The TikTok Experiments, Vol. 8: Posting Only 3 Minute Videos for a Week
/
All year long, I’ve been conducting a series of TikTok experiments in order to understand what kind of content the powerful TikTok algorithm prioritizes. I’ve tried an SEO trick, CapCut templates, shorter videos, a higher posting frequency, and more. In July, I experimented with longer videos, sharing only 3 minute videos for one week. The result? It worked better than I anticipated! Here’s what I learned from sharing longer videos on TikTok for a week.
Why longer videos?
Lately, I’ve seen TikTokkers claiming that the algorithm is pushing longer-form videos. Not YouTube length content–just original content that’s longer than 60 seconds. 3 minutes is actually the maximum length if you’re filming natively in TikTok; videos can be up to 10 minutes long if you upload clips to the app.
After experimenting with 15 second and 60 second videos, I decided it was time to try sharing 3 minute videos to see how the algorithm liked them. The 15 second and 60 second video experiments were both a bit of a flop, so I wasn’t expecting much.
What kinds of 3 minute videos did you create?
I shared 14 videos over the course of the experiment. I set a personal rule that the videos didn’t have to be exactly 3 minutes; as long as they ran over 2 minutes long, that was fair game. The videos I shared covered topics like:
Dogs – I didn’t share as much dog content this time because my dog videos are usually shorter; I shared one video offering advice about adopting a bonded pair of dogs.
Books/Movies – I shared 3 book reviews, a review of the Barbie movie, and thoughts on switching from Goodreads to StoryGraph.
Food Reviews – My husband and I reviewed the new Mario Oreos and candies from an international food store.
Social Media Advice – I announced the launch of this experiment, shared advice on choosing SEO keywords for TikTok videos, and offered insights on things to consider before interacting with a brand on social.
Which 3 minute videos performed the best?
Several of my 3 minute videos performed well! In my two most recent experiments, the best-performing videos received 3,900 views and 2,700 views. This time, my most popular video received 7,100 views! It was a video sharing my honest review of the Barbie movie, shared the day after it came out, so I think sharing about a topic that the whole internet was talking about helped it get great reach.
Interestingly, it didn’t generate great engagement: only 2 comments! Also worth noting: this video did not hit 7k within days of posting. TikTok’s algorithm still operates on a delayed release. The video hit 1k about a week after I posted it. By the end of the second week of being live, it hit 7k! I love that content has a longer lifespan on TikTok compared to other platforms.
talking about the barbie movie in front of a green screen photo from the panel discussion i attended
Other popular videos generated more engagement:
A review of the bestseller Tomorrow, and Tomorrow, and Tomorrow received 519 views, 21 comments, 2 saves, and 1 share. I argued that it did not live up to the hype and many readers agreed!
My explainer on switching from Goodreads to Storygraph received 503 views, 12 comments, 4 saves, and 2 shares. This information was helpful to a niche, engaged audience of readers who enjoy tracking their reading on social apps.
I reposted one of TikTok’s “on this day” videos, a review of Aubrey Gordon’s first book, and it received 448 views, 6 saves, 12 comments, and 1 share. The book is still highly relevant and folks enjoyed the video and expressed an interest in reading the book.
Interestingly, videos of screen recorded clips performed poorly. I shared clips of the Liberty’s Kids computer game and did not receive any comments or shares; a review of the app Kinder World received only 145 views. The algorithm does seem to prefer original content: humans talking. Not screen recordings, even if you think the content is interesting! I wonder if I had used a screenshot of the game as a green screen and talked over that, if it would have performed better. My guess is yes. People need a human to connect with.
a screen recorded tour of the kinder world app did not perform as well as i hoped
As I’ve mentioned before, I often get stuck in what folks call the “200 views jail,” where it seems like all your videos are only reaching 200 people. Lately, most of my videos are surpassing 300 views, so that’s been positive!
Should I be sharing more 3 minute videos?
Initially, I thought this experiment flopped. 300 views per video was good, but it didn’t seem particularly impressive. I gained 28 followers during the week of the experiment, which is more growth than I saw in past experiments. My theory is that these videos reached a more niche audience who were genuinely interested in my content, so they were more inclined to follow.
Once again, looking at the stats a few weeks later convinced me that this tactic is effective:
Video views up 36% – 24k views the week of the experiment, compared to 17k the week before and 20k the week after. It seems like TikTok may genuinely be serving longer videos to more people.
Unique video views up 32% – 16k the week of the experiment; 12k the week before; 16k the week after.
Likes up 27% – This content genuinely resonated with folks it reached! I received 1,096 likes during the experiment, compared to 861 likes the week before and 929 the week after.
Profile views up 50% – I found this stat pretty shocking! I received 268 profile views the week of the experiment; with 178 the week before and 167 the week after.
Comments up 21% – This was a smaller increase in engagement, but hey, I’ll take it! 118 comments the week of the experiment, compared to 97 the week before and 95 the week after.
Shares up 144% – I’ve often had trouble increasing my shares, and I’m not sure what it was that made these videos more shareable! Perhaps it’s because they provided genuinely useful information. I received 105 shares the week of the experiment, 43 the week before, and 76 the week after.
The growth here isn’t as dramatic as it was with my SEO keywords experiment (a 1,700% increase still has me blown away!), but a 36% increase in video views and a 50% increase in profile views is still great!
That being said, I did find this experiment much harder than past ones. For my previous experiments, I’ve been able to record most of the videos in advance and publish them throughout the week. This time, I struggled to think of topics that I could talk about for 3 minutes straight. I don’t think it’s worth publishing content if you don’t genuinely have value to add, so it was tough to generate enough valuable content.
I would not recommend pivoting to only 3 minute videos. Ultimately, this is something to implement in moderation, as a piece of the larger content puzzle. I’ll keep sharing a few 3 minute videos every week, but I won’t share 14 in one week again!
So far I’ve learned that CapCut templates and including SEO keywords can help boost your reach. And now I can add longer videos to my arsenal as well! What should my next TikTok experiment be? Send me an email if you have a suggestion for a future TikTok experiment.
Looking to up your TikTok game in 2023? Book a TikTok consultation!
