Last fall I audited 40 mid-market SaaS blogs as part of a research project. I picked the 40 from a list of YC-backed series-A companies, all in B2B, all with active content programs. The exercise was simple: for each blog, take three random posts. Strip any identifying information — company name, branding, screenshots, internal links. Hand the post to a peer and ask them to guess which company wrote it.
Out of 40 companies, 4 produced content that was identifiable. The other 36 produced content that read like generic SaaS marketing prose — competent enough, on-topic, factually correct, and totally indistinguishable from work that could have come from any of their direct competitors.
The frustrating part: those 36 companies were spending real money on content. Full-time content marketers, freelance writers, agency retainers, AI tools, editorial calendars. The investment was non-trivial. The output, viewed through this voice-distinctiveness lens, was indistinguishable from what an AI would generate with a one-line prompt.
Why content goes generic
The cause is upstream of the content team. When content is generic, the failure was made earlier:
- The brief is generic. “Write a blog post about [topic] for [persona].” Topic + persona without a specific angle, opinion, or worldview produces topic-summary writing — exactly the kind of content AI now generates for free.
- The reviewer is permissive. Content reviewers in most SaaS companies optimize for no obvious mistakes — accurate, polished, on-brand. They do not optimize for distinctive. So distinctive ideas get sanded down and removed in editorial passes, the way an essay loses voice through committee review.
- The writer has no skin in the game. Freelancers writing for 20 clients optimize for the median of what works across all 20, not for what works specifically for you. AI writes the same way — to the median.
- The company doesn’t actually believe anything. This is the root cause. If leadership can’t articulate three things that are true about the world that their competitors disagree with, the content team cannot manufacture distinctive voice. They will produce category-summary prose because that’s the safest output that won’t get flagged in review.
The fix is positioning, not better writers
When clients come to us frustrated with their content quality and asking us to write for them, we usually start somewhere else. We start with three questions:
- What does your category get wrong that you’re trying to correct?
- What does your typical buyer believe when they show up that you have to correct before you can sell to them?
- What would your strongest competitor never agree to publish? (And could you publish exactly that?)
If the founder can’t answer those questions, no amount of writing craft will produce distinctive content. We don’t accept ghostwriting engagements where the founder can’t answer those three questions; the work is unsalvageable.
If they can answer those questions but the content isn’t reflecting it, then we have a process problem we can fix. Usually it’s that the answers exist in the founder’s head but never make it into the content brief. We solve that with structured async founder interviews — voice memos, written prompts, transcript review — so the writing has a foundation of actual perspective to render.
A heuristic for whether your content has a voice problem
Take your last 5 blog posts. For each one, ask: would a reader who didn’t know which company published this be able to guess? If they couldn’t, ask: could we have written this post in a way where they could? If the answer is “no, because we don’t have a distinctive perspective on this topic,” then either:
- You shouldn’t be writing about this topic, or
- You should develop a perspective before writing about it, or
- You should write the post in a way that’s actually useful as a reference (the rare valid case for impersonal-summary writing)
The last case applies to ~20% of B2B content. The other 80% should have a perspective. If it doesn’t, you’re publishing wallpaper.
— Liu
