4 ms·
One major lesson to take from this: Make sure you fully understand the impact of changes you're considering for your site. It's easy to check one metric (in th
by jackowayed 16y ago
One major lesson to take from this: Make sure you fully understand the impact of changes you're considering for your site.
It's easy to check one metric (in this case, percent of cams) and declare that that metric proves that your change won't be a problem. But you should make sure to check every metric you can get (percent of sales and percent of traffic, in this case).
There's no reason you couldn't have tracked traffic to each cam and discovered that 60% of your traffic was to adult cams, right? And you may not want to track what cams each user views for privacy reasons, but you probably could have taken some anonymous data to see that adult cams make up a significant percentage of paid users' views, right?
- jacquesm 16y agoHindsight is always 20/20, the reason I write out this stuff is to make sure that others can learn the lessons without having to pay the price. Of course I could have done all those things, but what happened here is that after running the site for 12+ years I thought I had a good grip on this data and could bypass the measurement step. We've had a clean site policy before (until the .com crash made it impossible to afford people to oversee the site 24/7) and it worked reasonably well, I though I could simply turn back the clock on that but had missed out on a shift of the core demographics of the paying users.
- jackowayed 16y agoNot trying to blame you, just confirming that it was an avoidable mistake to support my point that you should gather as much information about changes as possible to avoid similar mistakes.