# How to warm up a TikTok account on autopilot (https://clout.wtf/guide/warmup-guide/) Warming up a TikTok account means spending time in the app the way a regular viewer would — scrolling the feed, watching videos, liking and following — before the account starts posting. Clout Uploader automates this on a physical iPhone: you pick a warmup template, choose feed or keyword-search activity, and the account builds its own watch history with no one at the screen. This guide covers both warmup types, every template parameter, and the presets we recommend. ## What warm-up is Warm-up is a mode where the account doesn't publish anything — it just spends time in TikTok like a regular user: * scrolls the For You feed * watches videos for a randomized duration * occasionally likes, follows, or saves to favorites * searches for videos by keywords * watches results from search **It does not upload or publish anything.** Warm-up exists to build up "life" on the account before its first publication and to teach TikTok's algorithm your niche. ### When to use it | Scenario | Warm-up needed? | | ------------------------------------------------ | --------------------- | | Freshly logged-in account before its first post | **Yes**, 1–2 cycles | | Account after a long pause (a week+) | **Yes**, 1 cycle | | Account that's been actively publishing recently | Optional | | You want to teach the feed your interests | Yes — keyword warm-up | > **Note:** Warm-up can run as a **standalone thread** before the main Upload thread, or be plugged into a Seeding cycle (after login) via `seeding_warm_*` settings. ## Two warm-up types The template has a **Feed** / **Search** toggle. One thread run uses **one of the two** (the choice is locked in the template). ### Feed — the For You feed The thread opens TikTok and scrolls the feed: 1. Watches the current video for a random duration 2. With probability X% likes the video 3. With probability Y% follows the author 4. With probability Z% saves to favorites 5. Swipes up to the next video 6. Repeats N times ### Search — keyword-driven The thread opens TikTok and runs a search: 1. Opens the search screen 2. Types **one random** keyword from the list (one is picked per run) 3. Switches to the "Videos" tab 4. Opens the first video in the results 5. Watches videos in fullscreen mode for a random duration 6. With probability X / Y / Z% likes / follows / saves to favorites 7. Swipes to the next video N times Search is more valuable when you want to nudge the feed toward a niche: cooking, fitness, a specific brand, etc. — TikTok's algorithm picks up the interests. ## Template setup In the UI: **Templates → create template → Run Mode = Warmup**. ### "Warmup Settings" card #### Feed / Search toggle You pick one type. Parameters of the other type are saved in the template but not used. #### Feed parameters | Parameter | What it controls | Default | Limits | | ----------------------------------- | ----------------------------------------------- | ------- | ------- | | **Videos in feed** (min / max) | How many videos to scroll (random within range) | 5 – 15 | 0 – 100 | | **Time per video** (min / max, sec) | Seconds per video | 2 – 10 | 3 – 20 | #### Search parameters | Parameter | What it controls | Default | Limits | | ----------------------------------- | ----------------------------------- | ------- | ------- | | **Keywords** | List of keywords (one per line) | empty | — | | **Videos in results** (min / max) | How many videos to watch in results | 3 – 8 | 0 – 100 | | **Time per video** (min / max, sec) | Seconds per video | 5 – 16 | 3 – 20 | > **Note:** **Keyword format:** one keyword per line. **One random keyword** is picked per thread run — so write similar / niche keywords. > Example: > > ``` > easy pasta recipe > 5 minute meal > homemade ramen > chef secrets > ``` ### "Feed Rates" / "Search Rates" cards Interaction probabilities are independent "coin flips" on each video: | Parameter | What it controls | Default | | ------------ | ----------------------------------------- | ------- | | **Like** | Chance of liking each video (0 – 100%) | 10% | | **Follow** | Chance of following the author (0 – 100%) | 0% | | **Favorite** | Chance of saving to favorites (0 – 100%) | 0% | > **Note:** **Heads up:** these are **independent chances** per video, not "one like per session." If you set Like = 50% and scroll 10 videos, statistically you'll get 5 likes. Don't push above 15–25% for Like or above 5% for Follow / Favorite — high values are too obviously different from a regular user. ### "Pause Between Accounts" card Pause between accounts when working with a multi-account list: | Value | When to pick | | ------ | ------------------------------------ | | 5 sec | Test run | | 10 sec | Normal mode, 2–3 accounts per device | | 30 sec | Large account list, cautious pace | | 60 sec | Maximum manual-switch imitation | ### "Debug" card * **Save screenshots** — screenshots at each step for diagnostics. On by default (matters for support). Turn off only when you're sure warm-up runs cleanly. ## Fields not used in warm-up Some template fields are inherited from Upload and ignored in warm-up. You can leave them alone: * Video / photo folder * Content type * Descriptions and everything related * Music * Move/Delete used * Publications per account, after upload wait ## What you need to launch a thread Beyond the template, the **Threads → Add Configuration** tab takes: | Field | What to pick | | ----------------------------------- | ------------------------------------------------- | | **Configuration Name** | Any name, e.g. "Warmup Phone 2" | | **Devices** | One or more devices to run on | | **Upload Template** | The template with Run Mode = Warmup | | **Account List** | List of accounts (login must already be done) | | **Proxy** | Optional — Stash / Happ / System / Manual | | **Cyclic Upload** | Turn on for an endless cycle through the accounts | | **Pause between accounts / cycles** | Your call | | **Schedule** | Optional — scheduled run | > **Note:** **Accounts for warm-up must already be logged into TikTok on the device** (or in Keychain). Warm-up does **not log in** on its own — for cold-start login use Seeding. ## What happens during a session ``` 1. The thread opens TikTok on the device 2. If Feed is selected: - Loop over N videos: - Watch X seconds - With configured chances: like / follow / save - Move to the next video 3. If Search is selected: - Open the search screen - Type one random keyword - Switch to the "Videos" tab - Open the first video - Loop over N videos (like the feed, with interactions) 4. Return to the TikTok home screen 5. If multiple accounts → pause → next account ``` If **Cyclic Upload** is on, after walking through all accounts the thread sleeps for "Pause between cycles" minutes and starts over. ## Common errors | Symptom | Cause | Solution | | --------------------------------------------------------------- | ------------------------------------------------------- | ------------------------------------------------------------------ | | Thread crashes immediately with "No keywords for warmup search" | Search mode without keywords | Fill in Keywords or switch to Feed | | Warm-up keeps using the same keyword | By design — one is picked per run | Restart the thread / enable Cyclic to rotate per run | | No likes at all | Like chance is 0% or it's a livestream (sidebar hidden) | Push Like to at least 5–10%; make sure the feed has regular videos | | Videos fly by too quickly | Min time per video too small | Push the minimum to 3–5 seconds | | Thread freezes for 30+ seconds between videos | Hit a livestream — the thread waits and swipes onward | Standard behavior, nothing to do | | Account engagement reach drops after a series of warm-ups | Too aggressive settings (Like / Follow > 30%) | Lower the interaction chances | ## Recommended presets ### "Safe warm-up for a new account" ``` Type: Feed Videos: 8 – 15 Time: 4 – 12 sec Like: 8% Follow: 1% Favorite: 1% ``` Mimics a first encounter with TikTok: a user lazily scrolls and rarely likes. ### "Niche feed shaping" ``` Type: Search Keywords: fitness motivation gym tips home workout beginner workout Videos in search: 5 – 10 Time: 6 – 15 sec Like: 15% Follow: 3% ``` The algorithm picks up the interests — the next feed is on-topic. ### "Active user" ``` Type: Feed Videos: 15 – 25 Time: 3 – 9 sec Like: 18% Follow: 5% Favorite: 2% ``` Mimics an active user who watches a lot and likes noticeably. ## "Warm-up configured correctly" checklist * [ ] Template created with Run Mode = **Warmup** * [ ] Type picked: Feed **or** Search * [ ] If Search — keywords filled in (one per line) * [ ] Video count and watch time within a realistic range (not 100 videos at 30 sec each) * [ ] Interaction chances ≤ 25% (Like) and ≤ 5% (Follow / Favorite) * [ ] Pause between accounts ≥ 10 sec for multiple accounts * [ ] Accounts already logged into TikTok on the device * [ ] Account List and Device set at thread launch * [ ] (Optional) proxy attached if you need to mask IP ## FAQ ### How long does it take to warm up a new TikTok account? For a freshly logged-in account, run 1–2 warmup cycles before the first post. An account coming back after a week or more of silence usually needs one cycle. Accounts that have been posting actively don't need dedicated warmup runs. ### Do I have to warm up before every post? No. Warmup is for fresh accounts, accounts returning from a pause, and for teaching the algorithm your niche via keyword search. An account in a steady posting rhythm doesn't need standalone warmup cycles. ### Does warming up guarantee views or account safety? No system can guarantee outcomes. Warmup does one specific thing: it builds normal viewing activity on the account and teaches TikTok's algorithm which niche the account belongs to. Treat it as building watch history, not as protection. ### Can I warm up several accounts at the same time? Yes. Each device runs its own thread, and accounts on a device take turns within TikTok's native multi-account support (up to 8 accounts per device). A fleet of iPhones warms up in parallel. ## Related guides * [Folder structure](https://clout.wtf/guide/folder-structure-guide/) — warm-up doesn't need media folders, but it's useful to see how warm-up differs from Upload and Seeding. * [Upload mode](https://clout.wtf/guide/upload-guide/) — the main publishing mode, often launched right after warm-up. ## Further reading on the blog * [TikTok Account Warmup: The Complete Step-by-Step Guide (2026)](https://clout.wtf/blog/tiktok-account-warmup-guide/) * [TikTok Shadowban: What It Is, How to Detect It, and How to Fix It](https://clout.wtf/blog/tiktok-shadowban-guide/)