How Direct Messaging Drives Creator Revenue: The Data
Quantifying the revenue lift from DM-based creator monetization using BSP analytics and creator survey data.
The statistic that does not mean what it says
Every article about messaging opens with the same number: a 98% open rate. It is the most cited figure in the category and the least understood.
The number is not fabricated. Benchmarks from Klaviyo, Attentive, Postscript and Validity converge on a 95-98% range for opt-in business messaging. What almost nobody states is how it is derived: it is estimated from delivery confirmations, response behaviour and lock-screen expansion analytics, not directly instrumented. It measures device-level message expansion, not active reading, and because SMS preview text is visible from a locked screen, an "open" can be recorded for a message nobody read.
So the honest reading is that a text message achieves near-total structural exposure. It lands on a surface the recipient will see, with no spam folder and no algorithmic feed deciding whether to show it. That is a genuine and unusual property. It is not a claim that 98 of every 100 recipients engaged.
The comparison to email makes the problem worse in both directions. Email open rates are reported anywhere from 20-28% to a 2025 average of 43.5%, with an industry range of 30.1% to 55.7% depending on the dataset. And email opens are themselves inflated: Apple's Mail Privacy Protection preloads message content and images for Apple Mail users whether or not anyone opens the message. Comparing an estimated SMS exposure figure to a pixel-inflated email open rate compares two different behaviours, measured by two unreliable instruments, and calls the difference a result.
The metrics that survive inspection
Click rate, reply rate and conversion are directly instrumented, and they are the numbers worth planning against. Klaviyo publishes benchmarks from platform-level data across its customer base:
- Flow-based messages, automated and triggered by something the recipient did, average click rates approaching 10%, roughly double campaign performance, with top performers above 16%.
- Campaign click-through bands at 8.9-14.5% for good performance and 14.6% or better for excellent.
- Flow click-through bands at 10.9-14.9%.
The gap between flows and campaigns is the most actionable finding in the dataset. A message triggered by an individual's behaviour outperforms a broadcast to the same list by roughly a factor of two. That is not a copywriting effect; it is a relevance and timing effect, and it is reproducible enough to appear in aggregate platform data.
Where the data runs out, stated plainly
There is no public dataset of per-creator direct-message conversion rates. None.
Klaviyo's benchmarks are drawn from a customer base weighted heavily toward ecommerce. Applying an ecommerce flow benchmark to creator fan messaging is an extrapolation across a different audience, a different purchase motive and a different price point. It is a defensible starting estimate and it is not a measurement of creator DMs.
Anyone quoting a specific creator DM conversion rate, "creators see a 34% response rate on paid messages" and similar, is quoting a vendor's marketing page. The recycled claims to distrust most are exactly the ones that travel without provenance: a bare "98% open rate", a "45% response rate", or a large ROI multiple with no statement of whether it came from a platform report, a survey, or a single case study.
| Metric | Figure | Instrument | What it actually measures |
|---|---|---|---|
| SMS "open rate" | 95-98% | Estimated from delivery confirmations and lock-screen expansion | Structural exposure, not reading |
| Email open rate | 20-28%, or 43.5% by one 2025 platform dataset | Tracking pixel | Inflated by Apple Mail Privacy Protection preloading |
| SMS campaign click rate | 8.9-14.5% good, 14.6%+ excellent | Klaviyo platform data | Clicks, directly instrumented |
| SMS flow click rate | About 10% average, 10.9-14.9% band, 16%+ top decile | Klaviyo platform data | Clicks on triggered messages |
| Creator DM conversion | No public figure exists | None published | Unmeasured |
The last row is the most important one in the table, and it is the row every other article omits.
What is measurable about creator direct-message revenue
The size of the pool is measurable, because one platform files audited accounts.
Fenix International Limited, which operates OnlyFans, reported $6.63 billion in gross payments for the year ended 30 November 2023, up 19% from $5.55 billion, with creators receiving 80%, roughly $5.30 billion. At year end the platform held 4.1 million creator accounts and 305 million registered users.
Subscriptions and pay-per-view messages both settle through that same pool, and the company does not break out message-driven revenue separately. So the pool is audited and the split inside it is not public. Any figure claiming that some specific percentage of creator revenue comes from paid DMs is an inference, not a disclosure, and the inference is usually made by someone selling messaging software.
The survey side supplies the other bound. Across more than 9,500 creators, Linktree found 46% of full-time creators earning under $1,000 a year and only 12% clearing $50,000, with 66% treating creating as a side hustle. Direct-message monetisation is therefore being attempted by a population whose median outcome is small. That is precisely why the lift question matters: a channel effect that is marginal for a top-decile creator can be decisive for someone in the bottom half.
The filter is real, it is just not where you think
The structural case for messaging is usually stated as "no spam folder, no algorithm". The first half is true of the inbox and false of the path to it.
US application-to-person messaging passes through 10DLC registration, where a campaign's trust score sets its throughput. Carriers apply their own content filtering above that. FCC robotext rules govern what may be sent and to whom. Apple screens messages from unknown senders into a separate list. So messages are filtered: the filtering happens at the carrier and the registration layer rather than in a spam folder, and it is unforgiving of unregistered or poorly configured senders.
On iMessage specifically: Apple provides no public API for sending iMessage. Messages for Business is an approved-partner programme oriented around conversations a customer starts, not an open send channel. Any product claiming to blast iMessage at a fan list is describing something outside Apple's sanctioned surface, and that is a compliance question before it is a deliverability question.
The part that genuinely compounds
Strip out the disputed statistics and one structural property remains, and it does not depend on any benchmark being accurate.
A contact list you own is portable. An audience reached through a platform's feed is not. If a platform changes its distribution tomorrow, the list still resolves to the same people and the feed does not. That is the argument for owned messaging, and it is an argument about asset ownership rather than conversion rate.
The strongest claim messaging can honestly make is not that it converts better. It is that it still works when the algorithm changes its mind.
How to measure your own, since no benchmark applies to you
Because there is no creator-specific dataset, the only trustworthy number is the one you generate. Four rules make it interpretable:
- Report conversion over messages DELIVERED, not messages sent. Sent counts your intent; delivered counts what arrived. The gap between them is carrier filtering, and hiding it flatters every downstream rate.
- Never compute a rate over "opens". The open is an estimate derived from lock-screen behaviour. A conversion rate whose denominator is an estimate cannot be compared to anything, including your own result from last month.
- Hold back a control group. Send to 90% of a segment and nothing to the other 10%. Revenue in the holdout is the revenue you would have earned anyway. Without it you are measuring your audience's baseline purchasing and calling it campaign lift, which is the single most common way messaging ROI gets overstated.
- State the denominator every time you report a number. "12% conversion" is uninterpretable. "12% of 2,400 delivered messages, over 14 days, against a 10% holdout" is a measurement.
Those four rules cost nothing to adopt and they make your own data more valuable than any published benchmark, because your data is measured on your audience.
The answer, bounded honestly
How much does direct messaging drive creator revenue? The defensible answer has three parts:
- The pool is real and audited at $6.63 billion of gross payments in one year on one platform, with 80% reaching creators.
- The engagement metrics that survive scrutiny are click rates around 10% on triggered messages, roughly double broadcast performance, from platform-level data on a largely ecommerce customer base.
- The per-creator revenue lift is unmeasured in public data, and the widely quoted open rate measures exposure rather than engagement.
Anyone offering a single percentage for the lift is not reporting a measurement. The useful work is to instrument your own list with a denominator and a holdout, and compare it to itself.
For the sizing behind the pool figures, see what the research actually says about creator economy market size. For how the take rates compare across every channel, see the ranked breakdown of creator revenue streams.
Sources
Klaviyo SMS marketing benchmarks and Help Center guidance on SMS click rates. Analysis of the 98% claim's derivation and its convergence across Klaviyo, Attentive, Postscript and Validity per MessageIQ and Digital Applied. Email open-rate benchmarks and Apple Mail Privacy Protection distortion per MailerLite and HubSpot. Fenix International Limited FY2023 accounts at Companies House, as reported by Variety and Music Ally. Linktree Creator Report 2022 via TechCrunch.
