c/algorithm-watch

The meme page admin who insisted engagement was all that mattered

I got into it with a guy who runs a local meme page in Columbus about how he decides what to boost. He told me straight up that he doesn't care if something is true or even funny, he only looks at the first hour retention rate and shares. I pushed back because I've seen his page push a fake story about a school closing that got thousands of likes but wrecked parents' mornings for no reason. He just shrugged and said the algorithm rewards reactions, not accuracy, and that's the game. It stuck with me because my students repeat stuff they see on his page like it's gospel, and we end up spending half a class untangling a rumor. I guess my question is, for folks who run pages or channels, where do you draw the line between playing the algorithm and not making things worse? Has anyone actually turned down a viral boost because the content felt too harmful?

The 2 a.m. deepfake clip that flipped my stance on verification

I used to think fact-checking viral clips was overkill, but then I saw a side-by-side of a politician's speech where the fake version had 400k more views than the real one, posted in a random Discord. Has anyone else noticed how the algorithm seems to reward the altered version just because it gets shared first?

The $180 boosted post that got 11 likes and zero sales

I run a small Etsy shop for handmade dice bags. Last month I thought I'd try boosting a post on Instagram to reach more people. Put $180 into it over 2 weeks, targeting people who follow tabletop game accounts. Got 11 likes, 2 comments, and my analytics showed almost no click throughs to my shop. I checked the audience breakdown after and it was mostly bots from other countries. Has anyone else had this happen with boosted posts, or is there a better way to spend that kind of money on social for a tiny shop?

Pro tip: test your retweets before the algorithm buries them

Last Tuesday I ran a small test with a local bakery account I manage. I posted the same photo of a croissant at 9 AM and 9 PM to see which slot got more traction. The morning one hit 1,200 impressions in two hours, the night one got 87 before I deleted it. Then I spent an hour talking to a guy who runs a meme page with 300k followers. He told me he schedules for 6 AM because the algorithm rewards early engagement from the west coast. Has anyone else found a time slot that works way better than the usual 'post when your audience is awake' advice?

Why did my pin with zero hashtags beat my hashtag-packed post by 4,000 views?

Deleted my 200-follower account and the algorithm treated me like a brand new person

Started fresh on Tuesday with zero history, posted the same kind of content I always did, and got 3 times the reach in a day, so has anyone else noticed that a clean slate seems to reset whatever invisible penalty an old account builds up?

Just realized hashtags in captions get throttled but hashtags in comments don't

Tried both on the same photo for a local food bank event last week, and the comment version hit 3x the reach with zero extra effort. Has anyone else seen this split on their own accounts, or is it just my niche?

Shoutout to the 40% engagement bump after I ditched the hashtag spam

For 6 months I was doing the usual thing, stacking 20 plus hashtags on every post. Numbers were flat, like 2% engagement. Last month I cut it down to 3 relevant tags per post and focused on answering comments in the first hour instead. Engagement jumped to 42% on the last two posts. It felt backwards but the algorithm seemed to care more about the reply chain than the tag count. Has anyone else seen a bigger boost from killing the tag pile and just talking to people?

That $35 TikTok analytics tool was pure garbage for my small channel

I paid $35 for a "viral prediction" service last month and it just told me to post at 6pm on Thursdays, which I already knew from free tools. The one thing that did help was their breakdown of comment sentiment, but even that was just a fancy word cloud. Has anyone found a budget analytics tool that actually tracks how hashtags trend over time?

A/B tested two hashtag strategies on IG, one crushed it by 4x

I run a small vintage shop on Instagram and I've been trying to figure out why some posts blow up and others flatline. Last month I decided to split test my approach. I posted the exact same photo of a 1970s denim jacket twice, once with 15 niche hashtags and once with 5 huge ones like #vintage. The niche tags pulled in 340 reach in 24 hours, the big ones got 82. The difference was engagement rate too, people actually commented on the niche one because it hit a smaller crowd that cared. I also noticed the algorithm pushed the niche post to explore after about 6 hours, maybe because the early engagement was stronger. Anyone else see better luck with hyper-specific tags over broad ones, or is it just my little shop?

Deleted my TikTok after seeing the same recipe 14 times in one hour

Last Tuesday I sat down to look up a simple pasta dish. Scrolled for maybe 60 minutes and the exact same cooking video looped back 14 times, just with different background music. That's when I realized the algorithm wasn't suggesting things I liked, it was testing which version of the same content kept my thumb still. I deleted the app that night after 3 years of daily use. Now I'm on YouTube Shorts and the repetition is way less aggressive, but I still see the same talking head in every other video. Has anyone else tracked how many times a single piece of content repeats in one session?

The algorithm noticed me after I stopped stuffing hashtags

I run a small local food blog and for months I'd slap 25 hashtags on every post about Portland restaurants. My reach stayed stuck around 300 views. Last week I tried the opposite approach, I used just 5 hashtags that matched specific niches like #pdxfoodies and #foodblogger. Then I spent 15 minutes actually replying to people in those tags instead of just posting and leaving. That single post hit 4,200 views and got 80 new followers. It finally clicked for me that the algorithm rewards engagement depth over keyword spam. Has anyone else found that commenting in niche tags works better than the shotgun method, or does that just depend on the platform now?

Hit 10k followers in 48 hours off one Reddit crosspost, then watched it vanish

My old cooking demo clip from 2019 randomly got pushed to 10k followers overnight, but by Friday the reach dropped to 40 views because the platform flagged it as recycled content. Does anyone else see their old stuff resurge like that, or was it just a glitch in the ranking?

Deleted my viral cooking video after the algorithm pushed it to the wrong audience

I posted a 90-second pasta tutorial from my kitchen in Chicago and it got 400k views in 2 days, but the comments were all people mad it wasn't a political rant, so I pulled it down and now I'm wondering if anyone else has had their content hijacked by the recommendation engine like that?

The 20-minute video that made me check every upload date

Last Tuesday I watched a breakdown of a viral political clip that had 3 million views in a day. The person who posted it to our forum spent 20 minutes showing how the original upload was from 2019 and got 400 views. They pointed out that the algorithm re-surfaced it because engagement from new accounts spiked after the topic trended. That stuck with me because I realized the platform is not showing me what is new, it is showing me what is being pushed. Has anyone else noticed old content getting recycled with fresh timestamps?

Hit 100k impressions on a random Tuesday and it felt weird

I been posting the same kind of stuff for months, mostly local news takes and weird niche memes, and the algorithm never cared. Then out of nowhere last week a post about a broken traffic light in my neighborhood got 100k impressions by noon. Nobody I know saw it, but the algorithm just grabbed it and ran. I checked the analytics and the spike came from a single share in a city group, not even a big one. Has anyone else had a totally random post blow up for no logical reason?

That viral 'shadow ban test' everyone shared last month is total nonsense

Saw a bunch of creators posting that phrase 'shadow ban check' and then claiming their reach dropped to zero. I tried it on my TikTok and my views actually went UP that day. Then I realized the algorithm was just pushing me because tons of people were searching that term. So the real trick is making content about the fake trick. Feels like the algorithm punishes people who chase trends and rewards people who talk about the chasers. Has anyone else noticed this loop where covering a viral topic makes your posts pop harder than the original?

The "#1 trending" tab is the worst thing I used to optimize for

I spent 6 months chasing whatever was trending in my niche, posting 3 times a day, and my engagement tanked... then I noticed a small cooking account with 2k followers getting 10k views on a 2-week-old recipe because the algorithm kept surfacing it for search intent. Tipping point was when one of my "trending" posts got zero shares while an old tutorial from 2022 suddenly blew up. Anyone else find that chasing the trending tab actually trains your account to be ignored?

Can we talk about how one bad caption wrecked my whole TikTok push?

I was posting from a coffee shop in Austin last month, same hashtags that got 50k views before, and suddenly my reach dropped to 300 because I wrote a caption with a typo that triggered the spam filter. It took me 4 hours to figure out it was the word "fre3" in the first line, not the content. Anyone else get shadowbanned over something this dumb?

My dog's account got more views than my cooking page in one week

I run a small cooking page on Instagram, maybe 400 followers, just me filming my recipes with a phone stand in my Austin kitchen. Last month I posted a video of my dog sitting on the counter licking a spoon by accident, I didn't even edit it, just threw it up as a story. Next morning it had 12,000 views and my regular 3 minute pasta tutorial from the same week had 87. Now the algorithm feeds my dog content to people who have never seen my food, and my reel of a beef stew recipe got 40 views. I started testing it, posted my dog sleeping for 10 seconds and it hit 2,000 in an hour. Has anyone else had a random, non-topic clip completely take over their page while their actual niche work gets buried?

The 2017 Facebook shift that killed my local news page

I ran a small neighborhood news page in Austin from 2014 to 2019. Last month I looked back at the numbers and saw the exact drop point: March 2017, when FB changed the news feed to favor friends over pages. My post reach went from 2,000 views to about 300 in three weeks, no matter what I posted. It wasn't the content, it was the algorithm straight up deciding local updates didn't matter. Has anyone else pinned down a single update that wrecked their reach like that?

My 3AM posting habit made my tweets flatline for 6 months

I always scheduled my tweets for 3AM because I thought fewer posts meant less competition. Then I saw my analytics and realized my audience in Chicago was asleep, and the algorithm just buried my stuff before anyone woke up. Now I post at 8PM local time and engagement tripled, has anyone else nailed their timezone sweet spot?

Vent: That time TikTok's algorithm decided I was a 'cottage cheese influencer' for a solid month

Last month I posted ONE video of me mixing cottage cheese with ranch powder as a snack for my kids. Within 3 days, TikTok decided that was my WHOLE personality. My feed became nothing but cottage cheese recipes. Cottage cheese smoothies. Cottage cheese frosting. I went from seeing cute dog videos to watching a woman explain how to make cottage cheese 'bread' that looked like a sponge. I counted 47 cottage cheese videos in one afternoon scrolling session. Has anyone else had the algorithm latch onto ONE random thing you did and just refuse to let it go?

Can we talk about how Instagram pushed that 'clean girl' aesthetic for 6 months straight?

I spent a full week trying to find makeup tutorials that weren't that sterile beige look and every single search result gave me the SAME 10 creators on repeat. Has anyone else noticed a platform just lock onto one vibe and refuse to show you anything else?

Spent $150 on a viral posting tool and my Instagram reach actually dropped

I saw this tool called PostMatic being pushed by every influencer in my niche (gardening content, mostly). Paid $150 for a month thinking it would schedule posts around peak times better than I could. Instead, my engagement tanked by about 40% in two weeks because the algorithm detected the automated posting patterns and just stopped showing my stuff to new people. Turns out the manual, messy approach where I post when something actually happens in my garden works way better. Has anyone else paid for one of these growth tools and seen the algorithm punish them for it?