Content Briefs and the On-Page Optimizer

Two tools for winning content

StretchSEM's content tools sit at the heart of the SEM-to-content workflow:

• Content brief — analyzes what already ranks for a keyword and tells you what to write to compete.
• On-page optimizer — scores an existing page against a target keyword and lists exactly what to fix.

Both cost 1 credit per run, and both are designed to hand off to StretchBAS so you can generate the fix on-brand, not just read a recommendation.

Content briefs

For a keyword, the content-brief engine fetches the SERP, then reads the top-ranking pages (their H2/H3 structure and length) and aggregates them into a brief.

1. Open Content brief and enter a keyword.
2. Run. StretchSEM analyzes the top ~6 results and reads up to 5 of them.
3. The brief returns:
- Difficulty and demand for the keyword (with bands).
- Competitors — who currently ranks (position, domain, title).
- Target word count — the median length of the ranking pages, rounded to the nearest 100.
- Subtopics — heading terms that appear across two or more ranking pages (the themes you must cover to compete), each with a page count.
- Example headings — real multi-word headings from the top pages to model your structure on.
- Related keywords — autocomplete-derived terms to weave in.

The brief tells you — and StretchBAS — what to write to actually rank, instead of generating content blind.

On-page optimizer

The optimizer fetches a single page and scores how well it's optimized for one target keyword, returning a 0–100 score and prioritized recommendations.

1. Open On-page and enter the page URL and a target keyword.
2. Run. StretchSEM fetches the page and analyzes its on-page signals.
3. You get a score plus a signals breakdown and a prioritized fix list.

How the score is built

The score is weighted toward the highest-impact signals:

• Keyword in the title — 25 points
• Keyword in an H1 — 20 points
• Keyword in the meta description — 15 points
• Keyword in the URL slug — 10 points
• Keyword in an H2/H3 subheading — 10 points
• Keyword present in body copy — 10 points
• Healthy keyword density (roughly 0.3%–2.5%) — 10 points

The signals panel also reports word count, exact body occurrences, density, and title/meta lengths.

Recommendations

Each recommendation is specific and prioritized high / medium / low — for example "Add 'industrial valves' to the page title — highest on-page signal," "Work the keyword into the H1," "Add a meta description (~120–155 chars) featuring the keyword," or "Keyword density is high (3.4%) — risk of keyword-stuffing; ease off." It also flags thin content (under 300 words) and over-long titles (which truncate in SERPs).

A realistic example

A content lead targets email marketing automation. The content brief shows a target length of 1,900 words, competitors like established SaaS blogs, and subtopics such as segmentation, workflows, triggers, and deliverability appearing across multiple ranking pages. She drafts an outline around those subtopics. After publishing, she runs the on-page optimizer on her new URL for the same keyword: it scores 72, flags that the keyword is missing from the meta description (medium) and from any H2 (medium). She adds both, re-checks, and the score climbs to 92.

Tips

• Build the content brief before writing so you cover the subtopics competitors treat as table stakes.
• Hit the target word count as a floor, not a ceiling — depth wins when it's genuinely useful.
• Run the on-page optimizer after publishing, apply the high-priority fixes first, then re-check.
• Aim for keyword density in the 0.3%–2.5% range — enough to signal relevance without stuffing.
• Send briefs and on-page fixes to StretchBAS to generate on-brand copy fast.

Troubleshooting

The on-page optimizer can't load my page. It uses a guarded crawler. Confirm the URL is public (not behind a login or firewall), include the full address, and check the page returns a normal 200. Pages that block automated fetches or time out return a fetch error (and refund the credit).

The content brief returned no pages analyzed. The SERP couldn't be fetched for that keyword at that moment, or the ranking pages blocked reads. Wait and retry; empty runs are refunded.

My density looks high but the page reads fine. Density counts exact keyword occurrences relative to word count. Prefer natural variants and synonyms over repeating the exact phrase.