Chase had ads on 400k sites, then on just 5k, with same results

Re:Chase had ads on 400k sites, then on just 5k, with same r... by stvswn on Thursday March 30, 2017 @02:34AM (#13993391)

1-5% click through rate on a display ad is pretty normal. If had impressions on 400k sites, the distribution of impressions per site had a long tail, and then you removed all of those that had no clicks, you'd expect to have a small set of placements with clicks. It doesn't mean they're spam sites, it's just math.

[ Parent | on HN ]

Re:Chase had ads on 400k sites, then on just 5k, with same r... by makomk on Thursday March 30, 2017 @05:28AM (#13993987)

Yeah. It feels like the New York Times is lying with statistics and fearmongering about "toxic sites" to encourage companies to only advertise with big sites - something that, as basically the biggest name in news in the US, they're positioned to benefit from financially.

[ Parent | on HN ]

Re:Chase had ads on 400k sites, then on just 5k, with same r... by roganp on Thursday March 30, 2017 @07:58AM (#13994594)

If 98% of the sites generate what is rounding error on your sales, then you measure that benefit against the risk that your brand gets associated with a hate speech site (maybe in the nytimes no less). It's math alright, but it means that the brokers are not adding a whole lot of value.

[ Parent | on HN ]

Re:Chase had ads on 400k sites, then on just 5k, with same r... by ssharp on Thursday March 30, 2017 @08:12AM (#13994675)

1-5% click through rate on display ads? You're missing a decimal point!

[ Parent | on HN ]