Dataset at a glance
16 June to 8 September 2026
| Tracked joins | 215,531 |
|---|---|
| Observed net campaign growth | 64,489 |
| Gross-to-net delivery ratio | 3.3× |
| Channels in the dataset | 78, of which 25 campaigns completed |
| 30-day post-delivery retention | 88.3% across 19 campaigns |
Net campaign growth is observed channel growth, not an attribution of every net member to TeleBoost. Full definitions and limitations are below.
Churn during delivery is heavy, and that is our cost, not yours
- Gross churn is high. Roughly seven out of ten people who join an adult Telegram channel through a shared folder leave again during active delivery, usually within days.
- What you buy is a net target. Delivery keeps running until your channel reaches the purchased increase, counted from where it started.
- The replacement load is ours. On the largest completed campaign, 13,835 tracked joins were handled to reach a 5,000-member net target.
This is why gross joins and net growth should not be treated as the same metric, and why a supplier quoting you a gross number is quoting you the number before any of this happens.
The number that actually matters: what is left after we stop
Delivery stopping is the real test, because that is when a bought audience either holds or evaporates.
| Time after delivery ended | Campaigns eligible | Decline | Still present |
|---|---|---|---|
| 7 days | 22 | −7.1% | 92.9% |
| 30 days | 19 | −11.7% | 88.3% |
| 60 days | 12 | −12.6% | 87.4% |
These are different cohorts, not three points from one retention curve.
Why 22, 19 and 12 are not the same cohort
A campaign can only appear in the 60-day row if it finished at least 60 days ago. The 12 campaigns in that row are therefore the oldest ones, not a random sample of the 22.
So comparing 88.3% at day 30 with 87.4% at day 60 and concluding that the curve flattens would be comparing two different populations. We publish all three with their counts attached so the reader can see the shape of the evidence rather than a smoothed line.
What “net growth” does and does not mean
What it means:
- The difference between the channel count recorded at campaign start and the observed live or final count.
- The number used to decide whether a campaign’s net target has been reached.
What it does not mean:
- It does not prove that every net member came from TeleBoost.
- It does not separate organic channel growth from promoted growth.
- It is not the same thing as the number of tracked joins our delivery system generated.
In September 2026 we audited this and found four channels where observed growth exceeded the joins we had pushed, by 180 members in total across the whole dataset, about 0.3%. One of them, a campaign delivered at no cost, reached its target with zero joins from our side. We changed the completion rule as a result: a campaign is no longer marked complete on the member count alone, it also has to have received the joins that were bought.
Method, in enough detail to argue with
One row per channel per UTC day, written by a monitor that reads the live member count from inside the channel rather than from a dashboard. Joins, leaves and owner removals are counted separately. There is no backfill before 16 June 2026, so a channel that started earlier has no history before that date and is excluded from anything time-based. Averages are divided by the number of days that actually have a row, never by a fixed seven. Where a rate would need a non-positive denominator we print nothing rather than an estimate. Figures on this page were last refreshed on 8 September 2026; the live counters keep moving and are at /live/.
Where this dataset is weak
It is one network’s traffic, aimed at one vertical, and it measures one hop. Members are people already inside adult Telegram channels we operate; the mechanism that moves them into a customer’s channel is Telegram’s shareable chat folders. That mechanism describes how a member arrives, not how the audience was built in the first place, and this dataset says nothing about the latter. It says nothing about members sourced by other methods and nothing about whether those members buy. Language is not nationality: the Telegram API exposes an app language, not a country, and we do not infer one from the other. And the post-delivery numbers rest on 12 to 22 campaigns depending on the window, which is enough to be worth publishing and not enough to be a law.
Check it yourself
Every campaign referenced here has its own public page with a day-by-day table. /live/TB-E4FA is the 5,000-member campaign that completed in 28 days; /live/TB-FBB8 is a 1,000-member campaign whose channel can be read directly on Telegram’s public preview. No login, no form, no request. If a number on this page and a number on those pages disagree, the campaign page is the one to believe.
No password, ever
The bot goes into your channel as admin in one tap, permissions already set, and you stay the owner. Never a password, never a login code: anyone asking for those wants the channel, not your growth.
Real 18+ users
Members come from adult Telegram channels we operate ourselves, so they already follow this kind of content.
Net target guarantee
We keep delivering until your channel reaches the full increase you bought, counted from where it started. Departures during delivery do not reduce your order.