Good practice: advertising honestly

There is a version of this article that writes itself and is useless: the one that says "be authentic" and "add value". I prefer the other one, which explains what concrete mechanism penalises you for each failure. Because almost none of these practices are good because they are noble — they are good because the opposite is sanctioned by someone: by the platform, by the ad reviewer, or by the subscriber.

1 · Frequency is your most dangerous variable

In a channel there is no penalty for posting too much, because there is no algorithm to penalise you. There is something worse and more final: the subscriber mutes the channel. And muting is not leaving, so you never even find out. The subscriber count stays flat while views per post fall.

So it is worth watching the ratio of views to subscribers more than the subscriber total. If the channel grows but each post reaches fewer people, you have your answer: you are publishing beyond what your audience wants to receive.

A muted channel looks a lot like a healthy one in the subscriber statistics. That is the trap.

2 · Do not buy subscribers

It is sold, it is cheap, and it ruins the only thing you had. A channel with thousands of inert accounts inside still gets the same real views as before, but now you cannot interpret any metric: you do not know whether a drop in reach is a content problem or the background noise of the people you bought.

And there is a side effect usually overlooked: if someone evaluates your channel from outside with the estimation tools, an odd ratio of subscribers to views is immediately visible. It is the first thing anyone considering working with you looks at. Buying subscribers does not just fail to help: it marks you.

3 · Say what is on the other side

This is the most important point in the article and the one with the most formal consequences.

An ad creates an expectation. The destination meets it or it does not. When it does not, three things happen at once: the people who arrive leave immediately, you paid for that click anyway, and — if the ad was running on a platform with review — you expose yourself to rejection or worse.

It is worth stressing that this is not a recommendation of mine. Meta says so in its own documentation on ad review: the process may include the components of the ad — images, video, text and targeting — and also the landing page associated with the ad, or other destinations.

In other words: the destination is part of what gets reviewed. It is not a free space you reach once you have passed the filter, and a mismatch between what the ad promises and what the destination delivers is grounds for rejection, with possible account restrictions if repeated.

A concrete example

Suppose you advertise "Free guide to setting up your first channel". Coherent is that the destination is a channel whose pinned first posts are that guide, reachable with no further steps. Incoherent is a generic channel where the guide was mentioned three weeks ago, or where you first have to message a bot to get it. The second is not a user-experience nuance: it is exactly the mismatch the policies describe.

4 · Remember where your ad appears

Sponsored messages show up inside channels somebody chose to read. You are not interrupting dead time: you are entering a space with its own context.

That makes tone matter more than usual. An ad written with the urgency of a clearance banner jars so badly that the dissonance itself defuses it. The ones that work tend to read like a brief, sensible recommendation — which is also what the format, short text with no image, is practically begging for.

5 · If you collect data, say so and honour it

The moment a bot stores anything about the person messaging it — even just the source they came from, as in the article on measurement — you are processing personal data, and whatever data protection rules apply in your territory apply to you.

The reasonable minimum, which is also cheap to implement:

• Explain what you store and why, in the bot's first message or a visible link.
• Store only what you will genuinely use.
• Handle deletion requests, and work out how before the first one arrives.

6 · Do not invent urgency or results

Countdowns that reset themselves, limited places that never run out, and result figures nobody can verify all work once. After that, what they build is an audience that has learned not to believe you, and that is the one thing on this list that is hard to reverse.

In terms of written rules, misleading claims are also expressly covered by the review criteria of advertising platforms. So even setting ethics aside — and there is no need to set it aside — it remains bad business.

The short version

• Publish less than you feel like, and watch views rather than subscribers.
• Do not buy audience: it leaves you without reliable metrics and it shows from outside.
• Make the destination deliver exactly what the ad promises. That is a rule, not advice.
• Write like someone entering another person's house, because that is what you are doing.
• If you store data, say so and know how to delete it.
• No invented urgency, and no numbers nobody can verify.

None of these is a growth hack. They are, rather, the list of known ways to waste money, written in the negative.