If you've read three blog posts from competing companies in the same category and couldn't tell which one came from where, you're not imagining things.
A 2024 Edelman and LinkedIn study found that 57% of B2B executives can't distinguish between competitors' content. Roughly six out of ten people who matter most to a B2B company's growth see the content put in front of them as interchangeable.
That's not a marketing problem. It's a system failure with three causes.
The volume trap
Marketing teams are under pressure to publish more. The pressure comes from somewhere reasonable. The CEO wants visibility, the sales team wants assets, the demand gen team wants pages for SEO. Volume becomes the goal because volume is measurable.
But publishing more usually means publishing faster. Publishing faster usually means publishing worse. The first content casualty is voice, the specific way a particular executive thinks and talks. Voice is slow. Volume is fast. They don't coexist.
When a marketing team has to ship four blog posts and twelve LinkedIn posts in a month, the posts that survive the production line are the ones that take the least time. The least time means the most generic.
The AI compound
Then add AI. As of 2025, more than half of long-form LinkedIn posts are AI-generated. Ahrefs found that 74% of new web pages in April 2025 contained detectable AI-written content. The web is being filled with what we'll call lowest-common-denominator content. Text that's grammatically fine, factually adequate, and indistinguishable from every other piece of text on the topic.
This is what happens when probabilistic text generators are pointed at the same prompts by everyone in a category. They converge on the same answers. The output looks like reasonable B2B writing because it is. It just isn't anyone's writing.
The result: a marketing manager scanning a competitor's blog has the unsettling experience of reading something that sounds exactly like what their own AI tool would have produced.
The agency factor
The third cause is the agency layer. Most B2B content agencies are built to optimize for output: pieces shipped, articles per month, posts per quarter. Their pricing reflects volume. Their workflows reflect volume. Their writers, often junior, are trained to produce passable content quickly across industries they don't deeply know.
This works for a marketing manager who needs to fill a content calendar. It doesn't work for a marketing manager who needs the content to actually sound like their CEO.
The agency optimization for output is the same optimization that AI tools are running on. Same prompts, same averaging effect, same indistinguishability. Hiring an agency to "produce content" without insisting on voice is functionally similar to using ChatGPT. Slightly more expensive, slightly more polished, identically generic.
What different looks like
If sameness is caused by speed, AI, and output optimization, the inverse looks like this: longer source material, deliberate human writing, agency models that optimize for voice instead of volume.
In practice, that means more time spent with the executive before any writing happens. It means fewer pieces produced per month. It means an editorial layer that catches the voice failures before they reach the audience. It means accepting that content programs run on quarterly rhythms instead of weekly ones.
Only 15% of B2B decision-makers rate the content they read as "very good or excellent." The other 85% is the market opportunity. The path to being in the 15% has nothing to do with publishing more. It has everything to do with publishing less, with more care, in a voice that's actually distinct.
That's not how most B2B content gets made. Which is why most B2B content sounds the same.