competitor-analysis

Overview of competitor-analysis skill

What competitor-analysis is for

The competitor-analysis skill helps an AI agent produce structured SEO and GEO competitor research instead of a loose “tell me what rivals are doing” summary. It is built for people asking practical questions such as “why do they outrank us,” “which keywords are they winning,” “what content formats work for them,” and “where is our easiest gap to close.”

Who should install competitor-analysis

Best fit readers are SEO leads, content strategists, founders, growth marketers, and researchers who need a repeatable competitor-analysis workflow for Competitive Analysis. It is especially useful if you want outputs that can become an internal report, battlecard, or content gap plan without rebuilding the format every time.

What makes this competitor-analysis skill different

The main differentiator is structure. This is not just a brainstorming prompt. The repository includes:

• analysis templates for competitor profiling, keyword gaps, content analysis, and opportunity scoring
• a battlecard template for downstream sales or strategy use
• an example report that shows the expected depth and formatting
• positioning frameworks that help turn raw findings into messaging and differentiation decisions

That makes the skill more install-worthy than a generic prompt when you need consistent deliverables, not just ideas.

What job it actually gets done

In practice, competitor-analysis is for turning scattered evidence into a decision-ready answer:

• who the real search competitors are
• why they perform better
• which gaps matter most
• what you should copy, avoid, or counter-position against
• how to convert findings into a usable report or battlecard

Limits to know before you adopt

This skill does not magically fetch reliable external SEO data by itself. The repo explicitly treats networked SEO tools as optional. If you do not provide SERP snapshots, tool exports, URLs, or notes, the model will still produce a framework, but confidence drops fast. Install it if you want a better analysis process; do not install it expecting proprietary data access out of the box.

How to Use competitor-analysis skill

competitor-analysis install context

The repository does not expose a skill-specific install command in SKILL.md, but the usual marketplace add flow for the parent repo is:

npx skills add aaron-he-zhu/seo-geo-claude-skills --skill competitor-analysis

Compatibility in the source points to Claude Code ≥1.0, skills.sh marketplace, ClawHub marketplace, and the Vercel Labs skills ecosystem. No system packages are required. Optional MCP/network access only matters if you want live SEO tool integrations.

Read these files first

For the fastest understanding, open files in this order:

1. research/competitor-analysis/SKILL.md
2. research/competitor-analysis/references/analysis-templates.md
3. research/competitor-analysis/references/example-report.md
4. research/competitor-analysis/references/battlecard-template.md
5. research/competitor-analysis/references/positioning-frameworks.md

That path gives you trigger logic, expected workflow, final output shape, and the strategy layer many users skip on first read.

What inputs the skill needs to work well

A good competitor-analysis usage pattern includes four input types:

• your site or product
• the competitors or SERP rivals to analyze
• the market or keyword scope
• evidence sources

Useful evidence can include:

• target keywords or topic clusters
• competitor URLs or specific pages
• exports from Ahrefs, Semrush, Search Console, Similarweb, etc.
• notes on pricing, audience, and positioning
• examples of AI citations or answer-engine mentions
• your own content inventory for gap comparison

If you provide only competitor names, expect broad findings. If you provide URLs, keywords, and a comparison baseline, the output becomes much more actionable.

Turn a rough request into a strong prompt

Weak prompt:

Analyze our competitors.

Stronger prompt:

Use the competitor-analysis skill to compare our domain example.com against competitor1.com, competitor2.com, and competitor3.com for the topics “project management software” and “kanban tools.” Focus on keyword gaps, highest-traffic pages, content patterns, backlinkable assets, pricing/positioning differences, and likely reasons they outrank us. Use a confidence label when data is inferred. End with the top 5 opportunities we can execute in the next 90 days.

Why this is better:

• defines comparison set
• narrows topic scope
• specifies output categories
• asks for prioritization
• distinguishes evidence from inference

Best workflow for competitor-analysis usage

A practical workflow is:

1. define the business question
2. identify true search competitors, not just commercial rivals
3. gather evidence by keyword set, pages, and market segment
4. run analysis with one output format in mind
5. convert findings into a gap list, action plan, or battlecard

This skill is strongest when used after step 3, not before. The more grounded the inputs, the less generic the report.

Choose the right output format

The references support at least three useful deliverables:

• full competitor report for strategy reviews
• battlecard for sales, marketing, or executive use
• positioning analysis for messaging changes

If your real need is “help the content team decide what to publish next,” ask for a concise opportunity matrix rather than a long narrative. If your real need is “help sales handle objections,” use the battlecard template instead.

How to analyze search competitors vs business competitors

One adoption blocker is mixing categories. In competitor-analysis for Competitive Analysis, separate:

• business competitors: sell similar products
• search competitors: rank for the same queries
• GEO competitors: get cited in AI answers and summaries

These sets overlap only sometimes. A company may be weak commercially but still dominate informational search. Tell the skill which type matters most for the task.

Practical prompt pattern that improves output quality

Use a prompt structure like:

• objective
• audience
• comparison set
• evidence provided
• required sections
• decision criteria
• confidence handling

Example:

Use competitor-analysis to produce a decision memo for our SEO lead. Compare us with three search competitors using the attached keyword export and page list. Include: keyword gaps, content patterns, linkable assets, positioning differences, and AI citation likelihood. Score opportunities by impact, ease, and time-to-value. Mark assumptions clearly.

This reduces filler and forces the model toward a useful artifact.

What the references are really good for

The reference files are not just examples; they solve common execution problems:

• analysis-templates.md prevents missing sections and weak comparisons
• example-report.md shows what “good enough” depth looks like
• battlecard-template.md helps move from research to operational use
• positioning-frameworks.md helps explain not only what competitors do, but how they frame themselves

If you skip the references, you will likely underuse the skill.

Common constraints and tradeoffs

This competitor-analysis guide is most reliable for pattern finding and structured synthesis, not exact SEO measurement. Watch for:

• stale or estimated traffic numbers
• overconfident backlink claims without source data
• shallow conclusions when competitor samples are too small
• strategy drift when the prompt mixes SEO, product, and brand goals without priority

Ask the model to separate confirmed data, reasonable inference, and open questions.

competitor-analysis skill FAQ

Is competitor-analysis better than a normal prompt?

Usually yes, if you need repeatability. A normal prompt can generate observations, but the competitor-analysis skill gives you a more consistent workflow and report shape. The gain is strongest when multiple teammates will reuse the method or when outputs need to be compared over time.

Is this competitor-analysis skill beginner-friendly?

Yes, but only if you already know what you want to compare. Beginners can use the templates well, yet may struggle with input selection. The easiest starting point is one domain, three competitors, one topic cluster, and a request for keyword gaps plus content patterns.

Do I need paid SEO tools for competitor-analysis install and usage?

No for installation, often yes for best results. The skill itself does not require system packages, and the repo notes external integrations as optional. But strong competitor-analysis usage usually depends on data from SERP observation, Search Console, or third-party SEO platforms.

When should I not use competitor-analysis?

Skip it if your need is:

• a quick one-off summary with no structured output
• primary user research or customer interviews
• deep financial or legal competitor due diligence
• exact market sizing without trustworthy external data

This skill is best for search visibility, content strategy, positioning, and competitive benchmarking.

Can it help with GEO and AI citation analysis too?

Yes. The source description explicitly includes AI citations and GEO alongside keywords, backlinks, and traffic share. That makes it more relevant than a pure SEO template if you care about answer-engine visibility, citation patterns, or why certain brands get mentioned in AI responses.

How many competitors should I compare at once?

Usually 3 to 5. Fewer than 3 can hide patterns. More than 5 often produces shallow summaries unless your evidence is already well organized. For a first pass, start narrow and expand only after you see where the real gap is.

How to Improve competitor-analysis skill

Give the skill evidence, not just goals

The biggest quality jump in competitor-analysis comes from providing source material. Better inputs include:

• a keyword gap export
• top pages by traffic for each competitor
• screenshots or notes from AI answer results
• pricing pages
• category and audience notes

Without that, the model leans on generic market logic.

Separate observations from recommendations

A common failure mode is jumping straight from “they rank well” to “we should copy them.” Ask the skill to produce:

1. observed patterns
2. likely causal explanations
3. recommended actions
4. confidence level

That structure makes the final advice more trustworthy.

Narrow the comparison scope

If your first output feels vague, your scope is probably too broad. Improve the competitor-analysis guide in practice by shrinking one dimension:

• one topic cluster instead of the whole market
• one funnel stage instead of all content
• one country or language instead of global
• one competitor segment instead of every rival

Tighter scope usually beats longer prompts.

Ask for prioritization, not just findings

Many reports fail because they list everything equally. Request scoring such as:

• impact
• implementation effort
• evidence strength
• time-to-value

This converts analysis into a roadmap rather than a reading exercise.

Use the battlecard and positioning frameworks after the first pass

Do not stop at the initial report. The best way to improve competitor-analysis for Competitive Analysis is to turn findings into the next artifact:

• battlecard for internal enablement
• positioning statement revision
• content brief backlog
• differentiation map

The repository’s supporting files make this transition easier than a generic SEO analysis prompt.

Correct common failure modes in follow-up prompts

If the first run is weak, use targeted follow-ups:

• “Separate commercial rivals from search rivals.”
• “Show only keywords where they rank top 10 and we do not rank top 30.”
• “Which competitor strengths are hard to copy vs easy to emulate?”
• “Convert this into a 90-day action plan for content and linkable assets.”
• “Flag claims that require external validation.”

These follow-ups are more effective than asking for “more detail.”

Improve trust with explicit confidence labels

Because some inputs may be inferred, ask the skill to tag sections as high, medium, or low confidence. This is especially important for estimated traffic, backlink quality, publishing cadence, and AI citation conclusions. Confidence labels make the output easier to share internally without overstating certainty.

Build a reusable competitor-analysis prompt template

If your team will use this repeatedly, save a house prompt with:

• standard sections
• scoring rules
• confidence language
• approved data sources
• expected output length
• preferred final artifact

That is where the competitor-analysis skill becomes operationally better than ad hoc prompting: same workflow, better comparisons, less reinvention.