Seven Questions Affiliate Managers Should Ask Before Launch

Before a test launches, the team and product can both speak confidently about a good result while meaning different things. Yulii, Senior Affiliate Manager at PIN-UP Partners, proposes seven questions to align the traffic source, market, budget, cap and scaling conditions. His recommendations follow in his own words, and at the end the editors answer two practical questions: how to record the agreement and how to reconcile events. Start with a shared picture of the test rather than discussing the rate in isolation.
Traffic source, market and test budget
1. Which traffic source and approach are you planning to use?
Simply answering "We run Facebook traffic" or "We work with In-App" is usually not enough to plan a test properly. The traffic source gives only an initial understanding. What matters next is how exactly the team plans to build its user funnel: which sources and formats it intends to use, which approaches and creatives it wants to test, and which of them have already delivered results.
You cannot give the same offer to two teams and expect identical launches. Funnels are built differently across digital sources and technical formats. What you can test in creatives, user journeys and individual campaign elements depends on the source, the technical solution, platform requirements and the rules of the relevant GEO.
In In-App channels, campaign configuration also depends on the availability of the relevant applications, so the user journey is planned with technical limitations, platform requirements and applicable advertising rules in mind.
So at this stage I work out how the team operates. Only then can we properly discuss which funnel to build, which approaches are worth testing, and where to start.
2. What do you already know about this GEO, and how much are you willing to spend on the test?
Next, I need to understand the team's experience in the specific GEO and whether the selected traffic source and advertising approach can be used under the applicable market requirements. Experience in a GEO alone does not give the full picture. What matters is what results the team achieved and what exactly worked or failed at the time.
If a previous test performed poorly, I first look for the reason. It could have been expensive traffic, the wrong approach, a low CR, or the funnel itself. Poor results in the past do not necessarily mean the GEO is unsuitable for the team: sometimes the problem was a particular launch rather than the market.
If the team already has successful experience and proven approaches, the launch usually moves faster. We can take a new offer and split-test it on approaches the team already knows how to run. If the team is entering a GEO for the first time, it makes sense to spend more time on preparation: reviewing current approaches, the budget and the first test together.
I also always clarify the test budget and acceptable CPA. These two figures decide what kind of launch is possible at all: the cap size, GEO, approach, and sometimes the funnel itself.
At PIN-UP Partners, we do not start a conversation with an offer and a rate without considering the team's economics. First, we look at the available budget, what CPA the team considers viable, and what experience it has in the GEO. Only then do we choose the testing format.
Agreeing on the test and useful volume
3. How exactly will the test be structured?
Before launch, I need to understand not only the budget but also how the team plans to allocate it:
- how many media buyers will take part;
- how many approaches and creatives will be used;
- what exactly we want to compare;
- how much time we are prepared to give each variation.
If several media buyers are involved, I usually suggest separating them by sub-IDs from the start. The same can be done for creatives and approaches. Once the test is complete, we see more than an average result for the whole team: we can tell what worked and what dragged performance down.
It is also worth agreeing on timelines and interim reviews before launch. Instead of waiting until the cap is used up to look at the numbers for the first time, compare the team's acquisition data with what the product is seeing as the test runs. This makes it possible to adjust the approach, creatives or traffic pace while the budget is still there.
Some starting conditions should be discussed separately. If the team relies on only a limited set of external analytics tools, tests a single approach, or plans to launch as quickly as possible, it is important to assess in advance whether that setup will produce representative results.
4. How much volume is needed for the test to actually provide meaningful results?
When setting a cap, it is important to avoid both very small and very large volumes. At PIN-UP Partners, we typically provide 50-100 FDs across all GEOs, as we work with Tier-3 markets. This benchmark is based on statistics from previous tests and an analysis of campaign results.
- 50 FDs are usually enough when one media buyer tests one approach: that is enough to collect the first meaningful data.
- 100 FDs make more sense when several buyers or approaches are involved. Otherwise, the sample for each variation is too small to compare results.
Too small a volume may not show an accurate picture of the audience. Too large a test on an unproven setup creates unnecessary risk for the product. The cap should give both sides enough data to make the next decision after the test.
One more point is worth discussing right away: what happens after the test cap. If the results are good, it is better to know in advance how much volume the team could deliver next.
Results, scaling and work after launch
5. What exactly will we consider a good result?
Before launch, I always try to agree with the team on what results would make the test successful for it. For some teams, ROI is the only critical metric. Others focus on potential scaling volumes or the chance of a higher rate. This needs to be discussed beforehand, because the team and the product evaluate the test from different perspectives, and the final decision has to take both sides' economics into account.
Product metrics become meaningful at different times:
- right away: mFD, mRD and the number of repeat deposits on the first day. They give an initial idea of the audience, but they are not enough for a final conclusion;
- later, once players have had time to engage with the product: average deposit value, the RD-to-FD ratio, betting turnover and margins.
In practice, the numbers do not always add up neatly. For example, the team may have high acquisition costs and a low ROI, while on the product side we see strong repeat deposits and retention. For me, this is not a reason to stop the test immediately: such traffic may prove valuable in the long run. In that case, it makes sense to extend the cap, watch further performance and discuss the terms. If the current economics of the test do not work for either side, we can discuss adjusting the commercial terms of future cooperation.
The opposite also happens: acquisition costs look attractive for the team, but retention and repeat deposits on the product side fall below expectations. Then it is equally important to look at the overall economics rather than CPA alone.
At PIN-UP Partners, a good test result means the economics work for both sides: the team gets an acceptable return, and the product gets quality players. If the numbers do not align yet, we work together to see whether the terms can be adjusted and the cooperation continued.
We covered other metrics that help you see the money in iGaming in a separate article.
6. When is a successful test ready to scale?
After a successful test, it is worth checking three things:
- Whether the team is satisfied with its ROI.
- Whether it can increase volume.
- Whether the metrics the product needs hold up after one or two weeks.
If all three are in place, the next step is to understand exactly how the team can scale the specific traffic source: what limitations it faces, how quickly it can increase volume, and what happens to acquisition costs and traffic quality as volume grows.
After that, we can build a plan not for a couple of days but for at least the next one, two and three months. At this stage, we look at longer-term metrics: how players carry over into the following month, the share of VIP players, whether audience quality holds, and whether the economics start to slip as volumes increase.
For me, scaling is healthy when the same logic that made the original test successful holds as volumes grow. If all key metrics deteriorate quickly after the budget is increased, the test was scaled either too early or too aggressively.
7. How will we work together after launch?
It is important to agree in advance on how we will communicate during the test. Usually, the main calls happen before launch or at the scaling stage, while after the test cap communication moves mostly to chat. The key is for both sides to exchange data regularly and be able to compare results quickly whenever a question comes up.
The data exchange usually looks like this:
- the product provides relevant aggregated campaign metrics, such as RD dynamics on days one, three and seven, the share of unique RDs, and other available traffic quality metrics. This information is provided to the extent necessary to analyse test results and in accordance with applicable data protection requirements;
- the team shares current acquisition costs, the situation with the traffic source, creatives and volumes, and whether the offer delivers the ROI the team considers viable.
The most useful discussions start when the two sides read the numbers differently, as in the examples from question five. That is when regular check-ins help decide whether to continue the test or stop.
This is the affiliate manager's role after launch: bringing the team's data and the product's data together into one picture. At PIN-UP Partners, we look at the economics of both sides so that, together with the team, we can understand whether to continue the test, what can be changed, and when to move on to scaling.
Questions before the first test
Who should record the agreement?
Both sides should confirm the same description of the test: source, market, budget, cap, events and the conditions for the next decision. A list of numbers without definitions is not a shared agreement.
Can a campaign scale immediately after its first FTDs?
First deposits are a reason to talk, but they do not replace the agreed observation period. Yulii recommends agreeing in advance which metrics the product uses to judge traffic quality and when there is enough data to decide.
What if tracker and affiliate platform figures differ?
Start by comparing event definitions, time zones and reporting periods. Check identifier passing and postbacks before changing rates or volumes: a dispute about results often starts with the two sides counting differently.
A shared agreement makes the launch useful
Yulii’s seven questions turn an offer discussion into a plan for working together. The buying team gets clear test boundaries, the product gets explicit quality expectations and both sides have a basis for their next decision. Write the answers down in one document and assign owners before launch. It will not guarantee a successful test, but it helps spot mismatched expectations early and spend the budget on testing a hypothesis instead of an argument.







