This content originally appeared on DEV Community and was authored by Drew Madore
LinkedIn just killed the carousel hack.
You know the one. That “swipe for more” format that dominated feeds for the past two years, promising insights while delivering mostly repackaged platitudes with nice gradients. The 2026 algorithm changes—rolling out in phases starting January—represent the most significant shift in how B2B content gets distributed since the platform decided it wanted to be a “creator economy” player back in 2023.
And honestly? It’s about time.
I’ve spent the last three weeks digging into LinkedIn’s official guidance, testing content variations, and talking to folks who’ve seen their reach either crater or explode in the beta rollout. Here’s what’s actually changing and what you need to do about it before your carefully crafted content strategy becomes expensive noise.
What LinkedIn Actually Changed (And Why)
The headline feature: “Depth Score.” LinkedIn’s new ranking signal measures how long people actually engage with your content—not just whether they clicked or tapped. Reading time, comment depth, saves for later, and shares to private messages all factor in.
Translation: the algorithm finally learned to spot the difference between “I stopped scrolling for 2 seconds” and “I actually read this and found it useful.”
The platform is also deprioritizing what they’re diplomatically calling “engagement optimization tactics.” That’s corporate speak for:
- Posts that explicitly beg for engagement (“Agree? Comment below!”)
- Artificial cliffhangers designed solely to drive comments
- Content formatted specifically to game the algorithm rather than serve readers
- Recycled viral content with minimal original insight
LinkedIn’s VP of Product mentioned in a December briefing that engagement bait had become so prevalent that genuine expertise was getting buried. Their internal data showed that 60% of high-engagement posts in Q3 2025 used at least one optimization tactic, but user satisfaction scores were declining. People were engaging out of habit, not because they valued what they saw.
The fix? Make the algorithm care about satisfaction, not just activity.
The Three Content Types That Now Win
After testing dozens of post variations and watching what’s performing in the beta, three formats are consistently outperforming everything else:
1. Detailed First-Person Experience
Not generic advice. Specific stories about what you tried, what failed, what worked, and what the actual numbers looked like. The algorithm seems particularly good at detecting specificity now—posts with concrete details (company names, exact metrics, specific timeframes) are getting 3-4x the reach of generic “here’s my framework” content.
Example: A B2B SaaS marketer I know posted about a failed product launch, including the exact budget ($47K), timeline (8 weeks), and what went wrong (positioning focused on features their target audience didn’t care about). It got 10x her normal reach and generated actual business conversations. Her previous “5 tips for product launches” carousel? Barely broke 1,000 impressions.
2. Original Research and Data
Anything involving proprietary data, original surveys, or novel analysis is getting massive distribution. LinkedIn’s betting that this type of content can’t be easily replicated or gamed. You actually have to do the work.
The bar isn’t impossibly high here. A 50-person survey of your customer base counts. An analysis of trends in your industry using public data counts. You don’t need a research team—you need a perspective and some numbers to back it up.
3. Substantive Commentary on Industry Developments
Not hot takes. Not “here’s what I think about [trending topic].” Deep, informed analysis that demonstrates you actually understand the nuances. The kind of post where someone reads it and thinks “okay, this person clearly knows what they’re talking about.”
This connects to broader shifts we’re seeing in how platforms evaluate expertise—AI in Content Marketing: 2025 Strategy Guide covers similar themes around demonstrating genuine knowledge versus surface-level content.
The algorithm appears to weight comments from people in your industry more heavily now. If your post generates discussion among people with relevant job titles and experience, it gets more distribution than if it gets generic “great post!” comments from engagement pods.
What’s Getting Buried (Finally)
Let’s talk about what doesn’t work anymore. Because if you’re still doing these things, your reach is about to fall off a cliff:
The “Swipe Carousel” Format: Unless your carousel contains genuinely unique information, it’s dead. LinkedIn’s specifically targeting this format because it became the poster child for style over substance. Those beautifully designed 10-slide decks that say what could’ve been said in two sentences? The algorithm now treats them as lower-quality content.
I tested this. Same information, three formats: a carousel, a text post, and a short article. The text post got 5x the reach of the carousel. The article got 8x.
Engagement Bait Questions: “Agree or disagree?” “What would you add?” “Tag someone who needs to see this!” The algorithm now detects these phrases and patterns. Posts containing them are getting deprioritized unless the surrounding content is exceptionally strong.
Reposted Viral Content: Sharing someone else’s viral post with “Thoughts?” as your only addition? That’s now treated as low-effort content. LinkedIn wants original perspectives, not content curation without commentary.
Generic Motivation Posts: The “Monday Motivation” industrial complex is finally getting what it deserves. Unless you’re adding genuine insight or personal story, inspirational quote graphics are basically invisible now.
Yes, another algorithm update that promises to reward “quality.” Where have we heard that before? But here’s the thing: the early data suggests this one might actually work. The correlation between content depth and reach is stronger than I’ve seen in years.
How to Adapt Your Content Strategy (Practically)
Alright, enough diagnosis. Here’s what to actually do:
Shift Your Content Calendar
Stop planning posts. Start planning perspectives.
Instead of “Monday: motivation post, Wednesday: industry news, Friday: team highlight,” think in terms of substantive pieces you can actually say something meaningful about. Quality over quantity isn’t just a platitude anymore—it’s algorithmically enforced.
I’m seeing better results from 2-3 substantial posts per week than daily content. The math has changed. One post that generates 20 minutes of total engagement from your network is worth more than five posts that get quick likes and nothing else.
Invest in Longer-Form Content
LinkedIn articles and newsletters are getting unprecedented distribution right now. The platform wants to compete with Medium and Substack, and they’re using the algorithm to make that happen.
Here’s what surprised me: articles don’t need to be 2,000 words to perform well. 600-800 words of actual insight outperforms 2,000 words of fluff. The algorithm seems to measure completion rate, not just length. Write as long as you need to make your point well, then stop.
Lead With Specificity
Your opening line matters more than ever. The algorithm is measuring how many people expand your “see more” fold. Generic openings kill that metric.
Bad: “Content marketing is changing in 2026.”
Good: “We spent $23K on LinkedIn ads last quarter and generated exactly zero qualified leads. Here’s what we should have done instead.”
The second one makes people want to know more. The first one sounds like every other post in their feed.
Build Real Engagement Loops
Comments still matter, but the quality bar is higher. A single thoughtful comment from someone in your industry is worth more than 50 “great post!” responses.
This means you need to:
- Ask specific questions that require thought to answer
- Respond substantively to comments (not just “thanks!”)
- Tag specific people when their expertise is genuinely relevant
- Create content that industry peers will want to engage with, not just like
Engagement pods are effectively dead. The algorithm can detect when the same group of people always engages with each other’s content within minutes of posting. Those signals now count for almost nothing.
Test Everything With New Metrics
Stop optimizing for likes and comments. Start tracking:
- Read time (LinkedIn now shows this in analytics)
- Saves and shares (especially private shares)
- Profile views from the post
- Actual business conversations initiated
A post with 50 likes that generated three qualified leads is infinitely more valuable than a post with 500 likes that did nothing. The new algorithm is trying to optimize for the former. Your strategy should too.
The Document Native Strategy
Here’s a tactical hack that’s working extremely well right now: LinkedIn’s native document uploads.
Upload a PDF directly to LinkedIn (not a link to an external document—actually upload it). These are getting absurd reach because:
- They require more investment to create (signals quality)
- They keep people on LinkedIn (platform likes that)
- They generate longer engagement times (the depth score loves this)
- They’re still relatively uncommon (first-mover advantage)
I’m seeing document posts outperform equivalent text posts by 5-10x. The format forces you to create something substantive—you can’t phone in a 300-word PDF the way you can phone in a text post.
Format ideas that work:
- One-page frameworks with actual detail
- Case study breakdowns with real numbers
- Research summaries with original analysis
- Process documents from your actual work
The key is making something you’d actually use yourself, not just something that looks good in a feed.
What This Means for B2B Brands
If you’re managing LinkedIn for a B2B company, this is both a challenge and an opportunity.
The challenge: You can’t just assign someone to “post daily on LinkedIn” anymore. That person needs actual expertise and insights to share. The algorithm won’t bail you out with distribution just because you showed up.
The opportunity: Genuine subject matter experts at your company can now build reach without gaming the system. The playing field is more level than it’s been in years. A well-informed post from someone with real expertise will outperform a perfectly designed carousel from a social media manager who doesn’t understand the topic deeply.
This means:
- Enabling employees with actual expertise to post (with support, not scripts)
- Investing in original research and data you can share
- Creating content that demonstrates knowledge, not just awareness
- Measuring success by business impact, not vanity metrics
The brands that will win on LinkedIn in 2026 are the ones that treat it as a platform for demonstrating expertise, not just broadcasting messages. Which, honestly, is what it should have been all along.
The Contrarian Take: Maybe This Won’t Last
Look, I’m optimistic about these changes, but I’m not naive. LinkedIn has announced “quality-focused” algorithm updates before. They don’t always stick.
The platform still needs engagement to sell ads. If deprioritizing engagement bait causes overall activity to drop significantly, they’ll adjust. They always do.
But here’s why I think this time might be different: LinkedIn’s competition isn’t other social platforms anymore—it’s Substack, Medium, and industry-specific communities. They need to become a destination for substantive professional content, not just a place to humble-brag about job changes.
The algorithm changes support that strategic direction. Whether they maintain this direction when engagement metrics inevitably dip in Q1? That’s the real test.
What to Do This Week
Stop reading and start adapting:
Audit your last 10 posts: How many contained genuine insight versus content for content’s sake? Be honest.
Identify your actual expertise: What do you know that others in your industry don’t? That’s your content foundation now.
Plan 3 substantive pieces: Not posts—pieces. Things that require thought and demonstrate knowledge.
Test the document format: Create one native PDF upload this week. See how it performs.
Change your metrics: Add depth metrics to whatever dashboard you’re using. Stop leading with likes.
The 2026 algorithm isn’t perfect. It’ll evolve, people will find new ways to game it, and LinkedIn will adjust again. That’s how this works.
But right now, there’s a window where substance actually wins. Where you can build reach by being genuinely helpful instead of algorithmically clever.
That window won’t stay open forever. Use it while you can.
This content originally appeared on DEV Community and was authored by Drew Madore