There are roughly a million people on LinkedIn who will tell you how LinkedIn works. Exactly one of them gets cited by the other 999,999.

Richard van der Blom publishes the Algorithm Insights Report, a 200-page annual study now in its seventh edition, built on 1.3 million posts and more than 1,400 hours of research. When the LinkedIn industry makes a claim about reach, this is the document it cites.

That sentence is the whole business model, and it's worth reading twice if you're a coach wondering why nobody treats your opinion as the last word on anything.

The trick is not the posting. It's the arithmetic.

Most experts compete on interpretation. They read the same public signals as everyone else, add a personal spin, and hope charisma closes the gap. It's an argument you can never definitively win, because the person replying in your comments is working from exactly the same information you are.

Van der Blom skipped that fight. His credibility rests on a measurement only he has run, so every claim he makes is checkable and nobody can contradict him without doing the work themselves.

It started small. He published his first LinkedIn algorithm report years ago as a curiosity project. This year, he and his team analyzed over 1.3 million posts from 50,000 creators for the 2026 edition, and the headline finding was brutal: reach is down 60% for active creators over the last two years.

Jay Clouse, who hosts Creator Science and has interviewed most of the serious operators in the creator economy, introduced him on the show this June with a line that is effectively the market's verdict. "For years, Richard has published the most comprehensive annual research report on the LinkedIn algorithm that I have personally found," he said.

What his own posting actually looks like

One analysis of his account, covering 97 posts published between April and July 2026, found he posted 6.2 times a week and averaged 273 reactions. Fifty-six of those posts mention the algorithm directly. Sunday is a full working day for him. Saturday is the only time he eases off.

The raw reaction numbers are not the interesting part. The composition is.

A median of 242 reactions on 272,000 followers, and a 42.1% comment ratio, roughly seven times the norm. "Interest" is his second most common reaction at 14.4%, the highest in that series of teardowns — the tell of an audience that reads to learn.

Translation for anyone selling expertise: his audience is not scrolling past and tapping a thumb. They're stopping to write something back. That is the metric coaches should actually envy, and it almost never comes from posting more often. It comes from posting things people didn't already know.

The post that proves the whole thesis

Look at what performed best for him, because it isn't a story, a hot take, or a carousel about morning routines.

His biggest post reported that connection-based distribution fell from 70% in 2024 to 54% in 2025 to 10% in 2026, while the Interest Graph climbed from 5% to 21% to 50%. It earned 1,458 reactions and 471 comments, roughly double anything else in the sample.

In plain terms: LinkedIn largely stopped showing your posts to your network and started showing them to people interested in your topic. Reach now follows interest rather than connections, and van der Blom is the person who measured it first.

The type of post that wins for him is a finding that overturns something people believe. Not a belief. A finding. There is a difference, and the difference is a spreadsheet.

The business underneath the PDF

The report is not a lead magnet. It's the product.

His site pitches it directly, with the social proof baked into the button: "Join + 4,000 clients. Pay once. All future updates included."

Then there's the corporate tier, which is where knowledge businesses usually find their real margin. Companies get the full Algorithm Report 2026, a co-branded cover page with their logo, internal distribution rights for all licensed employees, and a copyright-compliant license agreement. A 75-minute live session is available too: 60 minutes on key findings, 15 minutes of open Q&A, open to all employees, flat fee regardless of tier size.

One research asset. Consumer price point, enterprise price point, and a speaking fee attached to the same 200 pages.

Meanwhile the free version of his authority compounds without him. Agency blogs, consultants and newsletters cite his numbers constantly — one marketing analysis this year called the report "the most widely cited independent benchmark of LinkedIn performance", and another built an entire nine-point playbook around his finding that one external link in the post body reduces median reach by 18.8%.

Every one of those citations is a competitor voluntarily telling their audience who the real expert is.

It is not a comfortable position

This is the part most people skip when they romanticize the model.

He's been targeted by LinkedIn's legal team, banned from the platform, and watched the third-party tools he relied on get shut down one by one. He keeps publishing anyway.

He's also living inside the same degraded feed as everyone else. "I normally get around 20 to 30 comments in the first five minutes and 80% of them are AI written," he told Clouse. His one-word description of LinkedIn in 2026: "turbulent."

The part that transfers

Nothing about this requires 1.3 million data points or a platform obsession.

It requires a question in your niche that nobody has bothered to answer with real numbers. What do executive coaching packages actually cost in your city. How many discovery calls convert in your vertical. What happens to retention after month four. How many of your peers raised rates last year and what happened next.

Fifty honest responses, cleanly analyzed and published every year with the date on the cover, is a defensible position. A hundred is a franchise.

The reason it works is unglamorous: opinions invite argument, and measurements invite citation. Coaches spend fortunes trying to sound like authorities. The cheaper route is to own a number the rest of the industry needs to borrow.