Competitive Analysis Skill

Procedural guidance for turning competitor products into structured intelligence that feeds PRDs, design briefs, and visual direction decisions.

When to Use

Running /landscape [initiative] [competitors...]
Scoping a new initiative and need market context
Entering Define phase and the PRD needs competitive evidence
A customer mentions a competitor in research or Slack signals
Evaluating "build vs. match" decisions for a feature area

Inputs

Initiative name
List of competitor product names and/or URLs
Optional: specific feature area or flow to focus on (e.g., "onboarding", "dashboard", "agent configuration")

Required Context

Load before analysis:

pm-workspace-docs/company-context/product-vision.md - Know what AskElephant IS and IS NOT
pm-workspace-docs/company-context/strategic-guardrails.md - Red flags for copying vs. differentiating
pm-workspace-docs/company-context/personas.md - Who we build for (evaluate competitors through our persona lens)
Initiative's research.md if it exists - Customer-voiced competitive signals

Competitor Tiering

Classify every competitor before analysis:

Tier	Definition	Analysis Depth	Examples
Direct	Same product category, same target buyer	Full profile + UX deep dive	Momentum, Reevo, Gong
Indirect	Different product, solves same job-to-be-done	Profile + feature comparison	Day.ai, Clari, Chorus
Adjacent	Different category, shares a design/automation pattern	Pattern extraction only	Zapier, Make, Tray.ai, Relay.app

Methodology

1. Define Analysis Dimensions

Tie dimensions to the specific initiative, not a generic feature checklist. Ask:

What user problem does this initiative solve?
What flows/screens are most relevant to compare?
What decision are we trying to make? (build, match, leapfrog, ignore)

Example dimensions for an "agent automation" initiative:

Agent configuration UX (how complex is setup?)
Trigger/action model (visual builder vs. code vs. natural language)
Error handling and observability (can users debug failed automations?)
Integration depth (shallow webhook vs. deep CRM field mapping)
Trust signals (how does the product communicate what the AI will do?)

2. Gather Intelligence

Sources to check for each competitor:

Product website - Positioning, pricing, target persona messaging
Product screenshots/demos - Via web search for "[product] dashboard screenshot", "[product] UI demo"
G2 screenshot galleries - g2.com/products/[product]/screenshots (often 5-15 real product screenshots)
Interactive demos - Search for "[product] interactive demo" or "[product] product tour" (Navattic, Storylane, Reprise embeds)
YouTube walkthroughs - Search "[product] demo walkthrough [year]" for recent UI screenshots
G2/Capterra reviews - What real users praise and complain about
Job postings - What they're building next (hiring for = investing in)
Changelog/blog - Recent feature launches and roadmap signals (often include UI screenshots of new features)
Customer mentions - Search initiative's research.md and pm-workspace-docs/signals/ for competitor names
Social/community - Reddit, Twitter/X, LinkedIn posts comparing tools
Help docs/knowledge base - Often contain detailed UI screenshots showing actual product screens

Use web search extensively. Do NOT make up competitor features -- cite sources. Prioritize capturing real UI screenshots from these sources using the browser-use subagent.

3. Capture Real Competitor UI Screenshots

For Direct and Indirect competitors, go beyond feature lists. Capture actual competitor product UIs first, then generate mockups only as a supplement.

Step 3a: Screenshot Capture (Primary)

Use the browser-use subagent to navigate to competitor product pages and take real screenshots:

Search for screenshot sources: Web search for "[product] dashboard", "[product] UI", "[product] demo", "[product] product tour". Look for:
- Product marketing pages with embedded screenshots
- Demo/tour pages (many SaaS products have interactive demos or Navattic/Storylane embeds)
- G2 screenshot galleries (g2.com/products/[product]/screenshots)
- YouTube demo walkthrough thumbnails
- Product documentation with UI screenshots
- Blog posts announcing features (often include UI previews)
Use browser-use subagent to visit the best URLs and take screenshots:
- Navigate to the page
- Scroll to the relevant UI section
- Take a screenshot
- Save to assets/competitive/ with the naming convention below
Naming convention for real screenshots:
- [competitor]-[screen]-screenshot.png (e.g., gainsight-health-dashboard-screenshot.png)
- [competitor]-[flow]-screenshot-[N].png for multi-step flows (e.g., vitally-setup-wizard-screenshot-1.png)
Annotate each screenshot in the competitive landscape doc with:
- Source URL
- Date captured
- What it shows (screen name, key patterns visible)

Step 3b: AI-Generated Mockups (Supplement)

When real screenshots aren't available (login-gated, no public demos, or need to illustrate a pattern comparison across competitors):

Use image generation to create representative comparison mockups
Naming convention for generated mockups: [pattern]-comparison-mockup.png or [competitor]-[pattern]-mockup.png
Always label generated images clearly in the doc: "AI-generated representation based on public documentation and marketing materials"

General Guidelines

Always prefer real screenshots over generated mockups -- they're more credible and show actual UX details
Capture the flow, not just individual screens (onboarding sequence, configuration wizard, dashboard layout)
Note interaction patterns: drag-and-drop, form-based, natural language, wizard-style
Save all images to pm-workspace-docs/initiatives/active/[name]/assets/competitive/
For each initiative, aim for at least 2-3 real competitor screenshots per Direct competitor

4. Build Feature Matrix

Rows = capabilities relevant to THIS initiative (not a generic checklist) Columns = competitors + AskElephant (current) + AskElephant (proposed)

Use these ratings:

Leading - Best-in-class implementation
Parity - Meets market expectation
Basic - Functional but limited
Missing - Not available
N/A - Not applicable to this product

5. Map Differentiation

Categorize each capability:

Category	Meaning	Strategic Response
Table Stakes	Everyone has it, customers expect it	Must match, don't over-invest
Parity Zone	Most competitors have it, some don't	Match if evidence demands it
Opportunity Gap	Few or no competitors serve this well	Potential differentiator -- validate with users
AskElephant Unique	Only we have this (or could)	Protect and amplify

6. Extract Design Vocabulary

Identify the language and patterns competitors use:

Adopt: Patterns that are becoming user expectations (e.g., "visual workflow builder" for automations)
Reject: Patterns that conflict with our values (e.g., surveillance dashboards, complexity-as-power)
Leapfrog: Patterns we can do better because of our unique position (meeting context, CRM knowledge)

Required Output Sections

The competitive-landscape.md document MUST include:

1. TL;DR

2-3 sentence market position summary. Where does AskElephant sit? What's the primary differentiation opportunity?

2. Competitor Profiles

Per competitor (Direct and Indirect tiers):

Product: Name + URL
Tier: Direct / Indirect / Adjacent
Positioning: How they describe themselves (use their actual tagline)
Target Persona: Who they sell to
Key Strengths: 2-3 things they do well
Key Weaknesses: 2-3 gaps or complaints (cite G2/review sources)
Relevance to This Initiative: Why this competitor matters for this specific work

3. Feature Matrix

Table with initiative-specific capabilities as rows, competitors as columns.

4. UX Pattern Inventory

For each key flow relevant to the initiative:

How does Competitor A handle it?
How does Competitor B handle it?
What's the emerging "best practice" pattern?
Where are users frustrated? (from reviews)
Screenshot/mockup references

5. Visual Reference Gallery

Organized by flow or screen type, with clear labeling:

Real Competitor Screenshots (captured from product pages, demos, G2):

Link to each image with source URL, date captured, and what it demonstrates
These are primary references for design decisions

AI-Generated Comparison Mockups (created when real screenshots unavailable):

Clearly labeled as generated representations
Used to illustrate pattern comparisons across competitors or when products are login-gated

6. Differentiation Map

Table categorizing each capability as Table Stakes / Parity Zone / Opportunity Gap / AskElephant Unique.

7. Design Vocabulary

Patterns to Adopt: List with rationale
Patterns to Reject: List with rationale (cite anti-vision when relevant)
Patterns to Leapfrog: Where our unique context enables better solutions

8. Strategic Recommendations

What to match (table stakes we're missing)
What to leapfrog (opportunity gaps we can own)
What to ignore (competitor features that don't serve our personas)
Risks if we don't act

Save Locations

Analysis document: pm-workspace-docs/initiatives/active/[name]/competitive-landscape.md
Real competitor screenshots: pm-workspace-docs/initiatives/active/[name]/assets/competitive/[competitor]-[screen]-screenshot.png
Generated comparison mockups: pm-workspace-docs/initiatives/active/[name]/assets/competitive/[pattern]-comparison-mockup.png
Competitive signals from other sources: append to existing competitive-landscape.md or create if missing

Image Naming Convention

Type	Pattern	Example
Real screenshot	`[competitor]-[screen]-screenshot.png`	`gainsight-health-dashboard-screenshot.png`
Multi-step flow	`[competitor]-[flow]-screenshot-[N].png`	`vitally-setup-wizard-screenshot-1.png`
Generated mockup	`[pattern]-comparison-mockup.png`	`health-score-patterns-comparison-mockup.png`
Generated per-competitor	`[competitor]-[pattern]-mockup.png`	`churnzero-alert-ux-mockup.png`

Integration Points

Research Analyst: When analyzing transcripts, competitive mentions feed into the Competitor Profiles section
PRD Writer: The Feature Matrix and Differentiation Map provide "Competitive Evidence" for the PRD
Design Brief: Design Vocabulary section feeds directly into the brief's "References" and "Patterns to Adopt/Reject"
Visual Design: The UX Pattern Inventory and Visual Reference Gallery inform mockup generation directions

Anti-Patterns

Copying competitor features without understanding WHY they built them
Generic feature comparison that isn't tied to the specific initiative
Listing features without evaluating UX quality and user satisfaction
Treating "competitor has it" as sufficient evidence to build (needs user evidence too)
Ignoring adjacent competitors that share relevant design patterns
Analysis paralysis -- the goal is actionable intelligence, not an exhaustive report

When to Refresh

Before entering a new initiative phase (Discovery -> Define -> Build)
When a competitor launches a significant update in the same space
When customer research surfaces new competitor mentions
Quarterly for ongoing initiatives

competitive-analysis