Outsourcing Fan Chatting: What You Gain, What You Lose

Outsourcing fan chatting means paying another company to hold the conversations your revenue depends on. It is a normal decision, made every day, and it is not the same decision as hiring. When you hire, you buy hours you can watch. When you outsource, you buy an outcome you cannot watch, from people you will never meet, writing as creators who trusted you with their names.

What follows is what genuinely transfers, what stays behind, and the audit protocol that replaces the supervision you gave up.

What does outsourcing fan chatting actually cover?

It covers the writing and the rota, and almost nothing around them. The pitch that providers make (“we take care of the chatting”) describes one function out of the four that running a chatting agency consists of. Here is the split that holds in practice, whatever the sales deck says:

Function	Typically the provider	Typically still yours
Writing replies in the creator’s voice	Yes	No
Shift coverage and rota	Yes	No
Pitch timing, within a grid	Yes	The grid itself
Voice guide and stated limits	No	Yours to write and update
Price floor and discount rules	No	Yours to set
Fan history and conversation record	Sits in their tooling	Yours to demand an export of
Creator relationship and complaints	No	Yours, always
Quality control	Their internal QC	Your independent audit

Read the right-hand column before you sign anything. Every line in it is work that does not disappear when you outsource, and agencies that budget only for the left-hand column discover the rest in month two.

What do you actually gain?

Speed, and a cost that moves with volume. Those are the two real gains, and they are worth a lot when they are the constraint you have.

• Coverage in days rather than after a hiring cycle. A hiring pipeline is a sourcing, testing and training sequence. A provider already has people who know the job, even if they do not yet know your creator.
• Turnover stops being your problem. When someone quits at a provider, their rota gets patched by someone else. When someone quits in your team, your evening block goes dark.
• A variable cost instead of a fixed one. An in-house team costs the same in a bad month; a revenue share does not. If your volume is seasonal or unproven, that difference is the entire argument.
• Hours and languages you cannot staff. Evenings maximise conversion in our data and the 2am-6am window collapses it, so the coverage worth buying from a provider is somebody else’s evening, not your own night.

Note what is absent from that list: quality. Outsourcing is a capacity and cash-flow decision. Nothing about it makes the writing better, and the good providers will tell you so themselves.

What exactly do you lose control of?

Four things, and they are the four that decide whether a conversation ends in a sale.

1. The voice. Someone you have not met writes as a creator you represent. Fans can tell. Not all of them, but the ones who spend most, because they have read the most of her.
2. The timing of the pitch. This is the expensive one. Across our corpus, a sale pitched before the sixth message converts roughly a third worse than one placed after about ten exchanges. A provider under volume pressure pitches early, systematically, and never tells you.
3. The price floor. Discounting is the fastest way to hit a monthly target and the slowest thing to undo, because a fan who bought at a discount has learned what the real price is.
4. The record. Fan history, notes, segmentation, who has already been offered what. If it lives only in the provider’s tooling, then leaving them costs you the memory of every relationship, not just the labour.

None of the four shows up in a monthly revenue figure, which is the only thing most agencies ever look at.

Which contract clauses decide whether this works?

The commission base, transcript access, the named team and offboarding, and they are rarely the ones negotiated hardest.

• The commission base. Gross sales or sales net of the platform’s cut. Two different numbers, one recurring gap. Write it in words, not just a percentage. The same trap sits one level up, in the split you agreed with the creator herself, and the full stack is in creator agency commission.
• Transcript access and cadence. Named creator, named week, readable threads, an agreed export format. Without this clause you have no audit, only a report written by the party being audited.
• The named team and the substitution rule. How many people write for your creators, and what happens when one leaves. A provider who will not name a team size is describing a pool, and a pool is where voice consistency dies.
• Offboarding and data return. What comes back to you, in what format, how fast. Agree this while everyone is happy; nobody agrees it later.

Cost the arrangement the same way you would cost a team of your own: the six-line worksheet in what a chatting team costs applies to a provider too, with the supervision and audit lines filled in rather than deleted.

How do you check work you no longer see?

By sampling on a rule, not at random, and reading against a written grid. Random reading produces panic over one bad thread and blindness to everything else.

The protocol that works is small enough to actually run every week:

1. Pull four threads per creator: one where the fan pushed back on price or complained, one that sold, one that did not, one that went dormant. The first of the four is the one you would skip in-house and cannot skip here.
2. Read them against the same grid your own team is scored on: voice, timing, price floor, handover continuity.
3. Count the five markers below rather than forming an impression.
4. Send the count back to the provider with the threads attached. A count gives them something to argue with. An impression gives them nothing, which is why nothing changes.

This is the same discipline as managing fan conversations in-house. Outsourcing does not remove the reading; it removes your excuse for not writing down what you read.

Which markers in a transcript show a provider drifting?

Five, all countable by a person with no special tooling, all drawn from what we can measure across 42,852 conversations and 38,879 sales.

What you count in the sample	What it means
Position of the first pitch in the thread	Before the sixth message, conversion drops by roughly a third; the optimum sits after about ten exchanges
Sales messages ending in an ellipsis	The worst-performing closer we have measured
Closed questions at the moment of closing	Costs several points of conversion against an open one
Sales attempted between 2am and 6am	The window where conversion collapses; evenings maximise it
A personal callback placed at the exact moment of the pitch	Costs points, even though it feels like it should help: the most counter-intuitive result in our corpus

Four of the five are habits a provider can correct as soon as it is shown the count. The fifth (the 2am-6am attempts) is a scheduling problem on their side, and it tells you they are staffing for presence rather than for sales.

What can outsourcing never fix?

An account with no audience, a creator with no stated limits, a price grid that does not exist. A provider handed those three will improvise all of them, in your creator’s name, and you will find out at the point where a fan quotes a promise back to you.

It also cannot fix the arithmetic underneath. If chatting costs you a share of every sale whether you employ people or contract them, the question stops being who writes and becomes what share of each sale the writing costs. That is the reasoning behind pricing justonedash at 20% of sales with no subscription: a conversation layer that costs something only when it collects something.

Whether that beats keeping the work inside your own walls is a separate decision, and it has a clear answer in four specific cases (in-house versus outsourced chatting sets them out side by side).