Multiple creators, a second bot, multiple languages

What gets copied, what does not

Going from one creator to several is not a multiplication. It is inventory work: separating what is a method from what is a special case.

Copied as is. The script structure, that is, the six-step sequence from module 5. The daily routine from the first lesson of this module. The definition of the five indicators. The procedures: setup, account loss, access register. The sheet templates.

Not copied. The persona, obviously. The prices, which depend on the catalog and the actual audience. The media. And above all, the test results. This is the most frequent mistake at duplication time: a teaser that won on one creator is applied everywhere, without re-checking, and becomes an internal belief that nobody questions anymore.

The test register from module 5 exists for exactly this: it reminds you what a result was obtained on. A result without its scope is not transferable.

The second creator is the most expensive

Counterintuitive, but constant: going from one to two is the hardest step, not going from four to five.

Three reasons.

Everything that was implicit becomes explicit. With a single creator, the method is in your head. With two, it has to be written, otherwise you will apply to one what you learned on the other without noticing.

The tools change in nature. A spreadsheet with two tabs becomes a spreadsheet with two times five tabs. Accesses multiply. This is the moment when a multi-creator management tool starts to justify itself, and when the CRM category from the previous lesson, and from module 8, stops being a luxury.

Your time fragments. The hidden cost is not the extra work, it is the context switching. Two personas, two catalogs, two price ladders, two audiences.

The practical consequence is that you must write the method before the second creator, not after. That is the point of this lesson’s exercise. From the third onward, duplication becomes mechanical, and the marginal cost drops.

When to create a second Telegram bot

The question comes up as soon as the second creator is on Telegram, and the default answer is yes.

A Telegram bot carries a token, an identity and a history. In practice, three arguments favor one bot per creator.

Risk isolation. A problem on one bot does not touch the others. It is the same reasoning as the replacement account queue from module 4.

Identity consistency. A bot is an object the fan can see. The same bot serving several personas is an inconsistency, and inconsistencies get paid for in this trade.

Readable measurement. One bot per creator gives separate numbers with no accounting effort.

The counter-argument is the administrative load: one more bot is one more token, one more configuration, one more object to watch. That is real, and it is precisely what the setup checklist from the exercise must absorb.

A reminder of the official milestones, useful at configuration time: business accounts and bot connection date from version 7.2 of the API, published on March 31, 2024; paid media from version 7.6, published on July 1, 2024; subscriptions and gifts from version 8.0, published on November 17, 2024.

Opening a language

This is the worst-made decision in the industry, because it is almost always made backwards: you translate a script, activate it, and conclude that the market does not work.

The method that gives a usable answer fits in three steps, and it follows the funnel stages from module 4.

Step 1: validate acquisition. Open an acquisition account dedicated to that language and publish in that language for several weeks. Measure only the first three stages: seen, followed, clicked. If acquisition does not take, the language is closed, and you have written no script.

Step 2: validate the conversation, by hand. If acquisition takes, handle the first conversations yourself or with someone who speaks the language, with no formalized script. You are looking for three things: what time these people write, what topics they bring up on their own, at which tier they stop. Those are the three cutting criteria from module 6.

Step 3: write the version. Only now. And only if at least two of the three criteria differ from your main audience. Otherwise, a translation is enough, and a translation costs ten times less than a version.

This sequencing avoids the most expensive mistake: producing a full version for a market whose acquisition was never taking off.

The tipping point

There comes a moment when your job changes without warning. You stop writing the conversations and you start deciding which conversations should exist. You stop producing and you start checking.

That shift is a good one, but it has a downside you need to see coming. The more automated and duplicated the operation is, the more the gap between what you believe is happening and what is actually happening can grow without anything flagging it.

That is the subject of the last lesson of this module, When to add humans back.