# How Live is pushed: length, frequency, reach

You have exactly two dials you turn deliberately: how long a stream runs and how often it runs. How widely each one is shown is decided during the stream itself, by whether viewers stay, type and react.

**What you leave with:** Your own launch schedule, and a clear idea of what you are optimising when you change length or frequency.

## 01. How the platform widens a showing

Every launch runs the same widening cycle: a small showing, a check on how people react, then a wider showing or none. An empty stream is that cycle stopping after the first showing.

1. **A showing to a very small group** — The first viewers are a trial, not an audience.
2. **Watching the reaction** — Whether they come in, stay, type and react.
3. **Widening, or stopping** — A reaction means a wider showing. No reaction means the showing does not grow, however long you keep going.
4. **Repeat on new people** — And round again for as long as the stream runs: every widening is a fresh trial.

Which has a direct consequence for length: while the stream is running the platform has a reason to try again. A stream you closed is not tried again.

## 02. How long to hold a stream

A short stream is not “less traffic” — it is fewer attempts to show you to anyone.

| Length | What it buys |
| --- | --- |
| 5–10 minutes | Almost nothing: the widening cycle does not get a second run |
| 30–40 minutes | The least after which a stream can be judged |
| 30–60 minutes | The baseline format: enough for several trials |
| 1–2 hours | A strong streamer's working format: the room turns over and the stream is tried on different groups |

**Watch out:** Closing at five or ten minutes is the most common avoidable loss. There is one exception: if the room is still empty at the thirty-minute mark, close it — staying longer will not help. Then comes the loop from lesson 03: a second launch at a different time, then a different account.

## 03. How often to go live

Frequency depends on the hours you actually have. Below are a floor, the sweet spot and the rhythm of the strongest cases; the floor is the one to hold even in a bad week.

1. **One stream every two days** (Minimum) — Below that a setup gathers neither an audience nor the data to decide anything with.
2. **Two streams a day** (The sweet spot) — Afternoon and evening are different audiences, and the difference shows from the first launches.
3. **3–7 launches daily** (Strong cases) — No days off. That is the rhythm most often behind steady growth in reach and traffic — and the one that requires the team from lesson 10.

Three to seven launches a day is not where anybody starts. It is here to show the ceiling: growth in this source runs into hours on air rather than into knowing more tricks, and that many hours take a team of streamers. Where to find them, and whether they share one channel or run one each, is lesson 10.

The number of launches is a poor KPI: ten short streams and four long ones cannot be compared. Streams, weeks and streamers are compared by FTDs per hour on air — lesson 11.

## 04. Why random accounts take off

If reach is weak, reworking the profile will barely lift it: the old niche, old videos and the look of the profile hardly affect who the stream is shown to. The four things that do affect it all happen on air.

**Affects the showing**

- How long viewers stay
- Activity in the chat
- Reactions from the room
- No pauses and no stillness

**Barely matters**

- The account's old niche
- The old videos
- How the profile looks
- The handle and the avatar

The platform picks an audience for what is happening live right now rather than for the account's history. Which is why streams regularly take off from completely random accounts — and equally why a beautifully presented account does not save a quiet stream.

## 05. Diagnosing your reach

Six symptoms visible from a single stream. Each with its likely cause, what to change, and what to leave alone while you do.

**No reach at all, nobody in the room**

The platform never started pushing the stream. Usually that is the account: it is not being shown at all.

- What to change: The loop from lesson 03: a second launch, a different hour, then a different account.

- What to leave alone: The length. Sitting in an empty room for longer does not help.

**People come in and leave immediately**

Retention. The first seconds give a viewer nothing to hold on to.

- What to change: The opening: movement on screen and a voice from the first second.

- What to leave alone: How often you launch. More streams with the same opening give the same result.

**There are viewers, but the chat is silent**

Nothing to type about. The stream is running as a monologue.

- What to change: Questions to the room, letting them pick the pair, asking outright for an opinion.

- What to leave alone: The subject you are reading. That is not why they are quiet.

**The room grows for a few minutes and then stops**

The first trial worked and the second did not: the middle of the stream went flat.

- What to change: The middle: change pair, change scale, a new question to the room every few minutes.

- What to leave alone: The opening. That is the part that worked.

**Streams are short and reach has not grown in weeks**

The widening cycle does not have the time it needs.

- What to change: Length: 30–40 minutes as a floor, then an hour or two.

- What to leave alone: Your delivery. There is nothing to judge it on yet.

**The room is busy, but nobody goes to Telegram**

Reach is not the problem. The move is, or the streamer.

- What to change: How you word the move and the channel's first screen, and in the streamer — speech, personality, language, trust, the CTA and knowing the product.

- What to leave alone: Everything to do with reach: the algorithm has done its part.

## 06. Mini-task: your schedule

- [ ] Counted the hours a week I can genuinely be on air
- [ ] Laid them out as launches of 40–90 minutes
- [ ] Put at least one stream in the afternoon and one in the evening
- [ ] Wrote down the floor I do not drop below even in a bad week
- [ ] Marked, in the diagnostics, the symptom I see most often in my own streams

The second line matters more than the first: ten hours a week as four long streams and as twenty short ones are two completely different results.

---

[Back to the programme](https://gamechange.partners/en/academy/tiktok-live/)
