Writing content that satisfies search intent
A page can use a keyword exactly, repeatedly, and correctly, and still fail to rank — if the type of content doesn't match what a searcher actually wanted when they typed that phrase. Search intent is that underlying want, and matching it matters more than any on-page keyword placement.
The four types of search intent
- Informational — "how does dns resolution work" — wants an explanation, not a product or a specific site.
- Navigational — "github login" — wants a specific, known destination, not a comparison or explanation.
- Commercial investigation — "best python web frameworks 2026" — researching before a decision, wants a comparison.
- Transactional — "buy django hosting" — ready to act now, wants a place to actually do the thing.
The exact same subject — Django hosting — produces three completely different intents depending on phrasing: "what is django hosting" (informational), "best django hosting" (commercial investigation), "django hosting pricing" (transactional, close to a purchase decision). Content built for the wrong one of these, even if excellent on its own terms, satisfies the wrong searcher.
Reading intent directly from current search results
The fastest, most reliable way to determine a keyword's actual intent isn't guessing — it's searching the phrase and looking at what's already ranking. If the top results for a phrase are all step-by-step tutorials, that's a strong, direct signal the intent is informational and instructional; competing with a product landing page for that same phrase, however well-written, is targeting the wrong format entirely, regardless of content quality.
What "satisfying intent" looks like on a real page
A tutorial titled "Making HTTP requests in Python without the footguns" (a real page on this site) opens with a direct, one-sentence explanation of what the requests library actually is, before any setup instructions — matching informational intent immediately, rather than making a searcher scroll past unrelated framing to reach the actual answer. The heading structure from part 3 reinforces this: each <h2> answers one specific sub-question a reader with this intent would actually have next.
Keyword density is not the goal
Old, discredited advice: repeat the exact keyword phrase 2-3% of total word count.This was a real, common SEO tactic years ago, and it's since become actively counterproductive — modern ranking systems use natural language understanding sophisticated enough to recognize topical relevance from context, synonyms, and related terms, not literal phrase-matching frequency. Content that repeats an exact phrase unnaturally reads worse to an actual human and gains nothing from a ranking system built specifically to reward genuine topical depth over mechanical repetition.
Satisfying intent fully, not partially
A search for "python requests timeout" that leads to a page briefly mentioning timeouts in passing, buried in a much broader article, technically contains the keyword but doesn't fully satisfy the intent behind that specific search. A dedicated section — or a dedicated page — that treats the timeout behavior as the actual subject, with a working code example, closes that gap completely instead of partially.
Writing one broad, general page and expecting it to rank for every related keyword variation, including ones with a genuinely different intent behind them. A page optimized to inform ("what is X") rarely also satisfies a comparison search ("X vs Y") or a transactional one ("buy X") — those usually deserve their own dedicated content rather than being folded into one page trying to serve every intent at once.
The on-page content fundamentals are covered. Next: technical SEO — starting with whether a search engine can even reach a page in the first place.