Lead magnet on the first message, a timed pre-launch drip, segmented offers, launch-week broadcasts, and a read on who is actually paying attention.
A course funnel is a pacing problem: a lead magnet, a week of warming, a cart that opens and closes on a fixed date, a replay window. Email carries all of that, but the delivery layer works against you — promotions tabs, bounces, double opt-in friction, hours between sending and being read.
Telegram changes the mechanics, not the strategy. A deep link drops someone straight into a chat with your bot: no form, no confirmation email, no deliverability layer in between. Every later message lands in that same thread, directly under the lead magnet the person actually opened.
Be honest about the trade-offs. Telegram is a poor home for a long-form sales letter — no rich HTML, real length limits, a reading posture that suits short beats — and unsubscribing is one tap on Block, far more final than ignoring an email. So split the work: Telegram carries the sequence and the urgency, your sales page carries the argument and the checkout.
Onboarding runs in a fixed order: /start, language pick, an optional category pick, an optional channel-subscription gate, then the welcome or offer message. For an info product, that welcome message is the lead magnet. It can carry media — a PDF workbook, a short video, a template — plus inline URL buttons linking the resource or the first free lesson.
The channel-subscription gate is the piece most course creators underuse. Turn it on and the magnet is released only after the person joins your announcements channel, so one deep link grows the list and the channel at once. On cold ad traffic you may want it off, with the channel invite moved into the follow-up.
Each magnet is a campaign with its own deep link, shaped t.me/yourbot?start=WORKBOOK. A campaign holds a promo code, a per-language message, media, inline buttons and an active window. Put a distinct code behind each traffic source and one funnel serves a YouTube description, a paid ad and a podcast mention.
Auto-messages fire on a delay measured from the moment someone joins — minutes, hours or days — and can be targeted by language, source, tag or category. Because the clock starts at join time, everyone gets the same relative pacing whether they arrived on a quiet Tuesday or during a launch spike. A workable seven-day skeleton:
The boundary that matters: drips are relative, launches are absolute. Anything tied to a fixed calendar moment — cart open, cart close, the live session — belongs in a broadcast, because a day-6 auto-message will happily reach a new subscriber three weeks after the cart shut. Keep the drip evergreen and let broadcasts carry the dates; the step-by-step mechanics are in the guide to Telegram drip campaigns, and the concept in the drip campaign glossary entry.
Categories let a subscriber self-select an interest during onboarding, and each category can carry its own welcome offer. For a course business the useful split is rarely topic by topic — it is by starting point. “I already have an audience” and “I am starting from zero” need different proof and different price framing, and a two-button category pick is the cheapest way to ask.
Language works on a separate axis. Each bot defines its own language codes; the bot reads the Telegram client language, the subscriber confirms it, and every later message is served in that language. That is what makes a bilingual launch realistic without running two funnels. Source tagging adds a third axis, marking where each subscriber came from, so paid and organic traffic can get different day-4 messages. More on combining the three on the audience segmentation page.
Launch week is a series of one-off broadcasts, each with a live recipient count so you size the audience before you press send rather than after. Read the count as a signal in itself: if it falls between Monday and Thursday, people blocked the bot mid-launch, which is feedback about frequency.
Scheduled broadcasts run daily, weekly or monthly with a timezone attached, which suits the recurring parts of a cycle — weekly office hours, a monthly evergreen check-in. That timezone field matters: “doors close at midnight” means nothing to a list spread across three continents, so pick the timezone the offer is denominated in and say so in the copy.
Replay reminders follow the same shape: broadcast the replay right after the session, then again as it expires, targeted to the language or category that joined, with the watch link in an inline button.
Scarcity that depends on you remembering to switch something off is not scarcity. Every campaign has an active window, and outside it the deep link simply stops delivering the offer. Set the window when you build the campaign and the cart closes on time even if launch day runs long.
Two flags do the rest of the work. One-time activation per user means a bonus or promo code is claimed once per subscriber, so a code posted publicly cannot be farmed by one account. New users only reserves an offer for people arriving fresh through that link — a real first-timer discount your existing list cannot quietly claim. Activations are recorded per subscriber, so who took the offer is a list you read afterwards, not an estimate.
The analytics that matter during a launch are not subscriber counts. Three views carry most of the signal:
Add source attribution and you compare channels on activation rate rather than raw volume — the comparison that survives a refund window.
Most course lists go quiet the day the cart closes and are cold again by the next launch. Leave the machinery running instead: switch the cart campaign off, keep the drip live for new joiners with the last step pointing at a waitlist rather than a checkout, and let attribution keep tagging arrivals.
Give buyers a deep link of their own after purchase so they carry a distinct tag, then target — or exclude — that tag in later auto-messages and broadcasts. Nothing lands worse with a new student than a last-chance enrolment message for the course they already bought. Per-bot roles — owner, admin, editor, viewer — let a contractor edit copy without ever seeing the bot token.
Do subscribers need to give an email address? No. The subscriber is identified by their Telegram account — no form, no confirmation step, nothing to bounce.
Can one account run several course bots? Yes, and each keeps its own subscribers, texts, campaigns and analytics.
How do I keep the drip out of launch week? Target auto-messages by category, tag, source or language, and keep calendar-bound messages in broadcasts where you control the send time.
What happens when someone blocks the bot? They are dropped from further sends rather than retried, and the drop shows in delivery stats and in your next broadcast's recipient count.
mySkua is invite-only while we work closely with a small group of operators. If you run course launches on Telegram and want this shape of funnel — magnet, drip, segmented offers, launch broadcasts, honest reporting — join the waitlist and tell us what you are launching.
Related: Telegram drip campaigns · Audience segmentation · Drip campaign.
Want this on your own bots?
mySkua is in a closed beta and accounts are opened in small batches. Tell us what you run and we will come back to you.
Request access