Audit my site →

Blog

Writing for Search Without Writing for Robots

The writing habits that help search engines are the same ones that help hurried readers, and robotic SEO writing fails both.

You can usually tell within one sentence when a page was written for a search engine instead of a person. “Looking for the best affordable dog grooming services in your area? When it comes to dog grooming services, choosing the right dog grooming service matters.” No human writes like that to another human. It’s the prose equivalent of talking loudly and slowly at someone you assume doesn’t speak your language — and the assumption is about a decade out of date.

The robotic style is a fossil. It made a kind of sense once, when search engines matched keywords more literally and the way to rank for a phrase was to repeat it. Search engines have spent years getting better at understanding language precisely so they can stop rewarding that, because pages that read like keyword inventories are bad answers, and serving bad answers is the one thing a search engine can’t afford. So the writers still grinding out keyword-density prose are optimizing for an algorithm that no longer exists, and modern algorithms are tuned to prefer the thing they’re avoiding: plain writing that says something.

But it would be too easy to conclude “just write well and ignore search.” That’s the other fashionable error. Search engines understand language better than they used to, but they are not mind readers, and there’s a legitimate discipline of writing that meets them halfway. The interesting thing is what that discipline turns out to be. Almost every practice in it is also just a practice of considerate writing.

Take the foundational one: say what the page is about, plainly, early, in the places built for saying it. The title tag. The H1. The first paragraph. When my scanner reports “No H1 heading found” or “Missing meta description,” people sometimes hear it as robot-appeasement — tags to fill in for the algorithm’s benefit. But think about what those elements actually are. The title is what shows in the browser tab and on the search results page; it’s the label a stranger judges before clicking. The H1 is the headline a visitor reads to confirm they’re in the right place. The meta description is your two-sentence pitch under the blue link. Every one of these is a message to humans that the search engine happens to read over their shoulder. Writing them well isn’t writing for robots. It’s writing the parts of the page that hurried readers rely on most.

The same doubling holds all the way down. Using the words your readers actually use — “sore throat” rather than “pharyngeal discomfort” — is keyword research when a marketer does it and plain empathy when a writer does it; it’s one act. Answering the question in the first paragraph instead of after eight paragraphs of throat-clearing helps the skimmer and the snippet-extractor identically. Covering a topic thoroughly enough that the reader doesn’t need a second search is exactly what the algorithm’s satisfaction signals are trying to detect. In each case there’s no fork in the road where the robot’s path and the reader’s path diverge. The robot is a proxy for the reader, and it keeps getting better at being one.

So where does search-aware writing genuinely differ from ordinary good writing? In one place, mostly: the awareness that a page arrives without context. An essay in a book sits between chapters that frame it. A page arriving from a search result sits between nothing and nothing. The searcher lands mid-site, having read none of your other pages, often knowing nothing about you. Search-aware writing accounts for this by making each page self-sufficient — restating the essential context, naming the topic explicitly rather than through in-jokes or cleverness. A print magazine can title a piece “The Long Goodbye” and let the table of contents do the explaining. A web page with that title, about, say, ending a business partnership, is invisible: no one searching for advice on dissolving a partnership types anything resembling those words. Clever titles are a luxury of publications with captive readers. On the open web, clarity in the title isn’t a stylistic retreat; it’s how the piece and its readers find each other at all.

The synthesis, then, is not a compromise between two audiences. It’s a sequencing. Write the page for one specific person with one specific question, in the words they’d use, answering quickly and thoroughly. Then check the machine-facing surfaces — title, description, headings — and make sure they state honestly what the page delivers. That checking step is mechanical, which is why I built a tool to do it; the writing step is not, which is why no tool can. The pages that rank durably are the ones where both steps happened: something worth reading, labeled truthfully. Robotic writing fails because it does only the labeling, at great length, with nothing underneath. Precious writing fails because it does only the substance and hides it behind labels no one would search for. The web rewards writers who do the obvious thing that somehow remains rare: say something useful, and say what you said.

Test your own website

Type your website address and press Test. In about ten seconds you see a grade for your homepage. Then pick what to order. The first option is free.

More articles

← All posts