Diagnose an underperforming page with real search data, competitor scraping, SERP feature analysis, and a prioritized list of specific fixes ready to implement.
Give Juma a URL and what you've noticed. You get back a page-level diagnosis with query-by-query performance data, a competitor comparison showing who overtook you and why, and rewritten titles and metas ready to implement.
The team gets the full picture for one page: every query driving impressions, how CTR breaks down by query, and where position has moved over time. The actual page gets checked against what's live, not what the CMS says: the real title tag, meta description, heading structure, and content depth as Google sees them.
From there, the top 3 ranking competitors for the page's main keywords get scraped and compared side by side: what they have that this page does not. A comparison table the client's page is missing, FAQ schema earning People Also Ask placements, a fresher publish date, a more comprehensive treatment of the topic. The SERP itself gets mapped too, because a featured snippet or AI Overview answering the query changes what kind of fix matters. Every recommendation ties back to specific data: "this query has 13,000 impressions at position 16.4, your title doesn't include the primary keyword, and the top competitor has a comparison table you lack."
The natural instinct when a page underperforms is to rewrite the title or add more content. But the fix depends on the cause. A page stuck at position 16 has a content depth or authority problem, not a title tag problem. A page at position 4 with 2% CTR when the benchmark is 7% has a SERP presentation problem. A page losing ground to a specific competitor needs the specific elements that competitor added. The diagnosis separates these causes so the team works on the right fix, not the obvious one.
The diagnosis identified the title tag or meta description as a problem. This prompt produces specific rewrites with character counts, based on the queries actually driving impressions to the page.
Based on the diagnosis, rewrite the title tag and meta description. Give me 2-3 options for each, with character counts and reasoning for why each version should perform better for the queries driving the most impressions.
The diagnosis showed a competitor gaining ground. This prompt does a deep side-by-side comparison: content structure, word count, schema, backlinks, and the specific content elements the competitor has that the client's page is missing.
The competitor at [URL] is now ranking above our client for the main keyword. Do a full side-by-side: content depth, structure, schema markup, backlink profile, freshness signals. Tell us exactly what they have that we don't.
A Juma Project is a shared space where the team stores everything Juma needs to know about a client. Create one project per client, add context as you go, and Juma will use what's relevant every time the team runs a flow. If the project already exists from other work, just add the SEO-specific items below.
Target Keywords
The keywords the client cares about most. When diagnosing a page, Juma checks performance against these priority terms first, so the team knows whether the drop affects keywords that matter to the business.
Competitor URLs
The 3-5 domains the client competes with in search. Juma uses these to pull the right competitive data instead of guessing which sites to compare against.
Past Audit Results
Previous SEO audits or page-level performance snapshots. Juma compares against the baseline to show what changed: "This page was at position 4 in the last audit and is now at position 11."
Add a short description in the project's info field that tells Juma what each file contains and when to use it. For example: "Target Keywords: priority terms from Q1 planning, focus on these first. Competitor URLs: main organic competitors, always compare against these. Past Audit: March 2026 audit, use as performance baseline."
