SEO content prompt builder
Generating content that ranks is less about clever wording and more about structure: the right keyword in the right places, related terms for topical depth, a logical heading hierarchy, and a meta description that earns the click. This builder turns those requirements into a single, reusable prompt so an LLM produces a publish-ready, optimized draft instead of a generic blob of text you have to rework from scratch.
How it works
You provide the primary keyword, a list of LSI (semantically related) keywords, the content type, target word count, and heading structure. The tool composes a prompt that tells the model exactly where the primary keyword must appear — the title, an opening paragraph, and at least one heading — while keeping its density natural rather than stuffed. It instructs the model to weave the LSI terms in for topical depth, follow your chosen heading layout, and finish with a meta description under 160 characters. The output is a structured draft with title, headings, body, and meta tag, ready for your editing pass.
What LSI keywords do in the prompt
LSI (latent semantic indexing) keywords are terms conceptually related to your primary keyword — the words that naturally appear in well-rounded content on that topic. For a primary keyword like “noise-cancelling headphones,” LSI terms might include “active noise cancellation,” “Bluetooth range,” “battery life,” and “audio quality.”
Including these terms in the prompt serves two goals. First, it tells the model to cover the full topic rather than producing a narrow piece that happens to mention the primary keyword repeatedly. Second, it signals topical authority to search engines, which increasingly assess content by the range of related concepts it addresses, not just keyword frequency.
The prompt instruction wraps these terms as secondary signals — the model is told to include them naturally, not to force them in. Unnatural LSI placement reads exactly as awkward as keyword stuffing to both humans and crawlers.
Content types and how the prompt adapts
The content type setting changes the structural instructions the model receives:
- Blog post asks for an engaging opening that addresses search intent, H2 sections covering the topic’s main aspects, and a conclusion with a next step.
- Product page asks for benefit-led copy, feature specifics, trust signals, and a call to action above the fold.
- How-to guide asks for numbered steps, prerequisites, and common pitfalls — the structure readers expect when searching for procedural instructions.
- Comparison page asks for a structured table plus explanatory prose, addressing the specific comparison intent that readers bring when they type “X vs Y.”
Matching the content type to the search intent behind your keyword is the single most important structural decision. A how-to keyword served by product-page copy will bounce; a comparison keyword served by a listicle will fail to satisfy the reader’s actual question.
Tips and notes
The generated draft is a foundation, not the finished article. Search engines increasingly reward demonstrated experience and original insight, so add your own data, examples, and point of view on top of the structure — that is what separates a ranking page from generic AI text. Keep your primary keyword to one focused phrase per piece; trying to rank one article for five keywords usually ranks it for none. Use the deeper heading structure for longer how-to and guide content where readers and search engines both benefit from clear sub-sections, and always fact-check anything the model states as fact before publishing.