How to Grow on LinkedIn in 2026: What 360Brew Actually Rewards (and Punishes)

Your reach dropped and none of your old tricks are working anymore, right? You’re not imagining it. LinkedIn quietly rebuilt its entire feed system, and this time there’s an actual documented playbook behind what it rewards and what it punishes.

This post walks through exactly what changed, in plain English, plus the specific things you can do this week to grow instead of just guessing.

What Actually Happens the Moment You Hit Publish

Okay so here’s the part almost nobody explains clearly. On March 12, 2026, LinkedIn’s engineering lead Hristo Danchev confirmed the platform had rebuilt its ranking system from scratch, replacing several separate smaller systems with one unified AI model. People online call this new system “360Brew.” Every post you publish now goes through three stages.

Stage 1 – Quality classification. The moment you hit publish, the AI sorts your post into a bucket: spam, low-quality, or high-quality. Posts with real specifics – actual company names, exact numbers, a real example – pass through cleanly. Generic, copy-paste style content gets filtered here before it even reaches anyone.

Stage 2 – The golden hour test. Your post goes out to a small, targeted sample first. Think of it like a taste test at a grocery store – a few people try the sample before the store decides whether to put it on the main shelf.

During this window (roughly 60 to 90 minutes), the AI watches how long people actually read your post, whether they save it, and whether comments are substantial rather than just “great post!”

Stage 3 – Expanded distribution. If that small sample engages well, your post gets shown to more people with similar interests. And here’s a nice bit of good news – unlike the old system, a good post can now resurface and keep circulating for two to three weeks, not just die after 24 hours.

The Signals That Actually Move the Needle Now

Likes used to be the whole game. Not anymore. According to Richard van der Blom’s independently published Algorithm Insights Report 2025 – one of the most widely cited studies on LinkedIn performance – the platform now weighs signals very differently.

See that? A like is basically the weakest signal on the whole platform now. A single DM share or save is worth far more than a pile of likes. Makes sense when you think about it – a like takes zero effort, but someone bookmarking your post or sending it to a colleague means it was actually useful to them.

By the way, this also means personal profiles now reach people far more effectively than company pages. If you run a business, post as a person, not as the brand. The AI seems to trust individual voices more than branded accounts.

Which Content Formats Are Winning (and Which Are Dying)

Format choice matters more now because different formats naturally produce different reading time, and reading time is what the AI is watching closely.

Honestly, the newsletter format surprised a lot of people. It bypasses the whole ranking system because it lands directly in someone’s inbox. If you already write long-form posts, turning your best ones into a newsletter issue is one of the easiest wins available right now.

8 Things Quietly Killing Your Reach Right Now

This new system doesn’t just reward good behavior – it actively suppresses patterns it recognizes as manipulative or low-effort. Here’s what to avoid:

• Engagement pods. Coordinated early liking is now detected with high accuracy and can quietly limit your reach for weeks.
• Generic AI-written content. Not AI content itself, but content nobody actually finishes reading – thin, generic posts that don’t hold attention.
• More than 5 hashtags. Hashtags are no longer a real distribution tool. Piling them on now reads as a spam signal.
• Engagement bait phrases. Things like “agree or disagree?” or “drop a yes if you feel this” are recognized and down-weighted.
• Copy-paste frameworks. The classic “hook line, personal story, bullet list, CTA” template is now easy for the AI to pattern-match and suppress.
• Tagging 10+ people. Mass tagging reads as artificial amplification, not genuine relevance.
• Links in the middle of your post. External links anywhere in the main body text get a real reach penalty.
• “Comment X to get Y” inside your caption. This one’s big enough that it deserves its own section – keep reading.

The Lead Magnet Workaround Everyone’s Missing

Asking people to comment a word to unlock a free resource used to be one of the most effective growth tactics on LinkedIn. It still works – the trick is where you put the ask.

Here’s the thing. The AI reads your caption text to decide what your post is about and whether it looks like bait. A caption that says “comment AUDIT to get the guide” gets classified as engagement bait immediately, no matter how genuinely useful the resource actually is.

The workaround: put the call-to-action inside your image, not your caption.

1. Write your caption completely normally. No mention of “comment,” “DM,” or any trigger word anywhere in the text.
2. Put the actual instruction inside your visual – something like “Comment AUDIT to get the resource” printed directly on the image.
3. End your caption with something like “the image below tells you exactly how to get it” – this points people to the visual without tripping the penalty.

Why does this work? Because the AI reads your caption text, but it doesn’t read the words inside your image the same way a person does. Your caption stays clean in the algorithm’s eyes, while your actual audience – real humans scrolling – sees both the caption and the image and understands exactly what to do.

One real example: a post using this exact method went from under 2,000 impressions on a typical caption-CTA post to 16,000 impressions on the first post using the image-only CTA, with noticeably better engagement across the board. Wait – worth noting, this was one specific post, not a guaranteed outcome for every account, but it lines up with how the system is designed to read text versus images.

A Simple Weekly Plan You Can Actually Follow

You don’t need a complicated content calendar. Consistency matters more than volume under this new system. Here’s a simple version you can actually stick to:

• Post once a day, max. Posting multiple times a day is now penalized. One genuinely good post per week beats five shallow ones.
• Comment on 5 to 10 posts from people in your target audience before you publish your own post. This warms up your account’s signal.
• Reply to every comment within 30 minutes of posting. Fast replies generate meaningfully more total engagement on the thread.
• Follow up with people who engaged within 24 to 48 hours. A personalized connection note or DM to someone who commented has a far better response rate than cold outreach ever does.
• Stay on one topic. Jumping between unrelated subjects confuses the algorithm’s read on your expertise and caps how far your content travels.

No worries if you can’t do all five perfectly from day one. Pick two or three and build from there.

Common Questions People Ask

Do likes still matter at all on LinkedIn?

Barely. Likes are now the weakest engagement signal on the platform. Saves, DM shares, and substantial comments carry far more weight in deciding how far your post travels.

Is it still worth asking people to comment for a resource?

Yes, but not in your caption. Put the instruction inside your post image instead of the text, since the algorithm reads caption text closely but doesn’t treat image text the same way.

How often should I actually post?

Once a day at most, and one solid post a week is often enough to build momentum. Posting several times a day is now actively penalized rather than rewarded.

Does my follower count still matter for reach?

Less than before. The AI evaluates your content and profile signals directly, so even a smaller account can reach the right audience if the content is genuinely specific and useful.

Should I still use hashtags?

Sparingly. Since the AI reads the actual meaning of your post, hashtags aren’t needed as a topic signal anymore. Using more than five can actually work against you.

Quick Recap

• LinkedIn rebuilt its ranking system from scratch, confirmed by engineering lead Hristo Danchev on March 12, 2026.
• Every post passes through quality classification, a small test audience, then wider distribution – in that order.
• Saves and DM shares now outrank likes by a wide margin as distribution signals.
• Carousels, native video, and newsletters are performing well; link posts and hashtag stuffing are actively suppressed.
• Lead magnet CTAs still work – just move them into your image instead of your caption.

What Should You Do Next?

Your one action right now: take your next planned post and move any “comment X for Y” style CTA out of the caption and into the image instead. That’s it. You don’t need to rebuild your whole content strategy today. Just try this one adjustment on your next post and watch what happens to the reach.

If you’d rather have a second pair of eyes on your whole profile first, DM the word “audit” on LinkedIn. You’ll get a full written breakdown of what’s working, what’s confusing the algorithm, and where to focus next – free, no setup required.

By the way, if you’re curious how AI and automation can help outside of LinkedIn too, BULDRR AI has more beginner guides on that, plus a free library of ready-made automation workflows at BULDRR AI’s free workflows library if you ever want to see what’s possible without building anything from scratch.

FAQ

Q: What is 360Brew? A: 360Brew is the name widely used for LinkedIn’s rebuilt, unified AI ranking system, confirmed by LinkedIn engineering lead Hristo Danchev on March 12, 2026, which replaced several separate smaller ranking systems.

Q: What is the “golden hour” on LinkedIn? A: It’s the roughly 60 to 90 minute window after you post where the algorithm shows your content to a small test audience and watches read time, saves, and comment quality before deciding how far to expand distribution.

Q: Why did my LinkedIn reach drop even though I didn’t change anything? A: LinkedIn rebuilt its entire ranking system in March 2026. Tactics that worked under the old rule-based algorithm, like early engagement tricks, carry much less weight under the new AI system.

Q: Are engagement pods still effective on LinkedIn? A: No. Coordinated early engagement is now detected and can suppress a post’s reach for an extended period, since the system evaluates genuine relevance rather than just early activity.

Q: What’s the biggest content format opportunity right now? A: LinkedIn newsletters. They land directly in a subscriber’s inbox instead of competing for feed distribution, and converting your best long-form posts into newsletter issues is a low-effort way to use that.

Q: How do I ask for comments without getting penalized? A: Put the request inside your post’s image rather than your caption text. The algorithm reads captions closely for bait phrases but doesn’t evaluate image text the same way.