Before public launch, choose two subscription prices that are currently available in your creator settings, hold the page promise and launch inventory constant, and compare how your intended audience understands each offer. Use structured interviews plus a capacity model, record competing explanations, apply a written decision rule and choose one starting price. Treat the result as a launch hypothesis, not proof of future revenue.
This is a pre-launch experiment, not average-price reporting or a live price-change checklist. If the page already has subscribers, use the subscription price change checklist because notices, existing cohorts and renewal effects require a different decision process.
Test a defined offer, not a number in isolation
Write the subscription experience before choosing candidates: recurring content themes, visible launch inventory, expected interaction, creator boundaries and what is not included. A prospective subscriber evaluates a price in relation to that promise. Comparing two prices while changing the content description teaches nothing useful about the price itself.
Keep optional purchases out of the first question unless they are essential to understanding the page. The test asks whether the core subscription proposition is understandable and plausible at each candidate. It does not forecast every future purchase or lifetime relationship.
Use the OnlyFans pricing guide to map the wider monetization system. Return here with one stable core offer and the current price choices visible in the creator account.
Choose Candidate A and Candidate B deliberately
Candidate A should express the creator's current best starting hypothesis. Candidate B should test a meaningful alternative, not a trivial difference selected because it feels safe. Both must be eligible in the current account and compatible with the same content and interaction promise.
Write a reason for each. A may prioritise a lower commitment for an audience unfamiliar with the creator. B may signal a more focused experience supported by stronger launch proof. These are hypotheses to investigate, not universal pricing rules.
Do not copy an “average” from a roundup. Public creator pages differ in audience, offer, discounts, content visibility and additional purchases. An average cannot tell a new creator which promise they can fulfil or how their own intended audience interprets it.
Prepare two matched offer cards
Create two simple cards with identical identity, bio promise, launch inventory summary, boundaries and next step. Change only the displayed subscription price and a neutral label identifying the candidate. Keep typography, image, order and emphasis the same so visual treatment does not reveal which version the creator prefers.
Include enough page context for a reviewer to understand the proposition, but do not build a fake checkout or collect payment. This is a research artefact. Make clear that the page is not yet available and ask reviewers to respond to the offer as described.
Save the exact cards in the test record. If copy changes after the first conversations, start a new version instead of silently combining feedback from different offers.
Recruit relevant reviewers without treating them as buyers
Speak with adults who resemble the intended audience in interest and discovery context. Do not recruit only supportive friends or current collaborators. Record how each person was selected and whether they already knew the creator. A small relevant sample can reveal confusion, but it cannot estimate a reliable purchase rate.
Show candidate order in a balanced way so the same version is not always first. Ask what experience the offer appears to include, what feels unclear, which candidate they would investigate further and why. Then ask what information would change their view. Avoid “Would you pay this?” as the only question; hypothetical agreement is easy and not a transaction.
Keep personal data to the minimum needed for the research note. Do not record sexual preferences, private identity details or contact information merely to make the worksheet look rigorous.
Model capacity before selecting the starting price
Build a simple capacity view for each candidate using the same offer. List recurring production hours, fan-care commitments, tools, taxes and other ordinary costs relevant to the creator. Use ranges where inputs are uncertain. The purpose is to find promises that remain workable, not to manufacture a precise revenue forecast.
Write scenarios for fewer, expected and more subscribers without assigning unsupported probabilities. Ask whether the promised interaction and production remain sustainable in each. If a candidate only works when an unproven volume or purchase pattern occurs, note that dependency rather than hiding it.
Keep gross platform receipts, platform deductions, creator expenses and take-home amounts conceptually separate. This page does not provide tax advice. The capacity model exists to prevent a price decision from creating an offer the creator cannot deliver.
Separate observations from the decision
Log each reviewer's restatement of the offer, candidate preference, reason, missing information and prior familiarity. Summarise recurring themes without converting them into invented percentages. A repeated clarity problem may require an offer rewrite before either price can be chosen.
List confounders: different card order, different explanation, audience mismatch, existing creator loyalty or an offer edit. Decide whether the evidence is strong enough for a starting hypothesis, not whether it proves the objectively correct price.
If neither candidate is understood, stop and repair the promise. If both are plausible and evidence is mixed, choose the more sustainable starting hypothesis and record uncertainty. “Inconclusive” is preferable to selectively quoting the reviewer who agreed with the creator.
Copy the Pre-launch price-test card
| Field | What to record |
|---|---|
| Decision | Choose one starting subscription price before the page is publicly launched. |
| Audience | Describe the specific prospective subscriber and how the creator currently reaches them. |
| Offer held constant | Write the same content promise, visible proof, interaction boundary and launch inventory for both candidates. |
| Candidate A | Record one price currently eligible in the creator account and why it fits the offer. |
| Candidate B | Record one meaningfully different eligible price and the hypothesis it tests. |
| Evidence method | Use structured preference interviews, offer-comprehension review and a simple capacity model. |
| Confounders | List changes in content, presentation, audience sample or promotion that could distort the comparison. |
| Decision rule | State what evidence favours A, favours B or remains inconclusive. |
| Stop rule | End the pre-launch comparison on its stated review event; do not keep collecting opinions until one answer wins. |
| Launch record | Save the selected starting price, rationale, known uncertainty and next review event. |
Add reviewer IDs that do not reveal unnecessary personal information, card order, observations, interpretation, selected candidate, approval owner and launch date. Keep the rejected candidate so the decision can be understood later.
Set the launch and next review boundaries
Publish the chosen starting price with the offer used in the test. Record any change between research and launch. Do not launch both candidates through duplicate accounts or repeatedly switch a new public page to simulate a clean experiment; account context and subscriber effects make that comparison different.
Define the next review event using meaningful account evidence and delivery experience, not a promise that the price will change on a fixed date. Track visits or source context where available, subscriber starts, questions, delivery workload and cancellations as observations. Do not claim that the starting price caused each outcome.
Once live subscribers exist, move future decisions to the separate change checklist. Promotion efficiency belongs in the promotion ROI scorecard, not this pre-launch price decision.
Use a decision ladder when evidence is weak
First ask whether reviewers understood the same offer. If they did not, rewrite the promise and repeat the comprehension check before comparing prices. Second, remove reviewers who clearly fall outside the intended audience from the decision note without erasing their useful clarity feedback. Third, compare the two candidates against ordinary delivery capacity.
If one candidate is clearer and sustainable, select it as the starting hypothesis. If both are clear but preference is mixed, choose the candidate that leaves the creator more room to deliver the written promise consistently. If neither is sustainable, change the offer rather than searching for a price that disguises the mismatch.
Record the decision level reached, unresolved uncertainty and the evidence that would justify revisiting it. This prevents the creator from reopening the question after every isolated opinion. A disciplined inconclusive result is useful because it identifies the next missing decision instead of pretending the research was precise.
Sources and limitations
The UK Government Test and Learn guidance supports stating assumptions, testing them proportionately and using feedback to refine a decision. Verify eligible subscription controls and terms directly in the current creator account and the OnlyFans Terms of Service.
Limitations: preference interviews are not purchases, a capacity model is not a revenue forecast, and a pre-launch comparison cannot guarantee conversion or retention. The output is one documented starting hypothesis with a clear boundary for later review.
Interpret the test as an offer-comprehension check
Ask each eligible reviewer to restate what the subscription includes, what remains optional and why the candidate price feels consistent or inconsistent with that promise. Record the words they use before explaining the offer. Confusion about deliverables is a different finding from simple price preference and should trigger a clarity repair before another comparison.
Do not average a handful of opinions into a market benchmark. Note the reviewer context, the version shown, the date and whether the content proof was identical. A pre-launch comparison cannot predict retention, PPV behaviour or income. Its useful output is a supported starting decision, the assumptions behind it and a review date after real account evidence exists.