A few weeks ago, I uploaded a video of a $400 Miata I got from the junkyard idling. There was no background music, no edits. Just me quietly swearing, and finally starting on the third try. The video got 1.2 million views on TikTok. My “professional” cinematic reveal of my car from the week before, which took me nine hours to edit, got 3,400.
That gap taught me more about this platform than any growth course I’ve paid for.
Car content is not one thing
Everyone just clumps “automotive TikTok” together like it’s one audience, and I get annoyed by that. The guy who watches his LS swaps isn’t interested in the same thing as the guy watching his Radwood Volvo restoration, and neither is interested in the videos being put out by the dealer lot guys who just use https://celebian.com/buy-tiktok-followers to inflate their numbers. I made the mistake early on of trying to serve all three. My retention tanked. Once I picked a lane — budget builds, specifically stuff under $3k — my average view duration on Reels went from something like 4 seconds to closer to 11.
You don’t figure this out by guessing, either. I read my comments more than I’d like to admit. Not the “nice car bro” ones — the arguments. Someone will say “that’s not a real limited slip diff” and forty people will pile on, and that thread tells you exactly what your audience actually knows and cares about. That’s free market research most people scroll right past.
What actually performs (from someone who’s tested this too much)
Before/after clips still work. They just do. There’s a reason every rust-repair account with 200k followers posts the same format — grinder shot, weld shot, paint shot, done. It’s basically a cheat code, and I’m not too proud to use it.
POV driving clips with a weird audio choice outperform anything I plan carefully. My best-performing Short of the year is just onboard footage of a Civic Type R on a backroad with a sped-up Interstellar soundtrack over it — 890k views, almost no editing beyond syncing beats to shift points. I still don’t fully understand why that combination works as well as it does.
Myth-busting clips are the most reliable format I use, full stop. “Premium gas doesn’t actually make your car faster” type stuff. People comment to argue even when you’re right, and that argument is the algorithm’s favorite fuel.
The honesty thing isn’t a strategy, it’s just what happens when you stop performing
I bought a “clean title” Mustang off Facebook Marketplace for $6,800 and found out three weeks later it had a bent frame rail hidden under fresh undercoating. I filmed myself finding it. I filmed myself calling the seller, getting no answer, and eventually just eating the loss because chasing it legally wasn’t worth the hassle.That video remains, to this day, the video that people bring up to me when we talk on DM.
I don’t believe this happens because I’m an excellent story-teller. I believe it happens because in automotive content people only see the win but never the loss. They need to see the loss.
Posting consistently matters more than posting well, at least at first
I know that’s an unpopular thing to say. But I went from posting twice a week to daily for about four months, and even my mediocre videos during that stretch started outperforming my “good” videos from the sporadic-posting era. The algorithm seems to reward the pattern more than any individual video’s quality, which honestly feels a little unfair when you’ve spent real time on something.
Timing has two layers to it. Obviously you want to catch people scrolling — for me that’s late evening, 8 to 10pm, when my audience skews younger. But there’s a seasonal layer too that people underrate: winter driving content spikes hard every November whether it’s good or not, and I’ve gotten lazy videos to 500k+ just by posting them the week of the first snowstorm in my region.
When something actually goes viral
Here’s what I do the moment a video takes off, and I mean within the hour: I pin a comment that teases the next video, I go look at exactly where in the video people dropped off (usually it’s obvious once you check analytics), and I try the same core idea again with one thing changed instead of trying to recreate the whole video.
That viral window is short. Maybe 48 hours where new people are actually deciding whether to hit follow. If your profile grid is a mess of unrelated content when they land on it, you lose most of them. I learned this the hard way — had a video hit 2 million views and gained maybe 4,000 followers off it, way below what it should’ve converted, because my page looked like three different channels mashed together.
Look, I’m not going to pretend there’s a formula that guarantees Authentic Growth. I’ve had videos I thought were unremarkable blow up and videos I was proud of disappear into nothing. But the creators I know who’ve actually built something lasting all do the boring stuff: they know exactly who they’re talking to, they post more than feels comfortable, and they treat a viral hit as a starting line, not a trophy.
