If you run a consumer company and want to know how to go viral on TikTok, start with an awkward fact. Nobody can make one video go viral on purpose. What a company can do is build a format, test it against others, and repeat the one that holds attention. Do that for long enough and big videos arrive as a by-product.
This guide is written for a founder. You have an app, a DTC brand, a game or a marketplace, and you want TikTok to bring customers without buying ads. You are not trying to be famous. A creator wants attention. You want the right people to remember your name.
I run short-form video for consumer companies. Below is how TikTok hands out reach, what the first seconds must do, how to test formats, what to measure in the first two weeks, and what a view is worth. The wider case for formats over hits, on any platform, is a separate piece.
Read why formats come first and hits second →
How the TikTok algorithm works, in TikTok's own words
Most of what you read about the TikTok algorithm is guesswork with numbers attached. TikTok has published a short description of its own, and it is more useful than the guesses.
TikTok says the For You feed draws on three groups of factors. The first is what a person does: the videos they like or share, the accounts they follow, the comments they post. The second is information about the video, such as captions, sounds and hashtags. The third is device and account settings, such as language, country and device type. TikTok says that third group counts for less, because people do not choose those settings as preferences.
The factors are not equal. TikTok gives one example of a strong indicator of interest: whether a person finishes watching a longer video from beginning to end. Its example of a weak one is the viewer and the creator being in the same country.
TikTok's Transparency Center describes the same thing from the other side. The system predicts how likely a person is to do certain things with a video. The list includes finishing it or skipping it, spending a certain amount of time viewing it, and following its author.
Then comes the part that matters most to a new company. TikTok says neither follower count nor previous high-performing videos are direct factors in the recommendation system. It adds one honest caveat. An account with more followers is still likely to get more views, because it has a larger base of followers to begin with.
Three things follow for a founder. A new account can reach strangers on the TikTok For You page from its first upload. An old account with a past hit gets no free pass on the next video. And the response that counts most is a quiet one: people staying to the end.
Why one viral TikTok is luck and a format is a plan
A single video that takes off has many possible causes. The subject, the opening line, the sound, the hour, the mood of the feed that day. With one example you cannot tell them apart, so you cannot do it again on purpose.
A format fixes that. A format is a video you can make again next week with a new subject. On TikTok the fixed parts are the opening move, the length, the way it is shot, and the line of text on screen. The subject is the only part that changes. Because everything else is held still, a run of videos in one format can be compared, and a single video cannot.
A company format has one extra rule. The product, or the problem it solves, has to be inside the format itself. A creator can go viral with a dance and be happy. If your account does, you get viewers who came for the dance, and the next video about what you sell will bore them.
The hook: what the first seconds of a TikTok must do
A person who swipes away never sees the rest of the video. So the opening carries more weight than everything after it.
TikTok's published advice on this is written for advertisers, and I am flagging that plainly. In its creative guidance for performance ads, TikTok tells advertisers to put the hook in the first 6 seconds. It also says to introduce what the content offers in the first 3 seconds. That is advice for paid videos. I apply it to organic videos too, because they sit in the same feed, in front of the same thumbs.
In practice a TikTok opening has three parts, and all three land at once:
- The first frame. It should look like the middle of something. The product in a hand, the app screen with the result on it, the game at its tensest second. No logo and no greeting.
- The line of text on screen. One short sentence that says who this is for or what they will get. It has to work alone, for anyone watching with the sound off.
- The first spoken words. They should say the same thing as the text in a different way, or start the story mid-sentence.
For a consumer product the simplest hook is the product doing its job. A founder explaining why the company exists is asking a stranger for patience. The stranger has a thumb and a feed full of other options.
The hook is also the cheapest thing to test. Film one video and record three different openings for it. Post them on different days and compare how many people stayed. Most teams throw away a whole video when only the first line was wrong.
Retention and rewatch: how to get views on TikTok that count
Finishing a video is the one viewer action TikTok names as a strong indicator. So the plain answer to how to get views on TikTok is to make videos people finish. Those are the videos the system has most reason to show again.
Four habits help people reach the end:
- One idea per video. If a second idea shows up while you edit, it is next week's video.
- Cut the length to fit the idea. A video should be as long as its idea needs and no longer. Most that fail in the middle were too long for what they had to say.
- Give a reason to stay that is not paid off until the end. Show the result last, or open with a question and answer it in the final seconds.
- End on the payoff and stop. A long goodbye teaches people to leave early.
Rewatches need care. TikTok has not published how a second watch is counted against a first, so I will not pretend to know. A video someone watches twice is a video they finished, and that part TikTok has spelled out. Do not build tricks for it. Make the video dense enough that a second look is worth it: a fast before and after, or a result that is hard to believe the first time.
You can read all of this in the app. According to Buffer's guide to TikTok analytics, TikTok Studio shows the average watch time for each video. It also shows the share of viewers who watched the full video, and whether they came from the For You feed or from your profile. Those three lines tell you more than the view count does.
Test formats, not single videos
Here is the method I would give a founder starting from zero. It needs a phone, a free editing app and a notebook.
Write down three formats before you film
Describe each one in a single sentence that still makes sense when you swap the subject. For example: open on the finished result, then show the steps that got there. If you cannot write the sentence, it is a single idea and it will not repeat.
Output: Three one-sentence format descriptions
Make several videos in each format
One video per format tells you nothing, because you are back to judging single videos. Make at least four in each, with a new subject every time and everything else held still.
Output: A run of videos per format
Change one thing at a time
If you change the opening, the length and the subject in the same week, the result cannot tell you which one mattered. Pick one. The opening line is the best place to start.
Output: A clean comparison
Judge the run and ignore the best video
Look at the typical result across each format's videos. A format where every video does decent work beats a format with one giant video and nothing else. The giant one may have been luck.
Output: A ranking of the three formats
Keep the winner, drop the weakest, add a new one
Make more of the format that held people. Retire the worst one. Bring in a new challenger so the test never stops. When a video goes far, it will sit inside a format you already know how to make again.
Output: Next month's plan
That last point is the whole reason to work this way. A viral video inside a format tells you what to post tomorrow. A viral video on its own leaves you guessing, and guessing is where most company accounts spend their second month.
See how we build short-form formats →
How often to post on TikTok, without the myths
How often to post on TikTok is the question with the most confident answers and the least evidence. I read TikTok's own description of its recommendation system and its Transparency Center page for this guide. Neither one names a number of posts per day as a factor. Neither one names a best hour.
There is a posting-time chart for every day of the week. They do not agree with each other, which tells you most of what you need to know about them.
What TikTok does say points the other way from most cadence advice. Its For You feed generally will not show a viewer two videos in a row by the same creator. It also says it does not recommend duplicated content. So posting ten times a day does not put you in front of the same person ten times. And uploading the same clip again is not a second chance.
The useful question is a different one. How many attempts do you need to compare formats, and what pace can your team hold for months? A pace you keep for a year beats a sprint you drop in week three. Every extra video is one more attempt, as long as it is a real attempt in a format you are testing.
What to measure in the first two weeks
In the first two weeks you are not measuring the business. You are measuring formats. Signups from a new account will be small and noisy this early. By the end you should be able to say one sentence: this format holds people and that one does not.
| Easy to count | Worth counting in the first two weeks |
|---|---|
| Views on your best video | The typical views across each format's run |
| Likes | Average watch time and the share who watched the full video |
| Follower count | Where the views came from: the For You feed or your profile |
| Comments in total | Comments that ask what the product is or where to get it |
Write the numbers down yourself, once a week, in one sheet with a row per video and a column for its format. Buffer's guide notes that some per-post numbers in TikTok are only available for the past seven days. If you wait a month to look, part of the record is gone.
Read the source of views with care. Views from the For You feed are strangers, which is the point. Views from your profile are people who saw one video and went looking for more. When that second number rises, a format is doing its job.
What a TikTok view is worth
A view records that a video was shown. Between that and a customer there are several steps, and most people drop out at each one. They watch. A few tap through to the profile. A few of those tap the link. A few of those sign up, install or buy.
Nobody can tell you your rate in advance, and I will not invent one. You can measure your own with three cheap habits:
- Point the link in your bio at one place only: the store listing, the product page, or the waitlist if you have not launched.
- Ask one question at signup or checkout: where did you hear about us?
- Each week, set new signups that named TikTok next to that week's views. The ratio is your own number, and it will change by format.
Two receipts from my own record show how far apart views and customers can sit. The first is a dating app I built and own. Its TikTok account grew to 12,000 followers and its Instagram account to 18,000, with $0 paid. Behind them sat a waitlist of 10,000 people, collected in five months after three months of preparation. On launch, roughly 4,000 euros of produced and paid UGC converted zero users.
That receipt counts followers and signups. It does not record views, so I cannot give you a views to signups rate from it. What it shows is two outcomes on one product. The accounts people chose to follow came with a waitlist. The content I paid for brought nobody.
Read that with care. It is one gaming marketplace, and the receipt does not say which videos moved the search number. It promises your company nothing. The part worth copying is where we looked.
So add one more line to your sheet: searches for your name. Google Search Console shows which queries bring people to your site, with impressions and clicks. Filter it to queries that contain your brand name and note the count each month. Nobody searches for a name they did not notice.
How to go viral on TikTok for business: what not to do
Most guides on how to go viral on TikTok for business list things to add. These are the things to leave out.
Do not buy followers. TikTok says follower count is not a direct factor in recommendations, so a bought number buys you nothing in the feed. Bought accounts do not watch, do not finish and do not buy. They are the only audience that never complains, which should tell you something.
Do not chase trends that have no fit. A trending sound or joke can lift one video. If the trend has nothing to do with your product, the viewers it brings came for the trend. Use one only when you can put the product inside it without forcing it. If you have to explain the link, there is no link.
Do not delete videos that did badly. I found no TikTok statement that deleting a weak video helps an account or hurts it, so I will not claim either. My reason is practical. A weak video is a result in your test. Delete it and you are left comparing your formats on their good days only.
Do not post the same clip again and hope. TikTok says it does not recommend duplicated content. If you believe a video deserved better, change the opening and treat it as a new test.
Key takeaways
- Nobody can make one TikTok go viral on purpose. A company can build formats, test them and repeat the one that holds attention.
- TikTok says follower count and past hits are not direct factors in recommendations. Finishing a video is its own example of a strong indicator.
- The first seconds decide who sees the rest. Open on the product doing its job, with one line of text that works with the sound off.
- Judge a format by the typical result across its run, and ignore the best video in it.
- In the first two weeks measure watch time, full watches and where views came from. After that, count signups and searches for your name ahead of views.
- Leave out bought followers, trends with no fit and reposted clips. Keep your weak videos, because they are results.
If you want to know which formats your product could run on TikTok, ask for the free teardown. Send a link to the product and its channels. Within three days you get a written read on what brings people in today and which formats we would build first. No call, and the plan is yours whether or not we work together.