that are more tailored to the likely volume of relevant drink sales within those different segments. You could follow this by estimating the number of stores within each segment and then coming up with a total figure. The decision to segment must be considered in light of the particular facts you are facing and the time limit you are given (if you are given a time limit).