Ever feel overwhelmed by SEO problems? Low traffic, high bounce rates, zero rankings… the list feels endless. You know something’s broken, but finding the exact cause is like looking for a needle in a haystack.
Wasting hours on manual audits is outdated. You need a smarter system.
This prompt is that system. It transforms vague complaints into a precise, prioritized diagnosis and action plan in seconds. Stop troubleshooting; start solving.
📋 The Prompt
Follow this structured diagnosis:
1. **Immediate Scan:** List the top 3 most likely technical, on-page, and off-page root causes for this specific problem. Prioritize by probability and impact.
2. **Deep Dive Evidence:** For the #1 most likely cause, provide a specific checklist of data points to verify (e.g., exact Google Search Console queries, specific crawl errors, competing page features). Tell me exactly where to look and what to compare.
3. **Action Blueprint:** Give me a step-by-step correction plan for the #1 cause. Each step must be a concrete, executable task (e.g., "Run this specific Screaming Frog configuration," "Rewrite the meta title using this keyword pattern," "Build one backlink from a site in this niche").
4. **Preventive Protocol:** Based on this diagnosis, what 2-3 monitoring alerts or habitual checks should I set up to catch this problem earlier next time?
How It Works
Why does this prompt work so well? It forces structured thinking instead of random guessing. Let’s break it down.
The first instruction, ‘Act as an elite SEO diagnostician,’ establishes a high-expertise role. This primes the AI to think like a consultant, not just a text generator.
The core magic is in the four-step structure. It moves from hypothesis to verification to action to prevention. This mirrors how the best SEOs actually work. The ‘Immediate Scan’ generates intelligent hunches. ‘Deep Dive Evidence’ stops you from acting on a hunch without proof—it demands data validation. This is critical.
Most AI outputs are vague. By asking for a ‘specific checklist of data points,’ you get direct commands. Instead of ‘check for errors,’ it will say ‘Go to GSC > Coverage > Excluded and filter for ‘Crawled – currently not indexed’ pages from last month.’ That’s actionable.
The ‘Action Blueprint’ is where theory meets practice. Each step is a discrete task. This turns an overwhelming problem into a simple to-do list. Finally, the ‘Preventive Protocol’ applies the learning forward, turning a one-time fix into a system. It’s not just a band-aid; it’s building resilience.
This prompt is your productivity multiplier. It systematizes the diagnostic phase, freeing your brain for creative strategy. For a broader framework on how to construct prompts like this, see our complete AI SEO Mastery guide.
Pro Tips & Variations
Advanced Tweaks for Maximum Impact:
Feed it more context for a sharper diagnosis. Instead of just ‘traffic dropped,’ try: ‘Traffic from branded search terms dropped 40% in 3 weeks, but non-branded is stable. We migrated to a new CMS 4 weeks ago.’ The AI will instantly focus on redirect chains and indexation post-migration.
Common Mistake: Being too vague in the [INSERT PROBLEM] section. ‘My SEO is bad’ will get you a generic textbook answer. ‘My product pages have a 90% bounce rate and zero conversions’ triggers a focused analysis on intent mismatch and page speed.
Use the output as a hypothesis, not gospel. The AI’s #1 probable cause is an educated guess. Your job is to use the ‘Deep Dive Evidence’ checklist to confirm or reject it. This prompt is a co-pilot, not an autopilot.
For ongoing strategic issues, like identifying new opportunities, pair this diagnostic prompt with our guide on using an AI Prompt for SEO Trend Analysis. One solves fires; the other finds gold mines.
Frequently Asked Questions
Is this just for big, catastrophic SEO problems?
Not at all. It’s brilliantly effective for small, nagging issues too. ‘Why did this one page’s ranking drop 5 spots?’ or ‘Why is my click-through rate from search results so low?’ The structured approach ensures you don’t overlook simple fixes.
What's the best way to phrase the core problem for the AI?
Be specific and include metrics if you have them. Use the formula: “[Metric/Observation] has [changed] over [timeframe] for [specific pages/traffic type].” Example: “Organic conversions have stalled for 2 months, but traffic is up 10% for blog posts.”
Can I use this with any AI model?
It works best with advanced models like GPT-4, Claude 3, or Gemini Advanced, as they have stronger reasoning capabilities. Simpler models might not follow the complex, multi-step logic as precisely.
The AI gave me a technical action I don't understand. What now?
Perfect. That’s the prompt doing its job—surfacing knowledge gaps. Take that specific term or task (e.g., ‘audit Core Web Vitals LCP element’) and ask the AI or search to explain *that one thing*. It creates a personalized learning path.
How is this different from just asking 'How do I fix my SEO?'
Night and day. A vague question gets a generic blog post list. This prompt forces a differential diagnosis. It considers multiple causes, ranks them, and provides a verification method before any action. It’s the difference between getting a dictionary and getting a custom-built repair manual.
