Hello everyone,
I got this question in an interview and was stuck on how to solve it
How would you approach this problem?
Back to overview
2
Be the first to answer!
Nobody has responded to this question yet.
Hi, thanks for your question. I think the case is essentially testing your ability to segment a massive concept like 'internet websites' to its components. Below is my answer for this case:
Assume one website represents one particular entity (e.g. LinkedIn is one website and not multiple web pages)
No. of active internet website = [No. of government websites (e.g. US Department of Justice, Ukraine's Telecommunications Commission etc.) + No. of Private Sector websites (e.g. Facebook, McKinsey, UBER etc.) + No. of NGO websites (e.g. Oxfam, Red Cross etc.)] x 1.1 (10% for others that don't fall into this category like random personal websites you're developing out of interest, school work, pilot project etc.)
No. of government websites = No. of countries x Average no. of industries per country
No. of Private Sector websites = Average no. of companies x No. of industries
- If more detail is required, average no. of companies can be broken down to family businesses, SMEs, and large corporations. I would advise you to size based on a medium-sized country and just use that to multiply by the total number of countries
No. of NGO websites - Similar as No. of Private Sector websites
Hope this helps in any way and happy to get comments/feedback on this - thanks!
Hi,
I would try this:
- 1M words in English language
- 60% or 600k are nouns
- 1.2 M if every word has multiple forms (e.g. dog / dogs)
- How often do I see a single-word web-sites? Probably 1 in 50
- How many domain zones? .com is probably 60% of the internet
You can take it from here and add as many assumptions as you want (e.g. different languages, word permutations, etc). I think this case is more about the approach rather than having precise numbers.
Best
