4 ms·
The terrible numbers that Groupon doesn’t want you to focus on
- lazyjeff 15y ago"The median number of Groupons sold to each Groupon customer (someone who has bought anything): 1." "The median number of Groupons sold to each person on Groupon’s mailing list: 0." The median doesn't seem like a fair way to measure either of these. Obviously there the distribution of Groupons sold is skewed, so the mean is likely to be much higher than the median. It would probably paint a clearer picture to say "Of the 25% of people on the mailing list that bought something, half bought 1 groupon, and the other half bought on average 3.6 groupons." (numbers are made up)
- rockya 15y agoFor sold to each customer, I believe the median is a better measure. See http://blog.agrawals.org/2011/08/11/groupon-s-1-full-of-hole/ http://blog.agrawals.org/2011/08/11/groupon-s-1-full-of-hole...
- smackfu 15y agoMedian is pretty pointless, but so is mean. An actual distribution would be much better than both.
- rockya 15y agoyes, an actual distribution would be the most useful, but Groupon hasn't provided that data.
- lazyjeff 15y agoObviously a distribution would be best, but for a single number, the mean is better. For example, if groupon says they have 100,000 customers and they will have 200,000 customers next year, and the mean # of groupons per customer is 2.5, then I can compute that they sell 250,000 groupons per year, and if they get 200,000 groupons, they will sell 500,000 groupons per year. The median tells us none of that.
- qq66 15y agoI think we got the point. This is starting to sound like thinkcomp on Facebook.