Growth has flattened, views swing for no obvious reason, so you change everything in the same week: the topic, the timing, the hashtags, the edit. A month later you have changed five variables and you cannot tell which one did anything.
The direct answer: what stalls most accounts is not one large mistake, it is five or six small decisions that repeat every week. Below are twenty of them in four categories, each with one reason and one fix.
Before you start, the thing that separates this list from the others. Some of these items break a rule TikTok publishes in writing. Others are practical judgement built on observation. Which one you are looking at is marked in the Status column of every table, because confusing the two is where almost every algorithm myth comes from.
How to use the list: open your five worst-performing videos, mark which items are present, then fix one item per publishing cycle and measure. Changing everything at once hides what worked.
| Category | Items | The question it answers | Covered in depth by |
|---|---|---|---|
| The opening | 1-5 | Why do people leave before the video starts? | TikTok hooks |
| Strategy | 6-10 | Why does nothing compound even though I post? | TikTok content strategy |
| Production and publishing | 11-15 | What am I breaking without noticing? | TikTok video production |
| Account and analytics | 16-20 | Why do I keep repeating the same mistake? | TikTok analytics |
Mistakes 1-5: The Opening
Anyone who leaves in the first second never reaches anything else you made. TikTok describes a viewer watching a longer video from beginning to end as a strong indicator of interest, which makes the opening the gate to everything after it.
| # | Mistake | Status | Why it hurts | The fix |
|---|---|---|---|---|
| 1 | Opening with "Hey guys, welcome back to my channel" | Practical judgement | Spends your two most valuable seconds on words that carry no information | Open on the line that carries value. Introductions belong in your bio and your comments |
| 2 | A generic opening: "I'm going to show you something important" | Practical judgement | Addresses nobody in particular, so it gives nobody a reason to stay | Name the subject and the audience in the same sentence |
| 3 | A music intro, an animated logo, or a blank frame | Practical judgement | Two seconds with no reason to stay, which is a long time at scrolling speed | Start at frame zero with an image, a movement or a word |
| 4 | An opening that promises what the video never delivers | Practical judgement | No penalty is documented for this, but the effect shows in your own data: a good arrival then a fast exit. Repeat it and your audience stops believing your openings | Size the promise to what you actually deliver |
| 5 | On-screen text that cannot be read at a glance | Practical judgement | Long, thin or edge-hugging text disappears under the caption and the interaction buttons | One short sentence, heavy weight, clear of the edges and the button zone |
The formulas themselves, seven of them with examples, are in TikTok hooks.
Mistakes 6-10: Strategy
This category never shows up in a single video. It shows up two months later, when you realise the effort did not accumulate in any direction.
| # | Mistake | Status | Why it hurts | The fix |
|---|---|---|---|---|
| 6 | Switching topic every two weeks | Practical judgement | People who followed you for one subject get another, and you never collect comparable data on either | If the change is necessary, phase it in while the original line is still running |
| 7 | Posting with no rhythm | Practical judgement | The value of consistency is that it gives you a fair comparison between videos, not that it "pleases the algorithm" | Pick a cadence you can hold. The timing question has its own page: the best time to post on TikTok |
| 8 | Ignoring comments | Practical judgement | TikTok lists user interactions among its recommendation factors but publishes no weighting. The certain loss is simpler: your comment section is the cheapest audience research you own | Read comments as an idea source. Details in TikTok engagement strategy |
| 9 | Relying on trends alone | Practical judgement | A trend brings a passing visit and gives nobody a reason to follow an account with no identity | Use the trend as a format that carries your subject, not as a substitute for it |
| 10 | Padding inside the video | Practical judgement | Every second without a reason is another chance to leave | Watch your own video at 1.5x and cut every moment you can cut |
Mistakes 11-15: Production and Publishing
This is the category the myths cluster around, which is why most items here are documented rules, taken from TikTok's Community Guidelines and support pages rather than from observation.
| # | Mistake | Status | Why it hurts | The fix |
|---|---|---|---|---|
| 11 | Posting a clip carrying someone else's watermark or logo, or re-posting a clip with no real editing | Documented rule | TikTok lists reused or unoriginal content posted without creative edits, including clips that show someone else's watermark or logo, among the content that is not eligible for the For You feed | Export without a watermark, and if you build on someone else's material, add your own work to it. App comparison in best TikTok editing apps |
| 12 | Exporting with settings you never checked | Practical judgement | A file that ignores the platform's own guidance gets re-compressed, and text edges and fine detail pay for it | Open TikTok's recommended video specs page and match it. They are recommendations, not minimums and not a ranking factor. More in TikTok video production |
| 13 | Stuffing hashtags that do not describe the video | Practical judgement | A hashtag is a description. A tag that does not describe the video adds nothing, and tagging for an audience unrelated to your content edges toward deceptive behaviour | Describe the video accurately and stop. The full method is in TikTok hashtags |
| 14 | Deleting a weak video hoping for a reset | Practical judgement | No penalty is documented for deleting, and no reward either. What you do lose is the data that would have told you why it underperformed | Keep it and read its retention curve. The full question is in does deleting a TikTok video hurt |
| 15 | Publishing commercial content without disclosing it | Documented rule | TikTok lists commercial content that is not disclosed with its content disclosure setting among the content that is not eligible for the For You feed | Turn the disclosure setting on for any paid collaboration. Note too that business accounts use the Commercial Music Library rather than the general one: the difference is in business vs personal account |
Mistakes 16-20: Account and Analytics
The mistakes that make everything above repeat, because nothing ever gets diagnosed. #MistakeStatusWhy it hurtsThe fix 16Measuring success by views alonePractical judgementViews describe distribution, not impact. A video that reaches a lot of people and leaves none of them behind builds nothingRead alongside views: completion, saves, shares, and follows after watching 17Never opening the retention curvePractical judgementWithout it you do not know where people leave, so you fix the wrong sectionOpen the curve for each video and find the steepest drop. Reading it and the benchmark question are in what a good TikTok retention rate is 18Judging from one video, or quitting before you have a samplePractical judgementA single video is shaped by variables you do not control, so judging from it is judging noiseCompare within one content type and across several videos. There is no official threshold, and any "the algorithm needs N videos" is not published by the platform 19Expecting a purchased number to do the job of contentDocumented ruleTikTok's Integrity and Authenticity policy prohibits fake engagement, states that it removes fake followers, likes and other inflated metrics when it detects them, and that it may restrict the accounts involved. That is the whole of what is published: not "a permanent algorithm penalty", and not "nothing happens"Separate the two decisions. A content problem is solved with content. If the goal is a visible number for a campaign or an initial presence, that is a separate commercial decision. 20Rewriting everything before checking account statusDocumented ruleTikTok states that if content is made ineligible for the For You feed, that information appears in the analytics tool, and that few views alone do not necessarily mean a rule was broken. The app also has an account status screen showing violations, restrictions and the appeal routeCheck account status and the analytics notices first. Then TikTok shadowban for diagnosis and why TikTok views drop for the full order of checks Three Circulating "Rules" With Nothing Behind Them These three appear in almost every mistakes list, and all three were in the previous version of this page. None has an official source:
- "A watermark freezes the video at 200 views." No such number appears on any official page. What does exist is broader and clearer: reused content posted without creative edits, including clips showing someone else's watermark or logo, is listed as not eligible for the For You feed. The rule is about originality, not about a view counter.
- "Using TikTok's native tools earns a distribution bonus." TikTok documents its in-app editing tools and what they do, and states nowhere that using them improves distribution. They are convenient and fast, which is reason enough without inventing another one.
- "Deleting videos sends an instability signal." TikTok documents how to delete a post and what happens to it, and documents no penalty. The "set it to private instead" workaround was built on this same myth.
The general test that covers all three: if a claim describes an internal mechanism with a number, a percentage or a duration and no official page says it, it is a guess no matter how often it is repeated.
Frequently Asked Questions
What is the most common TikTok mistake?
Spending the first seconds on a greeting or a preamble instead of the subject. That is practical judgement rather than a published rule, but its effect is visible in almost any account's data: the steepest drop in the curve sits at the opening.
Does deleting underperforming videos hurt your TikTok account?
TikTok documents no penalty for deleting. Its support pages explain how to delete a post and what happens to it, and mention no effect on distribution. The real loss is the video's data, which is what would have told you why it underperformed. The full question is in does deleting a TikTok video hurt.
Does buying followers or views help growth?
The accurate answer is neither "nothing happens" nor "a permanent algorithm penalty", and both get repeated constantly. What TikTok publishes in its Integrity and Authenticity policy is this: fake engagement is against the guidelines, the platform states that it removes fake followers, likes and other inflated metrics when it detects them, and that it may restrict the accounts involved. That is platform policy as published, it can change at the platform's discretion, and reviewing it and the terms of service before using any service is the user's responsibility.
What we do not claim in either direction: that a purchased number raises or lowers reach, or changes recommendation or how long people watch. Services and content solve two different problems, and the decision is yours after reading the terms.
How many videos should I post before judging the result?
There is no official number, and every "the algorithm needs N videos" or "wait N days" is unpublished. The practical standard is to compare within one content type and across several videos rather than one, because a single video is shaped by variables you do not control.
What is the most important metric in TikTok analytics?
The retention curve, because it is the only one that tells you where the problem is rather than just that one exists. The steepest drop identifies the section to fix. As for what counts as a good number, TikTok publishes no benchmark for any metric, and the only useful comparison is against your own account's average.
Practical Summary
- What stalls most accounts is repeated small decisions, not one large mistake.
- Always separate a rule TikTok publishes from practical judgement. Most of what circulates about "algorithm penalties" is the second kind, and most of it has no source.
- Only four items on this list are documented rules: reused content posted without editing, undisclosed commercial content, fake engagement, and checking account status. The rest is useful judgement, not platform policy.
- Bottom line: open your five worst videos, mark which items are present, and fix one per publishing cycle.
- Your next step: start with one item from category three, because that is the category with actual rules you can break without noticing.
- The content types that structurally avoid half of these are in TikTok content ideas, and the full path from idea to publish is in TikTok content creation. If the deeper question is how distribution is decided at all, that is how the TikTok algorithm works.
Every other TikTok path is collected in the TikTok Resource Hub.