Affiliate Marketing Top 5 Mistakes When Testing an Offer or How Not To Fail at the First Steps

The first Covenant of traffic generation states: "Tests are our everything." And it is necessary to approach them with the maximum share of responsibility. 

Remember one truth: when you test an offer, you buy information, training, and experience. You shall not forget about the offer if you didn't get some good profit right on the first day. Your task is to get information as quickly as possible so that you can make money from this information, this is important. 

But let's not beat around the bush, let's get to the point.

1. Lack of a clear algorithm of actions

As we've written above, we buy information, but it also needs to be properly digested. If you lose $40-50 or make just nothing, then you should not immediately start testing another offer. Analyze everything as you may have made a mistake somewhere.

You should check all the segments of the campaign and see which one is failing. Check again: CTR, CPM, CPC, GEO, gender, age, etc. The algorithm shall look like a selling script. For example, you have done something - look at the result. If positive, then work this way, if negative, change it. If the offer is in the affiliate program, and the affiliate program acts as a direct advertiser, then the offer makes traffic and the traffic is good.

Don't despair, if you lose money while testing. Don't take it as money to waste - you've gained the experience.

2. Poor GEO work

The question of working with geo is in the first place. After all, it can be so simple that you remove a couple of geos and the campaign starts working like crazy. Let's look at an example.

You have an extremely low approval rate. Create labels in the affiliate program that will show traffic from certain GEO, and you will immediately see which geo is ruining your campaign. It may be that some Tier 3 country spoils everything and makes nothing but trash.

If we look at it in practice, it looks like this:

1. You launch almost every possible geo (multi-geo).
2. Put labels and divide the traffic by geo. You do this both on Facebook and in the affiliate program.
3. Compare the indicators for each GEO and choose the optimal ones.
4. Create a new campaign with profit data and without bad geo.

3. Click price

A webmaster should understand that each product and geo has its own price per click. The average value can be from $0.03 to $0.3. When you start testing an offer, the first thing you do is look at the CPC. After all, based on the CPC, you can already draw some conclusions.

If the product doesn't perform well in this geo, maybe you are presenting the information not so much interesting and/or creative doesn't catch much, it will be noticeable by the CPC indicator - it can be about $0.4 or more. Such ads campaigns must be stopped and edited. CPC rarely gets lower at all.

4. Disabling ads if the lead price is incorrectly calculated

Few traffic generators start calculating the optimal price for the lead before running the test. Not few but most of them shall do so. After all, thanks to easy mathematical calculations, you can make your work easier. Most people think: "Why count if the negative result will be clearly seen". But it is better to spend 2 minutes and know the exact numbers.

Everything is done according to the formula. Let's do the math. For example, our payment for the offer is $30, approval for the offer is 35%. We use the formula (30*35/100=10.5).

Our price per lead should not exceed $10.5. In other words, if the price for the lead is more than $10.5, we will work for minus, if less, that's going to be the profit. However, at first, we can afford to pay a little more for the lead because the pixel needs to catch on to the audience.

5. Testing of a huge number of creatives and combinations

While testing offers, we strongly don't recommend that you make 10 creatives, launch 10 different advertising campaigns for 1 offer, and start testing. Why you shouldn't do so When testing such a huge number of combinations, you simply will not have enough money to check each ad efficiently.

Because of this, you will not be able to get high-quality analytics and subsequent correction work. It may be that you choose a lower-quality and worse converting ad because of the simple haste and unwillingness to spend money on further testing.

For a successful test, make and test 3, well, a maximum of 4 combinations and approaches. If some of these plays right, scale up, if nothing does, come up with something new.

Share the article