Testing is one of the most complicated topics in affiliate marketing. There are no simple recipes or clear instructions. Even experienced affiliates often struggle to figure out how much money they're willing to spend to find a winning campaign and test different ad creative options.
This is even more challenging when it comes to push notifications and pop ads. These ad formats are cheap and can reach huge audiences, but they can't be deeply customized. So the performance of a particular pop-up ad campaign ends up depending on a lot of factors – not only on choosing a high-quality ad network and properly setting up blacklists, but also an element of luck.
We've looked at what different affiliates think about this. Some say you should spend about 30% of your advertising budget on tests. Others think the ideal amount is ten times the payout per lead. Let’s check out these different views and see which one makes more sense.
So, the basic cost of tests depends on the offer, geo, ad network, sources, and payout per lead offered by the advertiser. Some affiliates recommend spending ten times the payout per lead, but according to our stats, it usually costs about two to three times the payout per lead.
But you need to understand that this formula isn’t universal for everyone. Everything depends on the vertical, the geo, and the tests themselves. Let's look at this in more detail.
What factors does the test budget depend on?
There are many factors that influence how much budget you’ll need for tests:
- What exactly are you testing? The offer, the specific ad creative, or the landing pages.
- Which geo does the traffic come from? Tests in Tier-1 geos will be more expensive than tests in Tier-3 geos.
- How extensive are the tests? Are you testing two similar campaigns or six completely different ones?
- What vertical are you working in? Gambling tests usually cost a lot more than tests for health products or dating.
- What’s the payout per lead? This can vary even within the same vertical and directly affects your budget.
- How long will the tests take? A two-week test will be much more expensive than a two-day test.
The diagram above shows a simplified view of what affects the cost of the tests. For push notifications and pop ads, the average cost is about three times the payout per lead because these sources are cheap. For example, if an advertiser pays $20 per lead, you might spend $60 on tests. According to affiliates, this is usually enough for testing.
Vertical
The most important factor is the vertical you're working in. For example, if you're testing ads for cryptocurrency offers, you'll likely need a much bigger budget than if you're testing ads for utilities. Leads or potential customers are harder to come by in the crypto industry.
Following the general rule of spending ten times the payout per lead, with the high cost of a lead in crypto, you might have to spend between $1,500 and $10,000 on your testing. But sometimes, such large budgets aren't even necessary. Affiliate marketers from the LeadBit team said they only needed $690 for their entire ad campaign, with just $150 going towards the testing phase.
In the dating vertical, those kinds of budgets are more common. In this vertical, you can become profitable with just $200 in Tier-1 geos, which isn't even the best result. Affiliate marketer Eugene TrafficSourcer spent exactly $200 on testing before he found the winning ad for dating offers in English-speaking countries like the US, Australia, and New Zealand. Only after that did he start to scale up his campaign.
Geo
In the Tier-1 geos, the CPCs for ads are way higher than in the Tier-3 geos. This is because people in Tier-1 geos are more used to seeing advertising, so it's harder to get their attention.
This means you'll need to spend a lot more on testing your ads in Tier-1 geos compared to Tier-3. For example, affiliate marketer Varun Keskar said that for gambling ads in Tier-1 geos like Germany, you might need $1,000 or more just for the testing phase. His total monthly ad budget was $5,507, with $1,200 of that going towards testing.
In contrast, costs are much lower in Tier-3 geos. One affiliate from the Adsterra team spent only $338 on a gambling ad campaign in a Tier-3 geo, with just $100 of that going toward testing. And these tests only lasted for the first two days.
Scope and duration of tests
The more variations you test and the more they differ from each other, the more money you'll need to spend. Testing 6-7 different ad elements at the same time is overkill. Only one of them is likely to perform well, while the others will just eat up your budget. It's better to test 2-3 campaigns within the same geo, vertical, and offer, and just change the ad creative or design.
Some affiliates recommend testing for at least 2 weeks. But in our opinion, that's usually too long – a full-scale ad campaign can last two weeks, but for testing purposes, that's just too long. Other affiliates suggest spending no more than one day on the test.
On average, based on the experience of many affiliates, a normal campaign test takes 2-3 days – in push notifications, this is enough time to see if the ad creative works or not. However, in some niches like cryptocurrency, this short timeframe may not be enough. If you want to save time and find a profitable campaign faster, the managers at MyBid can tell you, based on other affiliates' experiences, what types of ad combinations are currently performing well and what you need to do to get profitable as quickly as possible.
What to test first
The screenshot below shows a post from a forum where a newbie wants to launch an ad campaign in pop ads but is worried that it will cost them more than $1,000 just for the testing phase.
The affiliate is forgetting that the most important thing is to first test the actual offer or product you're promoting. If the offer doesn't perform well, it will be very difficult to run a successful ad campaign for it no matter what else you try.
So the first step should be to test the offer itself. If the offer "flies" or performs well, then most of your other campaign elements like ad creatives and geos will likely be successful too. But if the offer doesn't work, it's going to be an uphill battle.
After testing the offer, you can then start testing different geos and ad creatives, gradually increasing your budget. The experience of other affiliates shows that $150-$200 is usually enough for the initial testing in push notifications and pop-ups, except for the most expensive verticals like cryptocurrency and gambling.
If you don't want to spend too much money testing new geos or offers, the network manager at MyBid can give you a rough idea of what a winning ad combination should look like, and can even help you set up the campaign themselves.
When to end the test
As we mentioned earlier, ad tests usually last between 2 to 3 days. This time period is generally enough to determine if you've found a successful campaign or not.
However, we have a recommendation to add. If on the very first day of testing, one of your ad variations performs so well that it's already generating a profit, then you can go ahead and finish your testing early and start scaling up that winning ad. This will save you money on additional testing and ultimately lead to a higher ROI.
An affiliate from the forum actually wrote about this strategy:
Bottom line
There isn't a universal answer to how much you should spend on testing ads. The budget can vary a lot depending on the geo, offer, vertical, and how extensive your testing is. However, most of the time, you can get by with spending $150 to $200 over a few days.
Most affiliates agree that this amount is usually enough to weed out what doesn’t work, find a good combination, and start making a profit. If you want to speed up this process and quickly find a successful approach, contact MyBid. We can tell you which strategies are currently working well for our affiliates. This can give you a head start instead of having to figure it all out from scratch.
