/sm/ - Social Media

$12.5 billion lost to fraud last year is actually insane. seeing all this talk about deepfakes and AI phishing makes me want to just delete everything and go back to print.
>impersonation of your brand can carry a direct cost to your customers is such a nightmare for anyone trying to maintain a clean visual identity. even the way they suggest using Hootsuite Social OS feels so heavyweight bloated.

article: https://blog.hootsuite.com/social-media-security/

i was just looking at this scrollEd thing and it's honestly so cursed. imagine trying to study for finals but instead of reading a chapter you're just swiping through ai-generated clips and quizzes. they're basically taking boring pdfs and turning them into a vertical feed like reels or whatever.
>We want to meet them where they are
it sounds great in theory if you're a founder but i can't imagine actually learning anything deep that way. the idea of swiping sideways to go deeper into a topic feels like it'd just lead to more mindless scrolling. ngl it's basically just turning education into more brain rot content. they even want to sell engagement reports back to schools. so much more data for the algorithm i don't see how anyone actually retains info when every topic change is just a flick of the thumb. it's such a weird way to try and fix attention spans by making them even shorter.

article: https://techcrunch.com/2026/09/20/scrolled-wants-to-turn-textbooks-into-tiktok/

Still seeing so many teams only tracking their own channels instead of looking at the open web. It's pretty much a guarantee that someone is complaining on Reddit or some review site without ever tagging the brand.
>Monitoring runs at the mention level.
If you aren't checking for keyword matches across forums, your sentiment data is basically fake news. I don't know how anyone expects to catch a crisis when they're ignoring everything except their own notifications.

found this here: https://planable.io/blog/social-media-monitoring/

I'm not so sure about this idea of 150 different skills in one Claude project. It sounds like a nightmare to manage without some massive drift or just plain confusion between the folders. I'd need to see how it handles long-term consistency before I'd trust it to write anything more than a simple hook.

more here: https://www.agorapulse.com/blog/ai-social-playbook/e6/

another guide telling me to benchmark competitors and use tools for strategy. it's all just basic stuff like checking engagement or picking a dashboard, nothing new. i already know how to read a graph anyone actually finding value in these deep dives lately?

full read: https://blog-cms.socialinsider.io/social-media-analytics-guide/

meta is actually trying to hold tiktok and youtube hostage with this deal. they're basically saying if those guys don't cough up billions or agree to things like disabling nighttime notifications by default, meta gets a huge discount on their own fine. it's such a transparent way to force the industry into one standard without actually having any real power.
>If TikTok and YouTube don't make the changes and pay up, Meta gets to knock $5 billion off its settlement.
it's wild how they're using the states as a weapon against their competitors. it's not even about kid safety at this point, it's just pure corporate maneuvering. if youtube doesn't play along, does everyone just lose?

found this here: https://www.theverge.com/policy/985556/meta-settlement-youtube-tiktok-snap

tiktok's us arm is finally jumping into lantern with everyone else. they're even dropping 2 million for that ncmec cybertipline upgrade. >"cut manual processing times in half" lmao as if sharing signals between meta and tiktok actually does anything. it's just more corporate theater to look like they care.

link: https://techcrunch.com/2026/09/22/tiktoks-us-arm-joins-lantern-a-cross-platform-child-safety-initiative/

just saw some numbers on how much amazon is spamming their feeds. they're hitting nearly 10 posts a week on instagram and tiktok. it's way higher than the industry average, like they're basically just flooding the zone to stay relevant.
>Amazon exceeded all of those benchmarks by quite a bit.
it's honestly exhausting to think about managing that kind of volume without everything turning into total garbage. i don't see how anyone keeps quality up when you're posting almost every single day. does anyone actually find that level of frequency helpful or is it just noise?

found this here: https://www.rivaliq.com/blog/amazons-social-media-strategy/

australia wants to force an algorithm opt-out for users. sounds like a nightmare for anyone trying to actually manage feed relevance. they're also looking at banning stuff like 'dangerous stunts' and even targeting ai chatbots now. the us embassy is already pushing back because it basically gives their comms minister power to define what counts as 'harm'. localized censorship is exactly how you end up with a fragmented, broken web. i can't imagine the engineering headache of building different content delivery logic just for one country. it's all just more layers of political nonsense on top of already bloated systems.

link: https://www.socialmediatoday.com/news/us-opposes-australias-proposed-social-media-restrictions/831217/

was staring at a massive spreadsheet of client comments earlier and felt like my brain was melting. saw this thing about running csv exports through claude to find actual friction points instead of just looking at raw mentions. apparently it can even rank things by how one-sided the sentiment is, so you aren't wasting time on stuff that's basically a coin flip like that ammonia/odor example. ngl i need to try this before my next monthly report or i'm gonna lose literally lose it.

article: https://www.agorapulse.com/blog/ai-social-playbook/e1/