A setup (or "bundle") in affiliate marketing is a combination of traffic source, offer, GEO, audience, creative and pre-lander that makes a profit. "Where do I get a working setup?" is the most common beginner question. The short answer: you don't find or buy a working setup — you build it from hypotheses and tests. Below: the setup formula, where to get ideas, how to test without burning budget, how to tell from the numbers what to fix, and how to scale what works.
What a setup is in plain words
A setup is a specific campaign recipe: where the traffic comes from, who sees it, what they see and where they go. If the recipe consistently brings back more money than the traffic costs, the setup works. If not, it's still just a hypothesis.
An example of how to write a setup: "Facebook → dating offer → Germany → men 35–55 → video with a meeting story → quiz pre-lander". Change even one element — say, the GEO to Austria — and it's a different setup that has to be tested again.
If you're new to affiliate marketing, start with Affiliate marketing from scratch and then come back here.
What a setup is made of: 6 elements
- Traffic source — where you buy or get traffic: Facebook, Google, TikTok, push, Telegram Ads or organic.
- Offer — what you sell and which action the program pays for: sign-up, lead, deposit. Payout models: CPA, CPL, CPS, RevShare and Hybrid.
- GEO — country and language. The GEO determines the payout, the click price and whether the offer is legal at all.
- Audience — targeting settings: gender, age, interests, devices, placements.
- Creative — the banner, video or copy that hooks and promises a benefit. How to make them: Creatives for affiliate marketing.
- Pre-lander — the page between the creative and the offer that warms the user up. More: Pre-landers in affiliate marketing.
The tracker isn't part of the setup, but without it you can't measure one. Set it up before the first launch — see Trackers for affiliate marketing and Postback and S2S tracking.
Where to look for working setups
None of the methods below gives you a ready working setup. They give you hypotheses — your test is what makes a setup work.
- The affiliate manager. Ask which GEOs and sources currently bring volume and good approval rates, and which creatives and pre-landers convert. The manager sees stats from every affiliate on the offer. How to choose a program: How to choose a CPA network; the list of programs is in our affiliate program catalog.
- Spy tools. Look for creatives that have run the longest: if an ad isn't switched off for weeks, it's most likely profitable. Overview: Spy tools for affiliate marketing.
- Cases and the forum. Don't copy the whole setup (by the time it's published it has often burned out) — take the mechanic: the creative angle, the pre-lander structure, the targeting approach. Ask in the Questions section.
- Adaptation. Move a working approach to a new GEO, language or source. It's one of the most reliable methods: the mechanic has already proven itself.
- Your own hypothesis. The formula: audience pain + offer benefit + platform format. For example: "people aged 25–40 tired of their job" + "free career course" + "short story video on TikTok".
Can you buy a setup? Almost never worth it. If a setup makes money, it pays the author more to run it than to sell it to dozens of people who will instantly "overheat" it. What gets sold has usually burned out already or never worked.
Example setup hypotheses
These are examples of how to write a hypothesis, not ready working setups — they need to be tested:
- Push In-Page → sweepstakes (SOI) → US → Android mobile → "check if you've won" banner → no pre-lander. On the vertical: What sweepstakes are.
- Facebook → dating → Germany → men 35–55 → story video → quiz pre-lander.
- TikTok (organic) → language-learning app → Brazil → 18–30 → "before and after" videos → profile link.
- Google Search → utility or VPN → Tier-1 → intent search queries → advertiser's landing page.
For gambling, sources and funnels are covered separately in Gambling affiliate marketing: how to start.
How to test a setup: step by step
- Write down the hypothesis. Source, offer, GEO, audience, creative angle. Without notes, in a week you won't remember what exactly you tested.
- Check tracking. The tracker, subid and postback must be verified with a test conversion before launch.
- Make 3–5 creatives with different hooks and keep the other elements the same.
- Set a budget. The rule of thumb is at least 2–3 offer payouts per hypothesis. With a $20 payout that's $40–60.
- Let the campaign gather data. Don't tweak settings every hour — wait for clicks and the first conversions.
- Decide by the numbers: cut, fix or scale.
The math: when a setup pays off
To see whether a setup has a chance, calculate the break-even point. The formulas:
- EPC (earnings per click) = payout × CR;
- a setup is profitable when EPC is higher than the click price;
- break-even CR = click price ÷ payout.
Example. The payout is $20 and a click costs $0.10. Break-even CR = 0.10 ÷ 20 = 0.5%, i.e. 1 conversion per 200 clicks. A $40–60 test budget buys 400–600 clicks, which at a 0.5% CR means 2–3 conversions. If that budget brings zero conversions, the hypothesis almost certainly needs to change. If there are conversions but CR is below 0.5%, the setup is still losing and needs fixing. The formulas are covered in detail in Affiliate math: CTR, CR, EPC, ROI.
Diagnostics: what to fix by the numbers
- Low CTR — the creative doesn't hook or the audience is wrong. Change hooks, visuals, targeting.
- CTR is fine, CR is low — the creative's promise doesn't match the page. Check the pre-lander, the offer and creative match.
- Leads but low approval — untargeted or incentivised traffic. Narrow targeting, cut placements, make the creative more honest.
- Conversions but negative ROI — expensive clicks or a low payout. Cut losing placements and ask the manager for a higher payout.
- Positive ROI — the setup works and can be scaled.
The main rule: change one element at a time. If you change the creative and the audience together, you won't know what worked.
How to scale a working setup
- Vertical scaling — same setup, more budget. Raise it gradually, in steps, and watch whether ROI holds. How Facebook campaigns are structured for scaling: Facebook Ads campaign structure.
- Horizontal scaling — new GEOs, creatives, placements and audiences built on what already works.
Burnout. No setup works forever. Signs: CTR drops, clicks get pricier, frequency rises. Refresh creatives ahead of time and keep the next hypothesis in testing while the current one is still profitable. AI tools help make new variants quickly — see AI tools for affiliate marketing: the full stack.
Common beginner mistakes
- Buying "working setups". You pay for something that has already burned out.
- Testing without a tracker. Without data you can't tell what worked.
- Too small a budget. A conclusion from 50 clicks is guesswork.
- Changing everything at once. One change at a time.
- Quitting after the first loss. A losing test with clear numbers is still a result: you know what to change.
- Running offers that are illegal in the GEO. Such traffic often goes unpaid, and accounts get banned.
FAQ
What is a setup in affiliate marketing in plain words?
It's a campaign recipe: traffic source + offer + GEO + audience + creative + pre-lander. If the recipe brings in more than the traffic costs, the setup works.
Where can a beginner get a working setup?
Ask your affiliate manager which GEOs and sources work right now, look at long-running creatives in spy tools and build your own hypothesis. Only your test makes it a working one.
How much money do you need to test one setup?
The benchmark is at least 2–3 offer payouts per hypothesis. With a $20 payout that's $40–60. The exact amount depends on the click price and expected CR — calculate it with the formula from the math section.
How long does a working setup last?
It varies: from a few days to a few months. It depends on competition, the source and how fast creatives burn out. That's why the next hypothesis is tested while the current one is still profitable.
Should you buy setups?
Almost never. A working setup pays more to run than to sell. Put that money into your own tests instead.
Can you find a setup with no budget?
Yes, with organic traffic: TikTok, Reels, Shorts, Telegram. You don't pay for clicks, but you need time and content. Where to start: Affiliate marketing without investment.
Discuss your hypotheses and get advice on a setup in the Questions forum section, and follow new cases and breakdowns in our Telegram channel.
The setup hypotheses in this article are illustrative and are not guaranteed working setups. The "2–3 payouts per test" rule is a practitioners' benchmark, not a standard; adjust it to the click price and vertical.
