Is Being a LinkedInfluencer the Key to Your B2B LinkedIn Strategy?

TL;DR: If your organic LinkedIn strategy consists of publishing 2-5 times per week on your company page, you’re swimming upstream... and that's by design. 

In early 2026, LinkedIn rebuilt its feed to rank content based on topical relevance rather than network relationships. This change structurally limits a company page because it posts about a wide variety of topics, like jobs, new hires, events, self-promotion, the community, and so on. 

So, regardless of how many followers your company page has, the algorithm is actively working against your company page. This shift means the platform is now rewarding narrow expertise from identifiable people, not businesses. Something you may have already noticed in your feed, but haven't updated your company's approach. 

You post to the company page three times a week. Someone writes the copy, someone designs the graphic, someone schedules and publishes it. Your engagement is four likes and a comment… and you recognize three of the names because they work with you.

Meanwhile a competitor’s VP of Engineering writes one unpolished post a week (with typos) from their personal profile and pulls forty comments from people that work at 60% your ICP list.

The easy takeaway is to look at that and conclude you need better content with more detail and depth. Heck, you'll throw in some typos too, if that'll help. Not only is that the wrong lesson, but chasing it will cost you another quarter of strategy sessions, SME interviews, and translating the engineering-speak into something you understand (and then back to engineering-speak to sound authentic) before trying to apply it to your company voice.

The real takeaway should be that your LinkedIn company page can’t win the game the same way anymore. At least, not alone.

What Actually Changed on LinkedIn?

In March 2026, LinkedIn’s AI team published an engineering post describing a rebuilt feed ranking system. The shift in what the system is looking for was, and is, far more important than the technical details. But don't let us stop you from a good time diving into the technical side of it.

The old feed leaned on the social graph. It asked who you know, who you follow, and who your connections engaged with. Primarily, it valued your network and prioritized the content it showed you based on what was valuable to it. The new system represents both posts and members as embeddings, which are mathematical representations of meaning, then reads your interaction history as a sequence rather than as isolated clicks. LinkedIn’s engineering lead described the goal as better understanding what a post is actually about and how that relates to a member’s evolving professional interests.

To summarize: LinkedIn used to show you more of what you already liked, based on clicks and comments you made. Now, it tries to work out what you’ll want next based on why you clicked, commented, and spent time with previous content.

Back in March, before the engineering post landed, we were reading the early reporting and working out what it would mean for clients. We were inferring that LinkedIn was moving from keyword-adjacent matching to full context matching, weighing dwell time and everything you’ve interacted with rather than a handful of tagged signals. That’s roughly what shipped and is now the current reality.

Two practical consequences follow:

• A post can reach people who’ve never heard of you (much more than before), if the system can confidently determine what the post is about.
• A post can fail to reach your own followers if the system can’t understand how it fits within their interests.

Relevance now runs far ahead of the relationships. Your follower count still matters, but topical interest signals matter much more.

Why Do Company Pages Lose Under This New System?

A company LinkedIn page loses in this new system because they are topically incoherent by design.

Look at what a healthy B2B company page publishes in any given month.

• Job openings
• Trade show booth photos and recap
• Product updates
• Holiday closure notice
• New hire announcements
• Culture reel
• Industry or workplace award

Each post is reasonable on its own but, other than being about the happenings at the company, the account is about no particular topic in general.

The new LinkedIn ranking system (built to distribute content by topic) can’t build a stable signal for an account like that. Every post seemingly resets their topical profile in LinkedIn's algorithm each time. There’s no accumulated topical authority to inherit from post to post, because the account has never been about one thing long enough to establish any. LinkedIn isn’t penalizing your company page directly. It just can’t assign your company page a category.

An individual profile, on the other hand, can do what the company page understandably can’t. A structural engineer who writes about specification-phase risk becomes a credibly consistent contributor. The system learns what she’s about and starts putting her in front of people who care about that subject, including people who’ve never heard of her or her firm.

So Should You Stop Posting On Your Company Page?

No. It’s still valuable and performs three jobs that personal profiles can't.

• It’s the credibility surface for due diligence. When a prospect, partner, or investor checks whether you’re a real company, they look at the page, and an abandoned one reads as a warning sign.
• It’s central to recruiting, because candidates evaluate the page before they apply.
• And it’s the mandatory host for LinkedIn advertising, including Thought Leader Ads, which promote an individual’s post through your ad account.

Think of it this way: the company page is now a destination, not an engine. People land on it to check you out, usually late in their evaluation, not necessarily to discover you. Judge it accordingly. A page nobody discovers you through can still be the page that closes the loop when someone’s already interested and just needs some additional credibility to push them to a decision.

Keep it current. Stop expecting organic reach from it and stressing about a lack of comments on your content.

How Do You Verify This on Your Own Account?

Most of the reach statistics circulating on this topic come from companies that sell employee advocacy software, which is worth knowing before you quote them in an organic social strategy pitch. The numbers get repeated so widely they start to feel like consensus, but the underlying methodology is usually nowhere in sight.

Run this test instead. It takes about thirty minutes and it uses data you already have access to.

1. Pull your last 20 company page posts. In LinkedIn Page Analytics, go to Content and set a date range that captures twenty posts.
2. Tag each post with one primary topic. One tag only. Force the choice. Recruiting, product, culture, event, industry commentary, and so on.
3. Count your distinct topics. That’s what LinkedIn is looking at to categorize you.
4. Record the median impressions across those twenty posts. Use median rather than average so one outlier doesn’t distort the picture.
5. Pick one team member who posts semi-regularly. Ask for their last ten posts and the impression count on each. Record their median.
6. Compare, then adjust for audience size. Divide each median by the relevant audience, meaning page followers for the page and approximate connections for the individual.

How to read the result:

If your topic count is high and one team member is outperforming a page with several times their audience, you have your answer, and you found it in your own data.

The audit tells you the page can’t carry the topic. The fix is to put the topics where the algorithm can actually read them, which means deciding what your company is going to be known for and whose name goes on it.

What About Letting Employees Take Center Stage?

Building your company's LinkedIn presence through individual profiles means giving up some message control. Your subject matter experts may phrase things differently than your brand guidelines would. They may talk more technically than the company page typically would. Losing some of that control may make you, and your leadership, uncomfortable. But that type of authenticity is kind of the point. The overly polished corporate phrasing is what the feed is essentially filtering out while letting (and promoting) the richly detailed technical authenticity flow through.

This also means the audience belongs to the person rather than the company. This is where the term "LinkedInfluencer" comes from. Their personal profiles may grow and surpass the scale of your company profile. As a result, they may have opportunities come to them directly. That’s a risk that many are not comfortable with. However, the profile of your company also grows, by proxy, alongside the profile of the person that is sharing their unique knowledge and perspectives. As a bonus, it typically comes with a far wider audience than your company page could ever reach on its own.

It's usually a trade worth making. The alternative is a company page that leverages the same resources, but fewer people see. Three things reduce your risk exposure in a LinkedInfluencer strategy:

• Tie content lanes to company expertise areas rather than personal brand ambitions. The lane survives the departure even when the follower list doesn’t.
• Put two people in adjacent territory within each theme, so no single lane has one point of failure.
• Use the company page as the archive. Reshare the strongest employee posts so the institutional record lives somewhere you control.

One benefit almost nobody expects: this reactivates dormant networks.

When one of our team started posting more regularly, the people who surfaced weren’t strangers. They were past clients and old colleagues he hadn’t heard from in years, quietly liking posts and commenting. No inbound lead came of it that week, but that isn’t the point. Somewhere down the line, a decision-maker at a company you’d love to work with remembered you exist, on a day you spent zero marketing dollars to reach them. Company pages don’t have that power. Your people do.

How Do You Pick Content Lanes?

Choose three to five lanes for the entire company. Not per person. Total.

Each lane gets one owner who has genuine first-hand claim to it, meaning they do the work, sit in the meetings, and see the successes and failures. A lane without an owner who actually lives it produces generic content, which is exactly what you’re trying to escape.

The test for whether a lane is narrow enough: if three or more people on your team could plausibly write the post, the lane is too broad. Add more guardrails and focus it more to a specialty that can be attacked from different angles.

An engineering firm might run:

• Why projects lose money in the specification phase
• Permitting and interconnection timelines
• Constructability review as a cost control

A manufacturer might run:

• Supplier qualification failures nobody documents
• What changed in your industry’s compliance landscape this quarter
• Total cost of ownership arguments that buyers get wrong

Each of those is specific enough for one person to own it credibly, and narrow enough that a ranking system can understand what the author is all about.

To be clear, a lane is not a gag order. Nobody has to stop posting about their kid’s robotics team or the conference they just spoke at. The occasional off-topic post won’t undo anything. The lane is about where the weight sits. If roughly three out of four of someone’s professional posts land in their lane, the system has plenty to work with and will be able to see that pattern over time. What breaks the signal isn’t a bit of variety; it’s a feed with no clear focus at all.

Expect the lane conversation to be harder than the content. When we mapped this for our own team, the sticking point wasn't picking topics; it was that naming a lane means saying no (or not right now) to topics outside it. Leaders, in particular, resist it, because a specific lane reads like a limit on what they, or the company, does.

The lanes still have to be adjacent without overlapping, or your people spend six months teaching the algorithm the same thing.

What Cadence Can a Small Team Actually Hold?

The advice circulating on LinkedIn says post three to five times a week. But, that advice is written for creators whose full-time job is filming, writing, and posting content to see themselves. It isn’t written for a marketing director with a website migration, a trade show, and a rebrand on their plate this week.

Start with one post per person per week, sustained over a few months rather than 6 weeks. This approach beats a 12-post, three-week sprint followed by nothing for 3 months. Consistency, not volume, is what allows the algorithm to establish a topical signal in the first place. To a topic-matching algorithm, inconsistency reads the same way scattered topics do: as an account it can’t, and doesn’t want to, place. Remember, LinkedIn’s product is your content. If they can’t trust you to publish consistent, valuable content, why would they promote you?

What's a reasonable time commitment, you ask? Per post, expect roughly 45 minutes to write and 30 minutes of comment response in the hour after publishing. That last part isn’t optional. Early engagement is what determines whether a post continues to spread, so the author has to be available when it goes live. Scheduling a post for a time when the author is in a client meeting wastes the potential for the post to exceed baseline reach.

Rough illustrative math: three lane owners posting weekly costs your company under four hours a week in total. Compare that against what you currently spend producing company page content that reaches 120 people and gets a like from Gary in Accounting. Gary's great, but not necessarily who you hope to be engaging with on LinkedIn.

In an effort to scale content production, it’s tempting to leverage AI and build content automations to push that 4 hour per week estimate down to 30 minutes. But there’s a real risk to that.

What Is the “AI Slop” Button, and Why Does It Matter Here?

On July 30, 2026, LinkedIn added an option to the three-dot menu on every post that lets any member flag the content as seeming like AI slop. Chief product officer Hari Srinivasan announced it in a post on LinkedIn, framing the button as a training signal: “Slop is hard to define and the definition changes; this lets us tune our models and make better feeds.” Reporting on the rollout goes a step further, describing new classifiers that will reduce how much low-quality content surfaces in recommendations from outside a member’s own network.

To be clear, LinkedIn isn’t banning AI-assisted writing. The company acknowledged that plenty of people use AI to proofread or pressure-test an argument. At the same time, LinkedIn retired its own “enhance your post” AI rewrite feature and replaced it with a proofreading tool designed to preserve the writer’s voice. Read those two decisions together and the position is clear enough. Assistance is fine. Substitution and automation are the problems.

The scale of the problem explains the urgency. Detection service Pangram estimated that more than 40% of long-form and 30% of short-form LinkedIn posts were fully AI-generated. Treat that as a vendor estimate from a company selling detection, but the direction is hard to argue with if you’ve opened the app recently.

For anyone building a content strategy, this closes off the obvious shortcut. You can’t solve the reach problem with volume, because volume produced without a first-hand point of view now carries a distribution risk that didn’t exist 6 months ago.

It also gives you a self-audit. Readers have learned the tells, and so has the flagging crowd:

• Emoji-prefixed bullet headers
• The three-item rhythm applied to everything, regardless of whether there are three things
• Sentences built on the “X is not Y, it’s Z” construction
• Rhetorical questions used as section transitions
• Takeaways that would be equally true for any company in any industry

That last one is the real test. If your post would survive a find-and-replace of your company name with a competitor’s, it has no first-hand insight in it, and no amount of formatting will fix that.

How to Get Started This Week

Run the thirty-minute audit above. Count your topics. Then pick three lanes and three names, and get the first post out before you build a process around it.

What!? Action before strategy? Gasp! Yep. Sometimes an action bias pays off.

Not because planning is bad, but because of what you’re planning against. You don’t yet know which lane your SME can sustain, which one draws comments from buyers instead of peers, or which one they lose interest in by week three. Those answers exist in published posts, not in a planning doc. A quarter mapped out in advance is a quarter of assumptions you may have to rewrite anyway, and the rewrite is demoralizing in a way that starting small (but quickly) never is.

However, there are some caveats worth mentioning to the "run fast + learn on the fly" advice. Regulated industries, public companies, and anyone whose SMEs testify as experts need a clear legal review and approval path before anything goes out. Some teams genuinely will not start without a framework to point at. If that’s you, build the smallest one that unblocks the first post: the lanes, the owners, who reviews, and what’s off-limits. That proof-of-concept can be built in an afternoon.

What we’d steer you away from is the middle path that feels the most responsible and works the least. Two or three weeks of calendar built before anyone has published anything, treated as the plan rather than a directional hypothesis. Plan out two weeks if it helps you start, as long as you hold it loosely. Then publish, watch what happens, and let the next two weeks be shaped by what actually happens in the feed.

FAQ

Does LinkedIn penalize AI-written posts?

Not directly. LinkedIn has stated it doesn’t prohibit AI-assisted writing and acknowledges legitimate uses like proofreading. What changed in July 2026 is that members can flag posts as seeming like AI slop, and that feedback influences distribution beyond the author’s network. Generic content also produces low dwell time and few saves, which independently suppresses reach.

How many topics should a company page cover?

Fewer than you currently do, though the company page will always be broader than an individual profile because it has to serve recruiting, announcements, and brand credibility, among other things. That proves precisely why the page is the wrong vehicle for reach with the updated recommendation algorithm. Concentrate topical focus at the individual level.

Does picking a lane mean that person can’t post about anything else?

No. The lane describes where the majority of someone’s professional content sits, not a restriction on what they’re allowed to publish. Personal posts, company milestones, and the occasional detour are fine and often perform well. The signal comes from the pattern, not from purity.

What if our subject matter experts refuse to post?

Start with one willing person rather than mandating participation. A single credible voice publishing weekly in a narrow lane outperforms six reluctant participants posting quarterly. Reluctance usually signals a fear of saying something wrong publicly, which a defined lane and a light review step will address. Get a small win first and use that to demonstrate to the reluctant SME that it can be done. Watch for the opposite failure too: a team that debates a single post for three weeks has already missed the window the post was for.

Should employees write their own posts or should marketing ghostwrite?

Either can work, provided the insight originates with the expert, and provided the expert keeps veto power over anything published under their name. The failure mode is marketing generating both the idea and the language, which produces exactly the generic output the feed now filters. A workable process is a fifteen-minute interview to capture one specific observation, then drafting from that and sending it back for a read.

How long before this shows results?

Plan on a few months before a topical signal is rooted and a bit longer before it compounds. Anyone promising a faster path is usually selling something.