# AI Search Optimization

Guide for understanding and optimizing visibility in AI search — Google AI Overviews, ChatGPT, Claude, Perplexity, and other AI assistants. This is sometimes called GEO (Generative Engine Optimization) or AEO (AI Engine Optimization).

## When to Use

- The user asks about AI search visibility or brand mentions in ChatGPT / Perplexity.
- The user wants to know if AI assistants recommend their product.
- The user wants to understand Google AI Overviews for their target keywords.
- The user asks about GEO or AEO strategy.

## Why AI Search Matters

AI assistants are becoming a significant discovery channel. Users ask ChatGPT, Perplexity, and Claude for product recommendations, how-to answers, and comparisons. Being mentioned (or missing) from these responses directly impacts brand visibility.

Key differences from traditional SEO:

- **No click-through**: AI responses answer directly. The goal is being cited, not ranking a blue link.
- **Source credibility matters**: AI models cite authoritative, well-structured content.
- **Different demand patterns**: some queries have high AI volume but low Google volume, and vice versa.
- **Multiple models, different results**: ChatGPT, Claude, and Perplexity may cite completely different sources for the same query.

## Workflow

### Step 1: Measure current AI visibility

Before optimizing, establish a baseline.

1. Run `hyperseo_mentions_track` with the user's brand name and 2 – 3 key competitor brands, using a query their customers would ask.
   - Example: `query="best project management tools"`, `brands=["Asana", "Monday.com", "ClickUp"]`.
2. This queries multiple LLMs (GPT, Claude, Perplexity) with web search enabled and checks which brands appear in responses.
3. Note: which models mention the brand, what context they provide, and which sources they cite.
4. **Note:** Perplexity results may intermittently return a routing error. If so, rely on the GPT and Claude results — brand mention data from two models is still actionable. Do not retry in a loop.

### Step 2: Check AI search demand

Understand how much query volume exists in AI channels.

1. Run `hyperseo_ai_search_volume_get` for the user's target keywords.
2. Compare with `hyperseo_search_volume_get` for the same keywords.
3. Look for keywords where AI volume is growing fast — these represent emerging opportunities.

**Interpreting AI volume:**

- AI search volume measures monthly queries to ChatGPT, Claude, Perplexity, and similar tools.
- Growing AI volume with stable Google volume suggests users are shifting to AI for that query type.
- Keywords with high AI volume and low Google difficulty are prime targets.

### Step 3: Analyze Google AI Overviews

Google's AI Overviews appear above traditional search results for many queries.

1. Run `hyperseo_ai_overviews_get` for the user's most important keywords.
2. Review: which sources does Google AI cite?
3. Note the domains and content types that get cited.
4. This reveals what kind of content earns AI citations.

### Step 4: Competitive AI visibility

Compare brand visibility across AI models.

1. Run `hyperseo_mentions_track` for multiple relevant queries with the user's brand and top competitors.
2. Build a visibility matrix: which brands are mentioned by which models for which queries.
3. Identify gaps: queries where competitors are mentioned but the user is not.

## Strategies for Earning AI Citations

Based on what sources AI models tend to cite:

**Structured, authoritative content**

- Clear headings that match question patterns.
- Definitive statements and data points (not vague qualifiers).
- Lists, comparisons, and structured data that AI can extract.

**Be the primary source**

- Original research, benchmarks, and proprietary data get cited more than aggregated content.
- First-party case studies and results are preferred over generic advice.

**Answer questions directly**

- AI models look for content that directly answers the query in the first 1 – 2 paragraphs.
- Use the question as a heading, then answer immediately.

**Build domain authority**

- AI models weight established, authoritative domains.
- Backlink profile and domain rank influence citation likelihood.
- This connects back to traditional SEO fundamentals.

## Output Format

**AI visibility report:**

1. **Brand mention status** — table showing which AI models mention the brand for key queries.
2. **AI vs traditional volume** — side-by-side comparison for target keywords.
3. **Google AI Overview sources** — which domains get cited for target keywords.
4. **Competitor comparison** — who has better AI visibility and why.
5. **Recommendations** — specific content improvements to earn AI citations.

```
| Query | GPT Mentions Brand? | Claude Mentions? | Perplexity Mentions? | Competitor Mentioned? |
|-------|---------------------|-----------------|---------------------|----------------------|
```

## Common Mistakes to Avoid

- **Treating AI search like Google SEO**: keyword stuffing and backlink tactics don't directly influence AI citations. Content quality and structure matter more.
- **Checking only one model**: different AI models cite different sources. Always check multiple.
- **Ignoring the gap**: if competitors are consistently mentioned and you're not, the content or authority gap needs addressing before any optimization.
- **Expecting quick changes**: AI model knowledge updates on different schedules. Improvements in content may take weeks to reflect in AI responses.

# Competitor Analysis

Step-by-step guide for identifying, analyzing, and benchmarking SEO competitors using the HyperSEO tools.

## When to Use

- The user asks "who are my competitors?".
- The user wants to understand why a competitor ranks and they don't.
- The user needs a competitive landscape overview before planning content.
- The user asks to compare two or more domains.

## Workflow

### Step 1: Identify competitors

Don't assume who the competitors are — discover them from data.

1. If you have the user's domain, run `hyperseo_competitor_domains_search` to find competitors based on shared organic keywords. This is the best starting point.
2. Alternatively, if you only have target keywords, run `hyperseo_competitors_search` to find domains that rank for similar terms.
3. The top 5 – 7 results are the real SEO competitors (these may differ from business competitors).

### Step 2: Domain comparison

For each competitor (and the user's own domain).

1. Run `hyperseo_traffic_get_bulk` with all domains at once to get a quick traffic comparison across all competitors in a single call.
2. Run `hyperseo_domain_overview_get` on each domain for detailed metrics (authority, backlinks, referring domains).
3. Present as a comparison matrix.

**Comparison matrix format:**

```
| Domain | Domain Rank | Organic Keywords | Est. Traffic | Backlinks | Referring Domains |
|--------|------------|-----------------|-------------|-----------|-------------------|
```

### Step 3: Keyword gap analysis

Find what competitors rank for that the user doesn't.

1. Run `hyperseo_domain_intersections_search` between the user's domain and each top competitor to find keywords where both appear in SERPs — this reveals direct keyword battlegrounds.
2. Run `hyperseo_domain_keywords_get` on the top 2 – 3 competitors (limit 50 each) to find keywords the user is missing entirely.
3. Compare: which high-value keywords do competitors have that the user is missing?
4. Cross-check the gaps with `hyperseo_keyword_difficulty_get` to find achievable targets.

### Step 4: Backlink comparison

Understand the link-building landscape.

1. Run `hyperseo_backlinks_history_get` on each competitor.
2. Compare link velocity: who is gaining or losing backlinks faster?
3. A competitor with strong backlink growth will be harder to overtake without a similar link strategy.

### Step 5: SERP position analysis

For the user's most important keywords.

1. Run `hyperseo_serp_results_get` to see exactly who holds the top positions.
2. Note the content types that rank (articles, product pages, videos, Reddit threads).
3. Identify patterns: do certain domains dominate across multiple keywords?

## Interpreting Results

**Domain rank**

- 0 – 20: very low authority, new or small site.
- 20 – 40: growing authority, can compete for low-difficulty keywords.
- 40 – 60: established site, can compete for medium-difficulty keywords.
- 60+: high authority, competitive for most keywords.

**Keyword gap opportunities**

Focus on gaps where:

- The competitor ranks in positions 5 – 20 (beatable).
- The keyword difficulty is under 50.
- Search volume is above 500.
- Your domain has some topical relevance.

**Backlink velocity**

If a competitor gains 100+ new referring domains per month and the user gains 5, content alone won't close the gap. Recommend link-building or pivot to less competitive terms.

## Output Format

Deliver a competitor analysis report with:

1. **Competitor overview table** — side-by-side domain metrics.
2. **Keyword gap list** — keywords competitors rank for that the user doesn't, sorted by opportunity (volume / difficulty).
3. **Backlink comparison** — link growth trends for each domain.
4. **SERP dominance** — which competitors appear most across target keywords.
5. **Actionable recommendations** — specific keywords to target, content to create, and where the user has a realistic chance of competing.

## Common Mistakes to Avoid

- **Comparing against impossible competitors**: a new blog should not benchmark against Wikipedia or Reddit. Focus on competitors of similar scale.
- **Ignoring content format**: a competitor may rank with video while you plan a blog post. Check the SERP format.
- **Treating all keyword gaps equally**: a gap is only valuable if the keyword has volume, is achievable at your authority level, and matches your business.
- **Skipping backlink context**: a domain with 10× more backlinks isn't beatable on the same keywords without a link strategy.

# Content SEO Planning

Step-by-step guide for building a data-informed content strategy using the HyperSEO tools. Focus on creating content that ranks by targeting validated keywords with the right content format.

## When to Use

- The user asks "what content should I create?".
- The user wants a content calendar with keyword targets.
- The user needs to plan blog posts, landing pages, or pillar content.
- The user wants to build topical authority in a specific area.

## Workflow

### Step 1: Define the topic territory

Understand what the user's content should cover.

1. Ask for their core topics or business areas (2 – 5 seed themes).
2. Run `hyperseo_keyword_ideas_generate` for each seed theme (limit 30 – 50).
3. This gives a broad view of the keyword landscape around their topics.

### Step 2: Validate demand and intent

Not every idea has search demand. Validate.

1. Run `hyperseo_search_volume_get` on the most relevant keyword candidates.
2. Run `hyperseo_keyword_difficulty_get` on the same set.
3. Run `hyperseo_intents_search` to classify intent at scale — this determines the right content format for each keyword.
4. Also run `hyperseo_ai_search_volume_get` to check AI channel demand.
5. Discard keywords with no meaningful volume (< 100 / month) unless strategically important.

### Step 3: Analyze the SERP

For each high-priority keyword, understand what Google rewards.

1. Run `hyperseo_serp_results_get` for the top 10 – 15 keyword targets.
2. Note the content format that ranks:
   - Long-form guides (2000+ words)?
   - Listicles ("10 best…")?
   - Product pages?
   - Video results?
   - Reddit / forum threads?
3. Match your content format recommendation to what actually wins.

### Step 4: Check AI Overviews

For informational keywords, check if Google generates an AI Overview.

1. Run `hyperseo_ai_overviews_get` for key terms.
2. Review which sources get cited in the AI overview.
3. Content that earns AI citations gets visibility even above traditional position 1.

### Step 5: Build topic clusters

Organize keywords into topic clusters, not flat lists.

**Pillar page**: broad topic with high volume and higher difficulty.

- Targets the main keyword.
- Comprehensive, long-form content.
- Links to all supporting content.

**Supporting pages**: specific subtopics with lower difficulty.

- Target long-tail variations.
- Each links back to the pillar page.
- Builds topical authority that helps the pillar rank.

**Example cluster:**

```
Pillar: "email marketing guide" (vol: 8,100, diff: 65)
  ├── "email marketing best practices" (vol: 1,900, diff: 35)
  ├── "email subject line tips"        (vol: 2,400, diff: 28)
  ├── "email marketing automation"     (vol: 3,600, diff: 42)
  └── "email list building strategies" (vol: 1,200, diff: 30)
```

Start by publishing the supporting pages. Once 3 – 5 are live and interlinked, publish the pillar.

### Step 6: Prioritize and schedule

Create a content calendar ordered by priority.

**Priority factors:**

1. Quick wins first (difficulty < 30, volume > 500).
2. Supporting content before pillar content.
3. Commercial-intent pages before purely informational ones (if conversions matter).
4. Topics where you already have some authority or existing content.

## Content Brief Template

For each piece of content, provide:

```
Target Keyword: [primary keyword]
Secondary Keywords: [2-3 related terms to include naturally]
Search Volume: [monthly]
Difficulty: [0-100]
Search Intent: [from hyperseo_intents_search: informational / commercial / transactional / navigational]
Recommended Format: [guide / listicle / comparison / landing page]
SERP Insight: [what currently ranks - based on hyperseo_serp_results_get]
Cluster Role: [pillar / supporting]
Links To: [pillar page or related supporting pages]
```

## Output Format

Deliver the content plan as:

1. **Topic cluster map** — visual grouping of pillar + supporting content.
2. **Content calendar table:**

```
| Priority | Keyword | Volume | Difficulty | Format | Cluster Role | Notes |
|----------|---------|--------|-----------|--------|-------------|-------|
```

3. **Content briefs** — one per piece, using the template above.
4. **Quick wins highlighted** — top 3 – 5 pieces to publish first.

## Common Mistakes to Avoid

- **Writing without keyword data**: every content piece should target a validated keyword with real search demand.
- **Starting with the hardest keyword**: build authority with easier supporting content first, then tackle the pillar.
- **Ignoring SERP format**: if the SERP rewards listicles, don't write a 3,000-word essay.
- **Creating orphan content**: every page must link to and from related content. Isolated pages struggle to rank.
- **Focusing only on Google**: check AI search volume too. A keyword with moderate Google volume but high AI volume may be worth targeting for emerging visibility.

# Keyword Research

Step-by-step guide for finding and prioritizing keyword opportunities using the HyperSEO tools exposed by the Hyper MCP.

## When to Use

- The user wants to find keywords for a new page, product, or campaign.
- The user asks "what keywords should I target?".
- The user has seed terms and wants to expand them.
- The user needs to validate keyword ideas with real data.

## Workflow

### Step 1: Seed expansion

Start with the user's seed terms and expand them.

1. Run `hyperseo_keyword_ideas_generate` with their seed keywords (set `limit` to 30–50 for a broad view).
2. Review the ideas for relevance — discard anything off-topic.
3. If the user has a domain, also run `hyperseo_site_keywords_search` to discover terms their site naturally aligns with.

### Step 2: Volume and commercial value

Take the most promising candidates and check real demand.

1. Run `hyperseo_search_volume_get` on the shortlisted keywords (batch up to 20 at a time).
2. Note search volume, CPC, and competition level.
3. High CPC often signals commercial intent — these keywords convert.

### Step 3: Difficulty and intent assessment

Filter by what's actually achievable and classify intent.

1. Run `hyperseo_keyword_difficulty_get` on the shortlist.
2. Run `hyperseo_intents_search` on the same keywords to classify intent (informational, navigational, commercial, transactional).
3. Score each keyword against the priority framework below.
4. Intent classification helps you decide the right content format before you start writing.

### Step 4: AI search demand (optional but recommended)

Check if these keywords are also being queried in AI assistants.

1. Run `hyperseo_ai_search_volume_get` on the top candidates.
2. Keywords with high AI volume may represent emerging demand not yet saturated in traditional search.

### Step 5: SERP validation

For top-priority keywords, check what currently ranks.

1. Run `hyperseo_serp_results_get` for your highest-priority keywords.
2. Look at: who ranks, what content format they use, whether there are featured snippets or AI overviews.
3. This tells you what you need to create to compete.

## Priority Framework

Classify every keyword into a tier based on difficulty and volume:

| Tier | Difficulty | Volume | Action |
| --- | --- | --- | --- |
| Quick Win | 0 – 30 | 500+ | Target immediately. These are low-hanging fruit. |
| Medium Play | 30 – 50 | 1,000+ | Target with strong content. Requires effort but achievable. |
| Long-Term Bet | 50 – 70 | 5,000+ | Build toward over months. Needs topical authority first. |
| Aspirational | 70+ | Any | Don't target directly. Build surrounding content first. |
| Skip | Any | < 100 | Not worth the effort unless strategically important. |

## Keyword Clustering

Don't present keywords as a flat list. Group them into clusters by intent.

**Informational cluster** — "what is X", "how to X", "X guide".

- Content format: blog posts, guides, tutorials.
- Goal: build authority, earn backlinks.

**Commercial investigation** — "best X", "X vs Y", "X reviews".

- Content format: comparison pages, listicles, reviews.
- Goal: capture mid-funnel traffic.

**Transactional** — "buy X", "X pricing", "X free trial".

- Content format: landing pages, product pages.
- Goal: convert directly.

**Navigational** — "X login", "X website".

- Usually not targetable unless it's your brand.
- Note: track these with `hyperseo_mentions_track` for brand visibility.

## Output Format

Use `hyperseo_intents_search` to auto-fill the Intent column. Present the final keyword research as a prioritized table:

```
| Keyword | Volume | Difficulty | CPC | AI Volume | Intent | Cluster | Priority |
|---------|--------|-----------|-----|-----------|--------|---------|----------|
```

Include a summary with:

- Total addressable search volume across all recommended keywords.
- Number of quick wins vs long-term plays.
- Recommended first 3 – 5 keywords to target.
- Suggested topic clusters.

## Common Mistakes to Avoid

- **Chasing volume alone**: a 50,000-volume keyword at difficulty 90 is useless if the domain has no authority.
- **Ignoring intent**: ranking for informational keywords won't drive sales if the user needs conversions.
- **Flat keyword lists**: always cluster and prioritize. A list of 100 unranked keywords is not actionable.
- **Skipping SERP validation**: the actual SERP tells you what Google rewards. Don't recommend a blog post if the top 10 results are all product pages.

# SEO Site Audit

Step-by-step guide for auditing a domain's organic search health using the HyperSEO tools. Covers domain authority, current rankings, keyword opportunities, backlink health, page speed, and AI visibility.

## When to Use

- The user asks "how is my site doing in search?".
- The user wants a health check on their domain's SEO performance.
- The user is evaluating a domain before acquisition or investment.
- The user wants to understand their current organic position before planning improvements.

## Workflow

### Step 1: Domain overview

Get a high-level snapshot of domain health.

1. Ask for the exact domain (e.g., `example.com` — don't guess).
2. Run `hyperseo_domain_overview_get` on the domain.
3. Review: organic keywords count, estimated traffic, backlinks, referring domains, domain rank.

**Quick health indicators:**

| Metric | Weak | Developing | Established | Strong |
| --- | --- | --- | --- | --- |
| Domain Rank | 0 – 10 | 10 – 30 | 30 – 50 | 50+ |
| Organic Keywords | < 100 | 100 – 1,000 | 1,000 – 10,000 | 10,000+ |
| Referring Domains | < 50 | 50 – 500 | 500 – 5,000 | 5,000+ |

### Step 1b: Real performance data (conditional — requires Google Search Console integration)

If the user has Google Search Console connected in Hyper:

1. Call `google_search_console_sites_list()` to confirm the domain is verified and get the exact `site_url` as registered in GSC.
2. Call `google_search_console_performance_get` for the search analytics. It reads the Search Console API directly, so there is no cache to warm and no sync step:
   ```
   google_search_console_performance_get(
       site_url="<site_url from step 1>",
       start_date="2025-01-01",
       end_date="2025-01-31",
       dimensions=["query"],
       row_limit=20
   )
   ```
   Pass `dimensions=["page"]` for landing-page performance, or `["query", "page"]` for both. Dates are `YYYY-MM-DD`.

   Returns actual clicks, impressions, CTR, and average position from GSC — ground-truth data vs. DataForSEO estimates.

   **If the call returns no rows:** GSC has no data for that range. Widen the date range and confirm `site_url` matches the verified property exactly. Do not substitute DataForSEO estimates and present them as GSC data.
3. Use this alongside `hyperseo_domain_overview_get`. Real GSC clicks will differ from ETV estimates — present both where available.

If GSC is not connected, skip this step and rely on DataForSEO estimates throughout.

### Step 2: Current rankings and trends

Understand what the domain already ranks for and how it's trending.

1. Run `hyperseo_domain_keywords_get` with `limit=50` to see top-ranked keywords.
2. Run `hyperseo_rank_history_get` to see organic traffic and keyword count trends over time — is the domain growing or declining?
3. Sort current keywords by traffic to find the pages driving the most organic visits.
4. Identify:
   - **Strengths**: keywords in positions 1 – 5 (defend these).
   - **Low-hanging fruit**: keywords in positions 6 – 20 (improve these with content updates).
   - **Declining**: keywords losing position (investigate and fix).
   - **Trend**: compare recent months to past months — a declining trend requires urgent action.

### Step 3: Keyword opportunities

Find untapped potential.

1. Run `hyperseo_site_keywords_search` to discover keywords the domain could target but isn't.
2. Cross-check the best opportunities with `hyperseo_keyword_difficulty_get`.
3. Focus on keywords where:
   - The domain has topical relevance.
   - Difficulty is achievable for the domain's authority level.
   - Volume justifies the effort.

### Step 4: Backlink health

Assess the domain's link profile.

1. Run `hyperseo_backlinks_history_get` for the domain.
2. Look for:
   - **Growth trend**: are backlinks growing, stable, or declining?
   - **Referring domain ratio**: healthy profiles have a diverse set of referring domains, not just many links from few domains.
   - **Recent losses**: a sudden drop in referring domains can signal lost partnerships or content removal.

### Step 5: Competitive context

No audit is complete without context.

1. Run `hyperseo_competitor_domains_search` to find competitors based on shared organic keywords.
2. Run `hyperseo_traffic_get_bulk` on the user's domain + top 3 – 5 competitors to compare traffic levels quickly.
3. Run `hyperseo_domain_overview_get` on the top 3 competitors for detailed metrics.
4. Compare: how does the domain stack up on authority, traffic, and backlinks?
5. This frames the audit findings — "you rank for 500 keywords" means different things if competitors rank for 500 vs 50,000.

### Step 6: Page speed check

Assess Core Web Vitals.

1. Run `hyperseo_pagespeed_get` on the domain's homepage and 2 – 3 key pages (both mobile and desktop).
2. Key metrics to evaluate:
   - **Performance score**: 90+ is good, 50 – 89 needs improvement, below 50 is poor.
   - **LCP (Largest Contentful Paint)**: should be under 2.5 s.
   - **TBT (Total Blocking Time)**: should be under 200 ms.
   - **CLS (Cumulative Layout Shift)**: should be under 0.1.
3. Poor page speed directly hurts rankings, especially on mobile.

### Step 7: AI visibility check

Evaluate presence in AI search channels.

1. Run `hyperseo_ai_search_volume_get` for the domain's top keywords to understand AI channel demand.
2. Run `hyperseo_mentions_track` with the brand name on 2 – 3 relevant queries.
3. Note whether AI assistants mention the brand, and how they describe it.

## Audit Report Structure

Present findings as a structured report.

**1. Executive summary**

- Overall health assessment (one paragraph).
- Top 3 strengths.
- Top 3 areas for improvement.

**2. Domain metrics**

```
| Metric | Value | Assessment |
|--------|-------|-----------|
| Domain Rank | [X] | [Weak/Developing/Established/Strong] |
| Organic Keywords | [X] | ... |
| Estimated Traffic | [X] | ... |
| Backlinks | [X] | ... |
| Referring Domains | [X] | ... |
```

**3. Top performing keywords**

```
| Keyword | Position | Volume | Traffic | URL |
|---------|---------|--------|---------|-----|
```

**4. Quick win opportunities**

Keywords in positions 6 – 20 that could move to page 1 with content improvements.

**5. Untapped keyword opportunities**

Keywords the domain should target but currently doesn't rank for.

**6. Historical trends**

Monthly organic traffic and keyword counts from `hyperseo_rank_history_get`. Is the domain growing or declining?

**7. Backlink health**

Trend chart summary, growth / decline assessment, referring domain diversity.

**8. Page speed**

Core Web Vitals for key pages (mobile and desktop). Flag any metrics outside acceptable ranges.

**9. Competitive benchmark**

Side-by-side comparison with top 3 competitors using `hyperseo_traffic_get_bulk` and `hyperseo_domain_overview_get`.

**10. AI visibility status**

Whether the brand appears in AI assistant responses for key queries.

**11. Prioritized recommendations**

Ordered list of specific actions, starting with highest-impact, lowest-effort items.

## Common Mistakes to Avoid

- **Auditing without context**: raw numbers mean nothing without competitive benchmarks. Always compare against competitors.
- **Focusing only on what's broken**: highlight strengths too. The user needs to know what to protect, not just what to fix.
- **Presenting data without interpretation**: don't dump raw tool output. Every metric needs a "so what" explanation.
- **Ignoring content quality**: metrics tell you what's happening, not why. If rankings are low despite good authority, the content itself may be the issue — note this for the user to investigate.
- **One-time mindset**: an audit is a snapshot. Recommend re-auditing quarterly to track progress.