Why no model is always the best
CPA, RevShare and Hybrid serve different purposes. CPA makes the result more predictable at the level of an approved qualifying action. RevShare links revenue to the audience’s subsequent value. Hybrid combines the two approaches. The choice depends on your source, acquisition costs, user quality and the time horizon you can sustain.
A comparison in one table
| Criterion | CPA | RevShare | Hybrid |
|---|---|---|---|
| Cash flow | Faster and clearer | Depends on subsequent activity | Part fixed, part variable |
| Affiliate risk | Lower after the action is approved | Higher over a short horizon | Moderate |
| Long-term revenue potential | No | Yes | Partial |
| Important metric | Cost per confirmed FTD | NGR and retention | Both groups of metrics |
| Calculation complexity | Low to moderate | Moderate | Above average |
Scenario 1: paid traffic with high acquisition costs
Suppose an affiliate spends $4,000 on advertising to generate 50 confirmed FTDs. At an illustrative CPA rate of $110, revenue is $5,500, leaving $1,500 before operating expenses. The calculation is straightforward: if the cost per FTD exceeds the affordable level, the campaign needs optimization.
RevShare may produce less revenue in the first month because the audience has not yet accumulated activity. With strong retention, however, some revenue continues to arrive later.
Scenario 2: an SEO project with long-lived pages
A content page can bring users for months. A fixed CPA captures each user’s value when the action is approved, while RevShare lets the affiliate participate in subsequent activity. If the source is sustainable and the audience is of good quality, a longer horizon becomes an argument for RevShare.
There is also a cost side: an SEO project requires content production, technical work, links and editing. Treating its traffic as free gives a misleading calculation.
Scenario 3: scaling after a successful test
Once a campaign consistently generates FTDs, Hybrid can ease some cash flow pressure. Its fixed component helps recover part of the advertising budget, while RevShare retains exposure to users’ long-term value.
An illustrative example:
- 40 FTD × $45 fixed component = $1,800;
- $7,000 NGR × 20% = $1,400;
- total under the Hybrid formula = $3,200.
Compare this result with CPA or pure RevShare over the same period and for audiences of comparable quality.
A simple selection framework
- If you need a transparent price for an approved action, examine CPA.
- If your source lasts a long time and audience quality matters more than rapid budget recovery, consider RevShare.
- If you want a fixed component alongside long-term revenue potential, discuss Hybrid.
Use the calculators for your own scenarios. The CPA, RevShare and Hybrid pages explain the individual models.
Compare the models over the same horizon
Comparing one week of CPA with one week of RevShare can be misleading. CPA shows a cash result sooner, while RevShare develops as audience activity accumulates. Choose a common horizon, such as 60 or 90 days, and compare what the same cohort generates over that period.
For paid traffic, record the cash flow gap separately. Even if RevShare performs better by day 90, the campaign may require more working capital during the first weeks. That affects the ability to scale.
Avoid choosing a model from one successful source
SEO, paid social and influencer traffic can have different lifespans and user profiles. Conclusions such as “we always need CPA” or “RevShare is always more profitable” are too broad. Match the model to the source and GEO.
A decision matrix
Score each model from 1 to 5 against four criteria: speed of budget recovery, predictability, participation in LTV and the effort needed to verify calculations. Then weight those criteria for your team. A media buyer with limited working capital may prioritize speed, while a content project may value longer-term revenue more highly.
The matrix does not produce a universal answer. It makes the reasons for choosing one option explicit.
Run a fair internal experiment
If several models are available, avoid switching all traffic after one week. Select comparable segments and define the comparison period in advance. For CPA, measure revenue from approved actions. For RevShare, measure accumulated revenue at a fixed cohort age. For Hybrid, measure the sum of both components.
Account for operational complexity separately. A model that generates 5% more on paper but requires much more working capital or manual reconciliation may be less suitable for your team.
Record the decision
Summarize the choice in a few sentences: the source, GEO and horizon; why this model is preferable; and which deterioration in metrics would trigger a review. This prevents the team from remembering only the selected rate a month later and forgetting the reasoning behind it.
LuckyBear RevShare rates
The current tier structure starts at 55% for V1 and reaches 75% for V5. The rate depends on your status; the maximum is not the starting rate for every account. Compare models using your actual rate rather than the maximum. See the affiliate tiers.
