Claude Code for Competitor Teardowns: Systematic Analysis in an Hour

Your competitor just shipped a major feature. Your CEO pinged the Slack channel: "What does this mean for us?" Your VP of Sales forwarded a deal you lost with one line: "They went with [Competitor]. Find out why." The clock is ticking and you're tabbing between their pricing page, a G2 search, and a half-finished Google Doc from last quarter.

I've trained over 100 founders, PMs, and product marketers on Claude Code. The ones who build a repeatable competitor teardown system stop scrambling. They respond to these moments in an hour, not a week.

What a Competitor Teardown Actually Covers

Most people think "competitor analysis" means glancing at a pricing page and skimming a blog post. A real teardown dissects five layers:

• Pricing: plan names, price points, feature gating, free tier limits, enterprise positioning. How they package value tells you who they're targeting.
• Positioning: homepage headline, hero copy, above-the-fold messaging, tagline, navigation labels. When these change, their strategy shifted.
• Feature set: what they ship, how they categorize it, what gets gated behind higher tiers, what's in beta. Their changelog is their roadmap in retrospect.
• Messaging: how they talk about themselves on landing pages, in ads, on social media. The language they use reveals who they think their buyer is.
• Reviews: what real customers say on G2, Capterra, Reddit, and Twitter. The complaints are more valuable than the praise because they expose cracks you can exploit.

Each layer tells you something different. Pricing tells you strategy. Positioning tells you audience. Features tell you investment. Messaging tells you narrative. Reviews tell you truth.

Setting Up Your Teardown Project

Create a folder that Claude Code can work inside. Structure matters here because you'll reuse this every quarter (or whenever a competitor makes noise):

competitor-teardowns/
├── CLAUDE.md
├── inputs/
│   ├── our-product.md
│   ├── competitors/
│   │   ├── competitor-a/
│   │   │   ├── pricing.md
│   │   │   ├── positioning.md
│   │   │   ├── features.md
│   │   │   └── reviews.md
│   │   └── competitor-b/
│   │       ├── pricing.md
│   │       ├── positioning.md
│   │       ├── features.md
│   │       └── reviews.md
├── templates/
│   └── teardown-template.md
└── output/
    └── (generated teardowns go here)

The CLAUDE.md file grounds every teardown in your product's context. Without it, Claude generates generic analysis. With it, every insight is framed against your specific strengths and weaknesses. The skills guide explains why this file is the most important 5 minutes you'll spend.

The CLAUDE.md That Drives Everything

Here's what your project-level context file should look like:

# Competitor Teardown System

## Our Product
- Name: Acme CRM
- Category: CRM for agencies under 50 people
- Key strengths: dead-simple onboarding, built-in invoicing, flat pricing
- Key weaknesses: no native phone dialer, limited reporting
- Pricing: $49/user/mo flat (no tiers)
- ICP: Agency owners and ops leads, 10-50 employees

## Teardown Format
Each teardown should include:
1. Executive summary (3 sentences: who they are, who they target, key threat level)
2. Pricing breakdown (plan names, prices, what's gated, comparison to us)
3. Positioning analysis (their headline, tagline, who they think their buyer is)
4. Feature comparison (their strengths vs ours, gaps on both sides)
5. Messaging analysis (language patterns, value props, emotional hooks)
6. Review intelligence (top 5 complaints from G2/Capterra, top 5 praises)
7. Threat assessment (low/medium/high with 2-3 sentence justification)
8. Recommended response (what we should do about this competitor)

## Tone
- Write for a founder who has 5 minutes to read before a board meeting
- Be direct and opinionated, take clear stances
- Quantify everything possible
- Flag anything that directly threatens our ICP or pricing

That file turns Claude from a generic AI into your competitive analyst. Every teardown it produces will reference your strengths, target your ICP, and flag threats specific to your business.

Gathering the Data

The teardown is only as good as the inputs. Here's how to populate each competitor folder quickly:

• Pricing: screenshot or copy their pricing page. Include plan names, prices, feature lists per tier, and any fine print about limits or overages.
• Positioning: grab their homepage headline, subheadline, hero section, and the first three navigation items. These reveal priority.
• Features: copy their features page or changelog. If they have a comparison page ("Us vs. Competitors"), grab that too. It shows what they think matters.
• Reviews: pull the most recent 20 to 30 reviews from G2 or Capterra. Copy the full text, not just star ratings. The words people use are where the signal lives.

This data collection takes 15 to 20 minutes per competitor. You can also ask Claude Code to help with the automated scraping approach if you want to streamline this step.

The Prompt That Generates a Teardown

With your inputs populated, open Claude Code in the competitor-teardowns/ directory and run this:

Read all files in inputs/competitors/competitor-a/.
Read our product context from inputs/our-product.md.
Read the teardown template from templates/teardown-template.md.

Generate a full competitor teardown for Competitor A following the
format in CLAUDE.md. Be specific and opinionated. Don't hedge.

For the review intelligence section, identify patterns across
multiple reviews, not just individual complaints. Group them
into themes.

For the threat assessment, focus on where they directly overlap
with our ICP (agencies under 50 people).

Save to output/competitor-a-teardown.md.

Claude reads every input file, cross-references their positioning against yours, identifies review patterns, and produces a structured teardown. First draft lands in about 90 seconds.

What the Output Looks Like

Here's a condensed version of what Claude generates:

# Competitor Teardown: HubSpot CRM

## Executive Summary
HubSpot targets SMBs with a freemium model that hooks users on free
tools then upsells aggressively into $800+/mo tiers. For our ICP
(agencies under 50), they're a medium threat: powerful but over-
engineered and expensive at scale. We win on simplicity and pricing.

## Pricing Breakdown
| Plan        | HubSpot         | Acme CRM       |
|-------------|-----------------|----------------|
| Free        | Yes (limited)   | No             |
| Starter     | $20/user/mo     | $49/user/mo    |
| Pro         | $100/user/mo    | $49/user/mo    |
| Enterprise  | $150/user/mo    | $49/user/mo    |

Key insight: HubSpot looks cheaper at Starter but costs 3x more
at Pro level, which is where agencies actually need to be for
reporting and automation features.

## Review Intelligence
Top complaint themes from 30 G2 reviews:
1. "Price jumps between tiers are brutal" (mentioned 14 times)
2. "Setup took our team 3 months" (mentioned 9 times)
3. "Features we need are always in the next tier up" (mentioned 8 times)
4. "Reporting is powerful but takes a HubSpot expert to configure"
5. "Contact limits on lower tiers force premature upgrades"

## Threat Assessment: MEDIUM
They compete for our ICP but lose on complexity and total cost of
ownership. Agency owners want something working by Friday, not a
3-month implementation. Our flat pricing eliminates the upgrade
anxiety that drives 47% of their negative reviews.

## Recommended Response
- Lead with "no hidden tiers" messaging in head-to-head deals
- Create a landing page: "Acme vs HubSpot for Agencies"
- Arm sales with the G2 complaint data as landmine questions

That's a board-ready teardown. Your CEO can scan it in 5 minutes and know exactly where you stand. Your sales team can pull landmine questions directly from it.

Comparing Multiple Competitors Side by Side

One teardown is useful. A comparative view across your full competitive set is a weapon. After generating individual teardowns, run this prompt:

Read all teardown files in the output/ folder.

Create a competitive landscape summary that includes:
1. Side-by-side pricing table across all competitors and us
2. Feature matrix: rows are key features, columns are competitors,
   cells are yes/no/partial with notes
3. Positioning map: where each competitor sits on two axes
   (simplicity vs power, and SMB vs enterprise)
4. Threat ranking: ordered list from highest to lowest threat
   with one-line justification for each
5. Top 3 strategic recommendations based on the full competitive picture

Save to output/competitive-landscape-summary.md.

The landscape summary is the document you bring to quarterly planning. It shows where you're winning, where you're exposed, and what to do about it.

Making It Repeatable

The real power is that this system runs every quarter with almost zero setup. When a competitor ships something big, you update their input files (10 minutes), run the same prompt (90 seconds), and have a fresh teardown ready. Compare that to the typical process: schedule a meeting, assign research to a PMM, wait a week for a first draft, revise it twice, and present it three weeks after the moment passed.

I recommend a quarterly cadence for full teardowns and an on-demand cadence for breaking moves. Set a calendar reminder. Spend one afternoon per quarter refreshing inputs and regenerating teardowns for your top 3 to 5 competitors.

Feeding Teardowns Into Other Workflows

Teardowns aren't a destination. They're raw material. Once you have them, you can feed the output directly into other Claude Code projects:

• Battle cards: pull the pricing comparison, review intelligence, and landmine questions straight into sales-ready cards
• Positioning updates: use the messaging analysis to sharpen your own homepage copy and ad language
• Product roadmap input: the feature comparison flags gaps your customers might care about
• Board decks: the competitive landscape summary drops right into a quarterly board presentation

One afternoon of work produces artifacts that serve your entire company for months.

What Founders Get Wrong

The biggest mistake: running a teardown once and never updating it. Competitive intelligence decays fast. A teardown from six months ago is worse than useless because it gives you false confidence. Build the system to be repeatable, not just impressive.

The second mistake: analyzing competitors you don't actually lose deals to. Focus on the 2 to 3 competitors that show up in your pipeline, not the 15 that show up on G2 category pages. Your sales team knows who they are. Ask them.

Start Your First Teardown

Pick your most dangerous competitor. Spend 15 minutes gathering their pricing, positioning, features, and reviews into markdown files. Set up the folder structure above. Run the prompt. You'll have a structured teardown before lunch.

New to Claude Code? The step-by-step tutorial gets you from zero to running in 10 minutes.

Want to go deeper? ClaudeFluent is our premium training program where I teach founders, PMs, and operators how to build real systems with Claude Code. Competitor teardowns, battle cards, dashboards, and dozens of other workflows. Join us for the next cohort.