Campaign ROI -80% (12)
Hi guys,
i started a campaign yesterday with frequent conversions. But my ROI is -80% at the moment (without optimizing anything).
So my question: Is it possible to get a positive ROI out of this by optimizing and testing or should I dump the whole campaign and try something else?
Thanks in advance,
Originally Posted by germoney
-80% looks bad, but sometimes its still doable.
What do the placements look like? All negative, some big ones negative, some green ones?
What traffic is this - pops, banners - shitty banners maybe? Or the LPs? Or you need LPs and you went for a direct link ...?
Offer? Did you pick a good one, maybe you've hit one thats not good?
Targeting - did you setup proper targeting? GEO, devices, carriers ... etc ?
There are too many questions here, be more specific.
It's poptraffic and direct linking. I already tested a landing page, but it performed really bad. I figured out the problem could be the loading time. My campaign runs on the other side of the globe, so loading time is really an issue. I guess its time for CDN or another VPS.
Some of my Placements are green and there are a lot of parameters to optimize so I'll give it a try.
Pops without an LP - definitely a lot of room for improvement here.
Regarding speed - what kinda loading times do you see for it when tested from various places in the world? You did test that right?
Pop traffic sources are flooded with bots, so there can still be some promise in this campaign if you manage to buy just real traffic.
Yeah I tested the loading time from various places:
Target Country: 2.5s
Frankfurt (near Server): 1.3s
New York: 0.9s
Waterfall view shows me I should optimize the image:
Loading time of html: 0.6s
Loading time of image: 1.3s
Compressed the image from 88kb to 23kb and changed the image hosting. Loading time of improved landingpage:
Target Country: 1.6s
Better but still not good, what should I do next? What kind of loading time is okay for pop traffic?
Tested loading time of google.com in my target country: 3.9s. Maybe my site isn't doing as bad as I thought.
Nevertheless did another version of my landing page without any images: 0.9s.
I'll now run some traffic to test this version. 
For loading times, anything over 2s will cause visitors to click away in droves.
For optimization tricks please go to the following thread and search for "Try decreasing your page load speed!":
http://stmforum.com/forum/showthread...-AM-FAQ-Thread!
Some offers can convert by direct-linking, some can't. Generally speaking, offers that don't require a lot of action/commitment from the visitor, has a chance of converting without a lander - e.g. free downloads. Then there are offers that just don't convert without a good lander - e.g. mobile subscription offers that require the visitor to pay. I'm making generalizations of course, but my point is that if you're direct-linking to an offer just to test it, and if that offer isn't the type that would convert without a lander, then you won't get conversions even if it's a good offer.
It's almost never a good idea to just test 1 landing page, unless it's a proven landing page that has a good track record. Try to test at least 3 from the very beginning, 5+ would be better - and borrowing from the most popular landers you see from spying (or get from spy tools) would be a good start, because if a lot of people are using them it means someone must be making money from them! They will help you to at least establish a benchmark before you spend time creating and testing custom landers.
So my question: Is it possible to get a positive ROI out of this by optimizing and testing or should I dump the whole campaign and try something else?
Basically you need to make sure that:
1)You've set up your testing correctly to properly gauge how promising the offer is.
2)You run enough traffic to the camp to give every offer+lander combination a chance to convert. Oh and add major traffic segments to the list too.
Maybe time to start a follow-along?
Amy
Thanks Amy,
I did find a gold nugget in your answer
Just tested the direct link loading time: 6.1s. Wow thats pretty slow and harmful for the conversion rate ... but I can't speed that up.
Since my landing page is loading faster but converting worser than the direct link, its probably just a bad landing page.
Does it makes sense to create more landing pages for this offer if it is loading so slow?
Where can I start a follow-along?
My own tests showed that the loading speed will stop to matter around 500ms, going faster is not going to give you anymore advantage so to speak. Looks like you should be able to improve this a bit.
If the offer really loads that slows, its not gonna work no matter what you do. Test it a few times across the day and with more than one testing tool. It could have been some server hiccup or any other problem, Im having a hard time believing it could really load that slow. If it shows such bad numbers all the time, try to find a better offer.
Did you only test 1 LP? you should test at least a handful to start
@germoney: I second what matuloo said. You probably should test other offers. You can choose to tell your AM about it too as a favor. I'm sure the advertiser would like to know about this so they can fix this issue.
As for follow alongs: If you go to the "Traffic Sources" section, and go into some of the subforums, you'll find "follow along" sections where you can post. If you do, please feel free to send me a PM and I'll do my best to give suggestions.
Amy
Yes I tested only one Landing Page. Stopped now the Campaign and told my Affiliate Manager about the issue. Asked him about good performing offers, too. I'll create an follow along as soon as possible.
Home > The Newbie Zone > Questions and Answers