What to look for before choosing which offer to keep going? (9)
Hey everyone,
After testing several offers and getting a few conversions (most of the time one per offer tested...) I'm wondering: what to look for in a campaign before sending it more traffic?
Let me explain.
When you do some testing and get too few conversions in different campaigns to really get significant data, what do you look at to decide that you're allowing all your testing budget to only one offer?
Traffic price?
Conversion rate?
ROI?
My theory is that if you pick an offer that's converting well but traffic is more expensive and your ROI is still really bad, wouldn't it be better to allocate all your traffic budget to that one offer with the good CR% and step by step optimize to improve your ROI?
Or should you rather test more offers and wait until you get at least a certain ROI to decide to keep it going and optimize?
From what I understand testing many offers + many landing pages + many geos is a great system. But what if your budget won't last forever and you want to go green fast?
What's the most cost efficient way to go green?
Depends on your own strategy ofcourse, there isn't a best way I think. I prefer to look at CR first, not checking the CPC cost at all when lauching to be honest.
If I see the campaigns converts like I wanted/expected, I check how I can optimize targeting/ads to get a low CPC and become profitable and start scaling from there.
Some I know launch ads upfront, without even linking to a real offer. Just to get a low cpc first and then find suitable offers. For me that doesn't work at all.
In a while you get a "feeling" campaigns will work upfront yes or no. Then it's just a matter of coming up with good angles or use older ones which worked in the past or in other geo's.
Specialisation in a certain niche/offer/product hereby is quite key to be honest. This way you recognize "good" and "bad" offers more easy.
Originally Posted by stickupkid
I launch most campaigns in niches that I'm familiar with, so I know straight away if an offer has promise or not. I already have proven funnels that I can use as benchmarks, so it all goes faster and easier.
As for the OP questions : you need to start somewhere, if you are only trying to figure what offer is best, keep the rest of the funnel the same and throw traffic at a bunch of offers ... pick the best ones and then play with your creatives, to push the ROI higher. Once you know that your funnel is better than before, you can retest the other offers again. Offer is probably the most important factor of a campaign, so I would recommend to find a few solid offers first and then play with the creatives.
When testing new offers, I will only continue with those that gave me multiple conversions, 1 or 2 random ones are really not representative. Usually, you will see one or two offers standing out, so continue with those. You can always retest the other ones later, but to cut the costs of testing, pick the ones that are giving you the most conversions and the best ROI. Amount of conversions has a slight preference with me, cause that's faster and easier to optimize - that's why low payout offers are better for beginners.
At the end of the day, your ultimate aim is to maximize daily profits. Not CTR, not CR, not even ROI, but profits.
There are so many factors that can come into play here, and you've listed some of them. As to your question "What's the most cost efficient way to go green?" Here are a few things that come to mind:
-Run in tier 3/4 geos that are less competitive (where traffic is therefore cheaper).
-Run low payout offers.
-Do testing on a traffic source that has reliable quality traffic (popads comes to mind if you're running pop).
-Spend more time on research before you set up a camp. For example using a spy tool like Adplexity to see which offers and creatives are doing volume. Or talking to AMs and other affiliates to find out which offers are hot right now, or browse this forum to do the same.
-Stay in the same vertical for a bit. You know what types of traffic works the best for the vertical. You know which landers work. This will give you an edge.
-Run in the same geos. Once you've experience success in a geo, just keep testing new offers when your old one dies. You've already cut the worst placements. Why not benefit from that again and again?
Lastly - your test approach will be key. Setting up camps is like fishing. You have a finite length of fishing line (i.e. your budget). You can either have a hundred lines but drop them all in shallow (i.e. test a bunch of stuff but only allocate a bit of budget towards each), or have a few lines that run deep (i.e test only a few things but thoroughly). Or you can choose a middle ground.
Even when it comes to testing many offers + many landers + many geos, there's a right and wrong way to do it. The wrong way would be to randomly grab an offer and a couple of landers and throw up a test, and then ditch it when you can't make it green. The right way would be to test all the most popular landers in use first to find a winner, then mass-test offers - this way you'd be maximizing the use of your investment in finding the best lander, i.e. by using it to test a lot of offers. Whereas with the first approach, you'd always be using unproven landers with unproven offers, thus starting from scratch every single time.
But testing lots of landers + offers + geos is only one approach. As I've mentioned above, you can choose to spend more time on research to zoom in on the most promising landers and offers to test. If you go on Adplexity and spend some time to replicate other people's success, you may find some gems without having to brute-force your way by testing many landers and offers. However, do be warned that replicating other people's camps can be hit and miss as well - in some cases you're not seeing the actual landers because the person may be cloaking from spy tools. Plus, I bet you and I are not the only people that go on Adplexity to copy camps. A better idea would be to look for TRENDS on what's working, instead of just going to the camps that have "received most traffic" and copying that along with a hundred other people.
Either way: You need to make sure your offer+lander is good enough to make a large enough portion of the total traffic profitable, for the camp to be worth your effort in running it.
The funnel is the most important factor to get right, aside from traffic quality. You basically need to make sure it can make a good part of the traffic profitable , so you can cut the unprofitable part without having to spend a lot of money.
So what is considered to be "good enough"? If your current camp will require a lot of cutting before getting green, AND you won't be left with a lot of traffic volume after the cutting, then your funnel is not good enough.
However, certain traffic networks, in particular some of the ones that broker traffic from many places, will require you to cut more placements before you'll see green. This will need to be taken into account as well, and will require having enough experience running on that source to know that you'd need to cut aggressively/extensively.
So we've talked about placements. You'll also want to drill down to the other major traffic segments such as OSs and carriers. In order to have a profitable camp, all you need is for a large-enough traffic segment to be green, for your best offer+lander.
Back to your original question: If you have a high-CR offer but traffic is expensive, what do you do? You'd test bids as you should anyways, and run at the best bid. You can drill down into major traffic segments to see if any of them is at least close to breaking even. You can also check placements stats to see if there are some big placements that suck, that you can nix. If it's one of those sources that requires a lot of placement-cutting, then you run and cut some more. But if none of your major traffic segments is close to green, and you're not seeing big bad placements you can cut, then it probably means you need to test more landers+offers. This will apply whether you're talking about a high-CR or low-CR offer, expensive traffic or not.
I don't know if I've answered your question - do feel free to continue the discussion. But I hope I've provided some food for thought on the matter.
Amy
ask your affiliate manager and see how the offer is doing on the network, usually they can give you some good advice on trending offers, what's doing well, epcs, etc.
Many good tips here already.
The breaking point is to find that one good offer that makes you profits.
Once you see how a good offer behaves you will have a benchmark / reference point you can rely on.
More of these small successes the better you will become.
That's why this industry is a lot about testing and experimenting.
BTW don't forget: most of the metrics are very relative, the one that is the most important is simply PROFIT.
Good luck!
Thank you all for the tips!
One strategy that comes to my mind is to go and test as much as possible one offer that has been listed as a top offer by 2 different people at my network.
If others are making it work I just have to find out the recipe by testing, right?
It goes against what I'm reading all over this forum to test many offers and only keep the winners.
In reality I'm afraid to burn my budget by testing too many offers and geos instead of focusing on one.
Being stubborn doesn't help haha but I'm forcing myself to apply the tips I'm reading here.
Try to find an offer which is "told" to be good & you fully understand the flow. No matter which traffic source you gonna use, make the flow from ad to conversion feel as natural as possible. Use some common sense and ask yourself is this convincing, trustworthy and appealing for the target group you have in mind.
To be honest it's still the good old AIDA-model. 1) Attention, 2)Interest, 3)Desire and 4)Action.
1) Good title + appealing image = attention
2) Explain "what's in it for me" (me=consumer), make them enthousastic/eager, tease em a bit
3) Use tactics as; scarcity/time pressure/examples of previous customers/show results/fear/if you don't do this, you regret bla bla
4) Button/link or whatever - > good working conversion page (trustworthy/optimized for device/not too many fields/ no suprises (say free everywhere, but ask them at the end for a fee/deposit/whatever)
Originally Posted by jelz03
The other consideration is how much effort you're needing to put into testing each offer. If you're fixing up 10 landers, then it would be such a waste of time NOT to mass-test offers after you've spent the money to cut down to a winning lander.
So you may ask, why not just test fewer landers? You could, but of course you'd be casting a smaller net. However, if your approach is to replicate camps you find through spying for example, then you can justify testing 1 offer + 1 lander. But in that case you'd STILL need to rip and test lots of camps if you want to find winners.
To summarize: Testing a lot is crucial to AM success. The question is what you choose to test and how. This industry certainly isn't for the faint-hearted, and I understand the pain of not having a big test budget. If you're really not comfortable testing wide, you can start slow. Do more research: More spying, talk to more people, browse more offers and see how other people are running them. And only set up a test if you truly believe it has a good chance of succeeding. It would be hard to hit high numbers though when you limit yourself like this, but to each his own.
Best of luck with whichever approach you may choose!
Amy
Home > General > Affiliate Marketing Forum