The warm hub: where the person who already subscribed lives

The warm hub: where the person who already subscribed lives

A messenger channel has no recommendation feed — the people here have already decided to read you. So content from a social feed does not work, and reach stops being a metric. Below: what replaces it and which formats actually lead to a direct message.

A messenger channel has no recommendation feed — the people here have already decided to read you. So content from a social feed does not work, and reach stops being a metric. Below: what replaces it and which formats actually lead to a direct message.

The channel has run for three years and nobody writes

A diagnosis I made of myself, looking at my own figures: three years of a channel and an empty inbox.

People read. They react. They do not unsubscribe. And not one question of "how do I order from you" or "how much is it". Not because it is uninteresting — because there is no occasion. The channel was a place where things are told rather than a place where things are asked.

That is the most common ailment of expert channels. A loyal, silent audience. And everything is measured by the wrong things: subscribers and views, figures that grow by themselves and say nothing.

How this differs from a social feed by function

The difference is not in formats but in who is reading.

A social feed is the cold top. The person landing there does not know you and is not looking for anything. The task is to catch him and take him further.

A channel is the warm hub. Here is somebody who has already subscribed. He knows you, he agreed to receive messages from you. He does not need convincing of your competence — that work is done.

The consequence almost everyone ignores: the content has to be different. A post that catches a stranger in a feed is, to a subscriber, a repetition of the known. And material useful to a subscriber will not stop a stranger.

Copying the same thing into both is the most widespread mistake. It produces a mediocre result in both places.

What is measured in a warm hub

Not views. They depend on the size of the channel rather than the subject. By my observations the median of normalised views across content categories falls within a very narrow range — that is, views will not tell you a good piece from a bad one.

Not subscriber count. It grows from external traffic and has no connection to money.

What actually shows whether the channel is working:

Direct messages. The main metric of a warm hub. My own baseline was three years of zero, and that is more honest than any analytics.

Shares. In a niche where everyone writes about the same things, a share is the scarcest reaction.

Substantive comments. Not "thanks" but their own case, their own figure, their own objection.

Answers to direct requests. If you asked people to name a sum or choose a direction and nobody answered, either the request is unclear or there is less trust than you thought.

What works in a warm hub

A full analysis of one case. A format almost nobody uses: from first contact to payment, with the platforms named, the sums and what went wrong. A subscriber does not need general advice — he needs to see how it actually happens.

Showing the failure first. An analysis of a mistake works better than average and occupies a small share of most feeds. For a warm audience it is also a signal of honesty.

Figures nobody else has. Your own observations, scrapes, counts. In my case, 5 986 jobs over four months; no competitor has that source, and it is what the channel is read for.

Direct questions that can be answered in one message. Not "what do you think?" but "tell me what you charged: the work, the sum, the platform". The difference is that the second has a specific answer.

One thesis per publication. A warm audience reads fast and does not finish long pieces. Two thoughts in one post means neither will be remembered.

What must not be done here

A product showcase. Promotional posts in a channel living on useful content get noticeably less reach and fewer reactions than the rest — by my observations roughly a quarter and a third respectively. With a small audience a regular showcase kills the channel within a month.

News and releases. Large aggregators cover those. And they address a rung your audience has passed.

A daily feed. The relationship between frequency and engagement is negative. Three or four publications a week is the working guide.

The "what do you think?" question at the end. It raises comments and lowers reactions.

Length and rhythm

Two technical things that strongly affect whether a piece is read.

Length. A warm audience reads fast and on a phone. Material that does not fit into a screen and a half gets finished noticeably less often. If the subject does not fit, that is not a reason to write longer but a reason to split it in two.

Rhythm. Short paragraphs of two or three sentences, emphasis on key thoughts, lists where there is enumeration. Solid text in a messenger does not get read at all — not from laziness but because the eye has nothing to catch on.

And one technique that almost always works: the first line should be a complete thought. Only that line is visible in the notification preview, and the decision to open or dismiss is made on it.

How a channel stops being silent

Three things, and the first matters more than the others.

Give people an occasion to write. Not "get in touch if anything" but a specific request: name a sum, choose a direction, send me your case. Silence more often means no occasion than no interest.

Publish things people can disagree with. Material everyone agrees with produces no dialogue. A position that can be disputed does.

Start a strand produced by the audience. A showcase of subscribers' work and jobs works three ways: it reinforces the author of the work, it serves as proof to the hesitant, and it gives you material you do not have to invent.

And an alarm threshold. If after five to seven consecutive publications the inbox is still empty, the problem is not the subjects. The problem is that the audience does not see you as somebody to ask. That is cured not by content but by directly inviting questions.

Strands instead of scattered posts

A separate thing that distinguishes a living channel from a feed of random notes.

A permanent strand creates expectation. The reader knows that Tuesday brings a case analysis and Friday a showcase of work. Expectation is what makes a channel get opened rather than scrolled past in a general list.

Two or three strands are enough. More and they stop being recognisable.

The best strand is the one the audience produces. Subscribers' results, their questions, their cases. It requires no ideas from you, reinforces those who appear in it, and works as proof for the rest.

And a rule: a strand starts once there is a month's worth ready. A strand announced and abandoned after two weeks does more harm than none at all.

What the business gets

By the methodology I use, this platform is part of the social packaging stage — 4–5 days. Part of full positioning: 3–4 weeks, 25–35 hours, around $2 000 for the package. Honestly: that is the price of the service by the Freelance 2.0 course methodology, not market statistics.

Do it yourself or commission it

Doing it yourself. The mechanics above are the "Channel" block of "Freelance 2.0": a post generator from a subject or a voice message, working from the knowledge base. Plus the machine cliché detector — in a warm channel artificial text is visible instantly and costs more than anywhere else.

The course costs $9 and is included in Basic. Registration is free. Three days of the Basic plan switch on with one button, no card needed: Basic includes 200+ AI assistants across 14 categories and the ten-lesson sales module.

Commissioning it. The service exists. The sign of it being done wrongly: if a contractor promises growth in subscribers and views, he is selling a metric with no connection to money.

Do one thing today: count how many direct messages came from the channel in the past month. If it is zero, publish one piece ending in a specific question with a short answer. Not "share your thoughts" but "tell me your figure". The difference will be visible the same day.

What this looks like on a real project

Below is the same work on a live client: a dental studio run by two doctors, a husband and wife. The work was delivered in eleven sections, the sixth being the channel as a warm hub.

Why this example is here. It shows that positioning is not an abstraction but a service with a clear composition. And it sells to any business with an audience and a product: a clinic, a school, an agency, a shop. The mechanics are the same; the content changes.

An analysis of a deal like that from first message to payment is the “First order” series module of the course “Sales”, Basic plan.

The platform's job

The person already knows you but has not yet chosen. Here he ripens: posts warm the trust and the bot takes him to a booking.

27 posts across nine weeks, three a week. Month one is who you are, the priority segment, differentiation, conversion. Month two goes deeper on five pillars.

The bot switches on in the second week and the cycle closes with conversion.

Two voices in the first person

The texts are written not "from the clinic" but from specific people, and the voices are kept apart.

Zakhar — calm and direct, about results. "I take on what others turned down." Surgery, bite, complex cases. The first line of a post: "They said five teeth had to come out. I saved them."

Alexandra — warm, an artist's hands, evidence-driven. "I check rather than take anyone's word." A gentle entry, treatment, aesthetics. The first line: "I take no authority on trust."

And paired posts — the two in turn, like a live conversation in front of a patient.

Five content lines and their own words

Fear → APPOINTMENT, TYPE, SOONER · Cause → BITE, ANALYSIS, FOCUS · Truth → CERAMIC, NOTES · Aesthetics → NATURAL, LESS.

The same words as in the lead magnets — a post leads seamlessly into the bot, with no separate "follow the link".

The prompt for posts and the data behind them

Freelance 2.0, prompt №32 "Channel posts from a subject or a voice message", Basic plan. The raw material is a subject or a voice transcript: what gets dictated on the move becomes a post.

Prompt №4 "Competitor analysis by channel" from the same course — about the platform rather than your own channel. The key thought from it: this is an honest environment for analysis, because views are not inflated and comments reflect real interest. For each competitor, the top ten posts with metrics and an analysis of which thematic categories reliably catch on.

Who helps run a channel. Assistants, Basic plan: Funnel analyst — every stage checked against a list. TED editor for messengers — the structure of a talk adapted to the channel format. Copywriter.

Skills for a warm channel

Both are advanced skills: the files come with the Creator plan, or for a one-off $9, or for 1000 XP.

What to take on the platform for this task

What to read next

Lead magnets behind code words — the main mechanic of this platform.

A bot that segments on the fly — what stands behind a code word.

The cold top of the funnel — where people arrive here from.

Four scripts and two simulators — what to do once somebody has written.

AI agents: what they are and what they pay — the full guide to agents, with market rates.

Also on AI SKILLS