Metrics and how do I calculate them? (4)
I have seen a lot of talk about metrics and how to calculate them. I have seen specific examples that people had knowing that their banner ctr was good, their landing page ctr was good, but they needed to test more offers to get profitable since the conversion rate was lower than desired.
My question is how does someone know that? Does experience with a platform tell you what is possible? Like I can get a 0.3 ctr on traffic junky with a bit of effort, but rarely have I seen a 0.4. I know that a site like Tube8 has poor ctrs and just as poor landing page ctrs.
But is there a better way to calculate this and understand which metric is most important for my campaign?
The definitive post on this topic is this epic one by caurmen.
http://stmforum.com/forum/showthread...ill-Make-Money
Originally Posted by cmdeal
A part of it comes down to experience and you just know what numbers are good or not when it comes to CTR and CR - which of course will vary from niche to niche and traffic source to traffic source.
A good campaign optimization strategy will aim to improve on each factor that affects ROI in a systematic way. e.g.
1st round of testing - mainly banner testing: 2 hottest offers + 2 ripped landers + 15 banners (5 angles x 3/angle)
2nd round of testing - lander testing: better offer from last round + better lander from last round + 10 ripped & custom landers + best banners from last round
3rd round of testing - banner testing: same offer + best lander from last round + same banners + bunch of new banners (some from best angle so far + some from completely new angles)
4th round of testing - offer testing: same offer + bunch of new offers + best lander so far + best banners so far
Etc. etc. so that you keep improving every element: offer, banners, and lander.
Then of course you can tweak the bid and campaign targeting (placements etc.).
The above is just an arbitrary/sample plan - so please don't get hung up on the details. Basically at any point in time you'll want to test the element that you believe will result in the largest potential increase in ROI. So depending on your particular situation the order of testing will differ.
To finally answer your question: Instead of asking the question "what CTR should I be able to achieve for my banners on this placement on this traffic source", you could just focus on trying to beat your previous record (i.e. your test control). Knowing what's possible can be inspiring, but it could also be discouraging if you can't reach the same numbers someone else is touting they can do. On the other hand, you may be able to do better if you don't let a certain "number to shoot for" limit you.
Also - make cutting decisions based on CR and not CTR. (There are very few exceptions to this rule.)
Amy
Home > Paid Traffic Sources > Adult Traffic (NSFW)