Replies · Purchases · Refusals · Time
Before changing a script, check replies. Before changing a PPV price, check purchases and objections. Before sending more follow-ups, check silence, refusals and the time since your last contact. Total revenue tells you something changed; these signals help you choose an action.
Start with Unly’s dashboard to see how your accounts are performing over the selected period. Then narrow your review to the account, media and conversations behind the movement. A quiet week does not tell you whether the problem is your opening message, the offer, its price or the audience receiving it.
Your review should end with one decision: keep the current setup, investigate a specific part of the conversation, or test one change. Write down the evidence and the number you expect that change to affect.
Choose the KPI that matches the decision
Group fans by purchase history first, then by recent activity: non-buyers who reply, non-buyers who stay silent, buyers who reply, and buyers who have gone quiet. Keep each fan in one group for the comparison. For a welcome-message review, narrow the group to new arrivals. A returning buyer already knows your content; a new fan may still need to understand what you offer.
| You want to change | Read first | What the conversations can explain |
|---|---|---|
| Opening message or script | Reply rate, time to reply and how far the conversation gets before an offer. | Is the question easy to answer? Does it respond to the fan’s actual message? Does the sale start before their preferences are clear? |
| PPV price | Purchases for the same offer, sales per recipient and explicit price objections. | Did the fan object to the price, ask what was included, or receive content they had already bought? |
| Follow-up frequency | Time since last contact, replies after a reminder, silence and refusals. | Was the follow-up tied to a request? Had the fan already declined? Are you repeating an unanswered pitch? |
| Media or pack | How often it was offered and purchased, the amount sold, and repeat purchase history. | Does the description match the content? Is the offer relevant to this fan? Is a pack being compared with a single item? |
If fans answer the opener but stop when the offer arrives, read the offer before rewriting the whole script. If they ask what the video contains, clarify the description before discounting it. An explicit “too expensive” gives you different evidence from an unopened message.
For the next action, use the guides to customizing AI replies and PPV prices and following up without spamming fans.
Write down what each rate counts
A rate needs a numerator, a denominator and a time window. Count fans once when measuring how many people reply or buy. Count transactions when measuring purchases. A fan who buys several items can raise the transaction count without adding another buyer.
Reply rate
Recipients who replied within the chosen window ÷ fans who received the selected message.
Choose the reply window before comparing messages. Exclude fans who never received the message; keep repeated sends to the same fan from inflating the audience count.
Buyer rate for one offer
Recipients who bought that offer within the chosen window ÷ fans who received it.
Keep the media, price and time allowed to purchase consistent. A rate based on message opens answers a different question from a rate based on all recipients.
Sales per recipient
Sales from those recipients for that offer within the chosen window ÷ fans who received it.
This helps you read price and purchase rate together. Use one currency and one definition of the sales amount throughout the comparison.
Refusals and review time
Record the reason for each clear refusal and minutes spent reviewing the group.
Silence is not a confirmed price objection. Separate pricing, content mismatch and requests to stop; include the time needed to correct confusing replies.
These are working definitions for your review. Use the product’s stated calculation when reading a dashboard rate, and name a separately calculated rate clearly. Exclude unrelated purchases and purchases linked to earlier sends. If you cannot establish the recipients or link purchases to the offer under review, keep the available counts; you cannot yet calculate that recipient-based buyer rate.
Compare equivalent periods and offers
Start with two completed periods of equal length covering the same weekdays. Keep the creator, platform, fan group, offer and send frequency as consistent as possible. Record a promotion or a change in incoming traffic: either can change the audience before your script gets a chance to work.
Give both groups the same time to respond and buy. Comparing last week’s offer after seven days with today’s offer after two hours makes the newer offer look weaker by construction. Check the counts behind a percentage; a single purchase can move a small group sharply.
The calculation before a price cut
If you test a single offer at $16 instead of $20, matching the previous gross sales requires $20 ÷ $16 = 1.25 times as many purchases. That is 25% more purchases, assuming every purchase uses the stated price.
This calculation matches gross sales at those two prices. After the test, check actual sales and net earnings; a higher purchase rate alone does not tell you whether the price change paid off.
Also inspect whether one large buyer accounts for most of the movement. Keep their purchase in the total, but distinguish it from a change across the wider fan group. For deciding whether a price or qualification problem comes first, read the PPV pricing guide for OnlyFans and MYM.
Keep sales, net earnings and payouts separate
A sale, the creator’s earnings from it, and a bank payout answer different questions. Use sales to study an offer. Use net earnings to see what remains after the platform’s deductions. Use payout status to track money available to withdraw or already paid.
Fansly’s Statistics page distinguishes Gross, Earnings, Purchases and Paying fans. Its Earnings figure is net of the platform’s share and refunds. Custom periods use UTC days, while Today and Yesterday earnings use local days. Match those date boundaries when comparing figures.
MYM explains its commission calculation separately from its payout process. A later bank transfer does not by itself mean the offer sold later. Compare the Unly sales figure with the corresponding sales activity, then check native net earnings and payouts on their own terms.
Turn your Unly review into a concrete action
Open Unly’s dashboard, choose the period and find the account you need to investigate. In that account’s statistics, open the Media tab. Compare Paid sends, Unique offers and Purchases, then look at reached fans, buying fans and revenue per reached fan. More sends and more distinct buyers are different outcomes.
Keep unique offers separate from unique fans when reading a rate: one fan can receive several different offers. Use the media ranking and script breakdown to find the offer or step worth reviewing. Keep the investigation on that account rather than changing every account because the overall total moved.
Read the relevant conversations in the platform’s own messaging interface. Note where fans stop replying, what they ask before buying, and whether a refusal concerns the price, the content or continued contact. Keep these notes beside your figures so a quiet conversation does not become a guessed price objection.
Connect the numbers to the settings
Unly lets you configure AI instructions and manage media offers. Once the review identifies the problem, change the relevant instruction, media description or offer price. Keep the other parts of the offer steady so the next review has a clear question to answer.
Explore Unly’s OFM software to bring account monitoring and offer configuration into your workflow.
Run one change with a clear review point
Write the question before editing the setting. “Will a clearer description help interested fans understand this pack?” is specific enough to test. “Will a better script make more money?” leaves the message, price, content and audience all open to change.
-
Save the starting point
Record the current text, media or pack, price, fan group and comparison dates. Save the counts and amount definition behind each KPI.
-
Change the suspected cause
If fans keep asking what is included, replace “You’ll love this” with a description of the actual content, format and length. Keep the price and recipient selection rules consistent; exclude fans who already bought that item.
-
Choose the measure and review window
For that description change, record clarification questions, purchases and sales per recipient. Allow the same time to buy in both groups. Keep an eye on refusals and extra work answering confused fans.
-
Decide what the evidence supports
Keep, revise or undo the change based on the figures and conversations together. If the group is too small or traffic changed substantially, extend the observation or repeat the comparison before drawing a conclusion.
A before-and-after comparison helps you choose your next move; simultaneous changes to traffic, audience or content can still explain the result. Record them alongside the test so you know which conclusion you can carry forward.
Start your next review in Unly with one account, one offer and one question. Try Unly, record your starting figures, and use the conversations to decide what deserves to change.