Hi,I had an interview yesterday, where I was asked to estimate "how many TV sets are sold each year in the UK". The question was focused on both the residential and commercial markets (e.g. offices, hotels).I found it fairly straightforward to estimate the residential market for TV sets.However, I really struggled in estimating TV sales in the commercial market. I didn't know where to start in terms of estimating the number of offices or hotels in the UK for example.Could anyone please advise on how they would have estimated annual TV sales in the UK commercial market?Thanks very much, John

Get some expertise and people will stop asking you this kind of questions. Then the questions will be more specifice, more like, "Here is a piece of paper. Write an SQL query to merge these two tables."

*specific

Hello mate.. i'd do a rough on the back of envolpe number crunching on the following linesI 800 X 50 Small hotels 5000 X 250 B&Bs 200000 X 1New Hotels on an average 10 X500 500 X25 100 X 50 Category :- Companies Subcategory big 8000 x 6Medium 50000x 2Small 20000x10 Total market in Numbers = Approx 2.1 Million... dont think will be millions of miles from the actual figures. hope that helps

Jon,Thank you very much for your help, I appreciate it. I guess the bit I found hardest about estimating the size of the commercial market was justifying why I believe that there are 'x' thousand hotels in the UK.When I was pushed to justify my assumptions, I used a 'bottoms up' analysis i.e. 'my hometown has 'x' pop'n & 'y' hotels so I will extrapolate that for the UK pop'n. What do you think of this method for justifying assumptions.Polish Plumber - sorry, I didn't find your response that helpful. I'd certainly be surpised if I were asked an SQL question for a strategy interview!

out of curiosity, having got a market population, how did you estimate the frequency of refresh to get an annual figure?

Always try take a demand-based approach to mkt sizing if possible. It's a bit more "technical" and impressive. i.e. X amount of hotel rooms (you are more interested in # hotel rooms reqd than # hotels) will be reqd for a demand of Y (based on local (business & leisure) and foreign (business & leisure) trips. Very difficult though, as the same room will host many stays. Maybe you've fo factor in an occupancy rate.But yeah, it is a tough one. Give me golf balls in a boeing 747 any day.Regarding refresh, assume hotels change TVs every X yrs -say 3-, then in any one yr. about a third should be renewing

John,Some of my numbers may be crazy but overall they seem to balance each other. That is one of the important things ( at least that is what i think) while solving these cases. The approach i have taken for the quick number crunching example above is as followsWho needs Tellys - Hotels, companies . Then categorise them briefly. You will have to do a bit of a number crunching for number of big hotels in the UK, B&Bs etc ( so a case study within a case study). Once you get past that stage then as some pointed out in one of these posts you need to come up with rate of refresh per hotel /compny. A quick and reasonable assumption will do. Add to it the number of tellys a new hotel or a new company will buy. Again two sets of numbers to be considered here. Average number of new hotels/ companies in a year. And then their demand. So 10 new B&Bs would, say, order 5 TVs on an av. then your totall is 50. Your numbers can go bonkers here. But as i numbers will get balanced if you use a lot of commonsense. The most impotant thing here is your crystal clear approach and your assumptions. Take care of these two aspects and numbers will do justice to you. Hope that helps.

anyone ever had the how do you fill a room with sand?

Package the sand up in small bags (e.g. paper - such as flour or sugar comes in, or plastic), then stack starting from the back. Or is that cheating?

Wet the sand. ... or is that also cheating?

who cares - simply lock the door and put a notice on saying 'Danger Keep Out - Room Filled With Sand'

Why do you want to fill the room with sand?

Sorry guys but I think you all went the wrong way about these.First of all, thats not a case study, its just a simple question trying to understand your thought process. You give these simple questions too much credit by calling it a Case Study.I got asked a similar question about a month ago for an MC I have accepeted Q: right now, how many ipods are being used in London?and found that just as simple as your TV one. Remember, MCs tend to overcomplicate everything, as an ex Industry guy I like to keep it simple.First of all, always tell interviewer you want to lock down the scope before moving forward. In my case I confirmed: "London" is inside M25 ring. "Ipod" includes apple only, none of those silly M3 thingys. By "used right now" you mean 0815hrs GMT.Next, go through potential worst case scenario. There are 60million people in the uk, 15million in M25 ring. But out of my 15mil, looking at the tube journeys and walking around, 10% of people have an ipod. Thats potentially 1.5mil users "right now" 0815hrs GMT. However, not all use it to go to work, some are still sleeping etc hence 10% of those are using it, hence 150k are using their ipod right now.The answer is not important, how you got there is what counts. Do not overcomplicate things like a typical MC.

Anon FX has the approach spot on in my view. I run through these exact types of questions in my case study briefing and this angle of attack is exactly what I advocate. The next session is in May or a recording of the last session is available immediately. See: <a href=http://events.top-consultant.com/UK/careerconference.aspx?ID=140>Revealed: How to Ace your Case Study Interview</a>Tony RestellTop-Consultant.com

Thanks Tony.Always good to hear positive feedback if/when I get things right.While I&apos;m here, good luck to anyone going through the various rounds of interviews that makes up joining MCs.