How often will the LP with lower CTR convert better than the LP with higher CTR? (7)
How often will the LP with lower CTR convert better than the LP with higher CTR?
This is something I've been wondering. I'm running a high payout offer on native ads. It's CPC. Without statistically significant data on conversions, I'm forced to place a heavy focus on ad ctr and LP ctr. My biggest focus now is to find which ads, when clicked on, will lead to the most clickthroughs through the LP to the offer. I'm doing this because I figure the more eyes I get on the offer, the greater number of conversions I will have. But, I don't want to risk cutting an ad whose users click through to the LP less, yet still have a higher conversion rate.
Example:
Ad 1:
100,000 impressions
Ad CTR: .2% (200 clicks to lander)
LP CTR: 50% (100 clicks to offer)
Percent of users that buy: 5% (5 conversions)
Ad 2:
100,000 impressions
Ad CTR: .2% (200 clicks to lander)
LP CTR: 20% (40 clicks to offer)
Percent of users that buy: 20% (8 conversions)
As you can see in my hypothetical situation, for whatever reason, the people who clicked on the second ad were more interested in actually buying, and even though there were less eyes on the offer, they ended up converting more than the larger amount of people that clicked on the first ad.
What I'm wondering: Does a situation like this ever happen to you? Should I adjust my strategy, or is it a solid idea to try to get as many eyes on the offer as I can for now?
Yes - that actually happens more often than you may think. 
This is why CR should be the main metric to use when cutting banners and landers, and not CTR.
Some advertisers aiming to increase brand awareness, will do their best to attract eyeballs by maximizing CTR. However, as affiliate marketers, our aim is to maximize profits or conversions.
It's not difficult to increase CTR - for example by making unrealistic promises - but when visitors get to the offer lander, they may not convert because they don't see any claims to back up those promises. Another example is to increase CTR by piquing people's curiosity, but when they land on the offer page and find out what it actually is, they may sneer and leave (unless it's done in good taste - enough to make people chuckle and convert for example).
And then there's the other end of the spectrum: Some banners/landers do a good job of prequalifying prospects, such that only the "right" sorts of visitors will click through (i.e. those that are most likely to convert). Not many people will click through to the offer, but of those that do, many will convert.
In summary, trying to increase CTR is a good thing, but only as long as CR is increased at the same time. Ask the question "What sorts of people do I want my creatives to attract?" And "What sorts of people would be more likely to sign up for this offer?" Then craft your message to cater to those people. Also, making the transition from landing page to offer page as smooth as possible will help - using similar words/phrases, colors, etc. (if the visitor can't tell the difference between the landing page and the offer page and thinks they're on the same site that'd be even better!) An exit popup that instructs the visitor on what they need to do on the offer page will often help as well.
As to what sorts of people to target, there are basically 2 approaches: Either target the masses (lots of traffic = lots of conversion, but lower conversion rates) or niche demographics (less traffic = fewer conversions, but higher conversion rates). If you're working with a traffic source like facebook where you can target by different demographics, taking the trouble to set up a bunch of campaigns to target different people will often give you better conversion rates. Or you can choose to make one large campaign to target the masses using generic angles - conversion rates may be lower but you may be able to make 1000 times as many conversions as going niche, and if you can make that profitable then you may decide to save yourself the trouble of going niche. Then there's pop traffic for example where you can't target by demographics, in which case generic angles aimed at converting the masses will usually work better.
Hope that helps! And in case you need some direction on how to cut banners/landers, this thread may help:
http://stmforum.com/forum/showthread...Banners-Part-1
Amy
Thanks Amy. I'm going to switch up my strategy a little.
I've spent $70 on some ads because they have high CTRs through to the offer. No conversions have come from them. These ads have "shocking" images and claims that aren't entirely focused on what my product does, but I threw them in because they offer to solve a problem many are insecure about and would like to fix. Since it's CPC traffic, these ads have ended up costing me more compared to my other ads because of the increased amount of clicks, and they've gotten me nothing in return.
I have a few conversions on a couple of the lesser CTR ads now, and I've made some variations of them with new images & headlines, so we'll see how they do.
It's difficult and risky to optimize without much statistical significance, but I don't have much of a choice here. I'm seeing patterns in the data and trying to attribute these patterns to possible user thought processes, and then I'm testing new possibilities from there. Hopefully I'm correct.
LOL I know what you're talking about regarding the high CTR ads costing you a fortune, because I did adwords for a while back in 2008 - you have my sympathy for sure! Something similar can happen on CPM networks as well, because some networks would automatically show high CTR ads more often, so if they don't convert they end up burning a hole in your pocket.
This is a common problem faced by many newer affiliates in the beginning of a campaign - a catch 22 where none of the multiple variables has a track record. So if the CR is low there's no telling whether it's the offer, banners, landers, or placements that's causing the problem. To combat this, I would suggest the following:
-Stick to "hot" offers recommended by AMs, and test at least 2 offers from the beginning, preferrably from separate aff networks to minimize chances that you're pushing duds.
-Spy a lot to make sure similar offers are already being promoted on the same traffic network you're planning to use - that's an indication that that type of offer is "working" on that traffic source.
-Rip the most-often-seen banners and landers and mod them up - those can serve as a benchmark. Also come up with own angles and create banners/landers from those. Including popular bannrs and landers can mitigate risks of using totally unproven creatives.
-When testing new offers, try to use a traffic source you have experience with, that you know has good quality traffic. This will allow you to rule out the possibility that traffic quality is causing a low CR.
Good luck with your testing!
Amy
Your plan sounds great Kevin! However, it's not necessary to test another vertical unless you have time and money to spare.
Amy
Ok, I don't have the cash yet so I'll hold off on it for now. I'll probably just try it if these offers don't work out.
Thanks again
Home > The Newbie Zone > Questions and Answers