B2B Competitor Teardown — AI Battlecard Generator

Give it a competitor's domain. Get back a finished, one-page sales battlecard — how they position themselves, their tech stack, their pricing, their recent moves, their strengths and weaknesses, and exactly how to win against them. It orchestrates four research Actors and lets an AI assemble the teardown, so you skip the manual competitive-intelligence grind and go straight to the talk track.

You bring the competitor. It hands back a battlecard your reps can paste into the CRM.

---

The flow

competitor domain                    4 research organs                    AI battlecard
  (e.g. notion.so)      →      firmographics · tech stack ·        →      positioning · strengths ·
                               pricing · recent news                      weaknesses · how to win ·
                                                                          objections · landmines

1. Paste competitor domains into domains (one per line).
2. Tell it what you sell in yourOffering (recommended) so "how to win", objection handling and landmine questions are framed to beat them on your behalf.
3. Run it. Each competitor comes back as a structured record and a ready-to-paste battlecardMarkdown doc.

---

What you get per competitor

1. oneLineSummary — the whole competitor in one punchy sentence.
2. positioning — how they position themselves (their own message).
3. whatTheyDo / targetCustomers — the factual core, grounded in firmographic data.
4. techStackSummary — what they run (from the tech organ), or null if unknown.
5. pricingSummary — their published pricing and plans (from the pricing organ), or null.
6. strengths / weaknesses — concrete, evidence-based bullets.
7. recentMoves — recent headlines about them (launches, funding, pivots), [] if none.
8. differentiators — what sets them apart in their market.
9. howToWin — concrete talking points to beat them, framed vs your offering.
10. objectionHandling — likely pro-competitor objections, each with a crisp rebuttal.
11. landminesToSet — sharp discovery questions that expose their weaknesses.
12. battlecardMarkdown — the whole thing assembled as a clean one-page markdown doc, ready for a CRM, Slack or your sales wiki.

Every statement is grounded only in what the research organs actually returned — the AI never invents pricing, customers or news.

---

Three ways people use it

• Sales battlecards / win rate — arm every AE with a "competitor teardown" and a tested talk track for "how to beat
• Product & strategy competitive analysis — track how rivals position themselves, what they charge, which tech they run and what they just shipped. Market positioning on one page, refreshed on demand.
• Competitive intelligence at scale — feed a whole list of competitors and get a battlecard for each in one run, indexed by domain, ready to drop into your CRM or enablement hub.

---

How the orchestration works

This is a super-actor: it runs four specialist Actors as "organs" (best-effort — if one returns nothing, the teardown degrades gracefully and still ships), then an OpenAI model synthesizes the battlecard from whatever came back.

Organ	Actor (via Actor.call)	What it contributes
Firmographics	noniner/ai-company-enrichment	What they do, target customers, industry, size
Technographics	noniner/b2b-tech-stack-detector	Detected technology stack
Pricing	noniner/b2b-pricing-extractor	Published pricing and plans
Recent moves	noniner/google-news-scraper	Recent headlines about the competitor

Each organ runs once across all domains and its results are indexed by domain, so a 25-competitor run is just a handful of sub-runs, not 100.

---

Input

Field	Type	Notes
domains	array	Required. Competitor domains, website URLs or emails (domain is extracted). One per line.
yourOffering	string	Recommended. What YOU sell — frames howToWin, objectionHandling and landminesToSet to beat each competitor.
maxResults	integer	Max competitors to tear down. Default 25, max 500. One charge per competitor.
maxConcurrency	integer	Competitors synthesized in parallel in the AI stage. Default 3, max 8.
model	string	OpenAI model. Default gpt-4o-mini.
useEnrichment	boolean	Use the firmographic organ. Default true.
useTechStack	boolean	Use the technographic organ. Default true.
usePricing	boolean	Use the pricing organ. Default true.
useNews	boolean	Use the recent-news organ. Default true.

Minimal input example

{
  "domains": ["notion.so", "asana.com"],
  "yourOffering": "We sell a project management tool built for engineering teams."
}

---

Output example

{
  "domain": "notion.so",
  "companyName": "Notion",
  "oneLineSummary": "An all-in-one connected workspace for notes, docs, wikis and lightweight project management, aimed at teams that want flexibility over structure.",
  "positioning": "Positions itself as the single flexible workspace that replaces a stack of separate tools, emphasizing customizable building blocks over rigid, purpose-built software.",
  "targetCustomers": "Startups, product and knowledge-work teams, and individuals who want a highly customizable workspace.",
  "whatTheyDo": "Provides a block-based workspace combining docs, wikis, databases and basic project views.",
  "techStackSummary": "Runs a modern web stack with its own real-time collaboration layer.",
  "pricingSummary": null,
  "strengths": [
    "Extremely flexible, customizable building blocks",
    "Strong brand and bottom-up adoption among startups",
    "One tool spanning docs, wikis and light project tracking"
  ],
  "weaknesses": [
    "Flexibility means teams must build their own structure",
    "Not a dedicated engineering project tool (no sprints, boards tuned for dev)",
    "Can get slow and messy at scale"
  ],
  "recentMoves": [],
  "differentiators": ["All-in-one workspace rather than a single-purpose PM tool"],
  "howToWin": [
    "Lead with purpose-built engineering workflows (sprints, cycles, dependencies) vs blank-canvas setup",
    "Show time-to-value: structured out of the box vs 'build it yourself' in Notion",
    "Target teams frustrated that Notion slows down as their project data grows"
  ],
  "objectionHandling": [
    {
      "objection": "We already run everything in Notion.",
      "response": "Keep Notion for docs — we plug into it and give engineering the structured sprint and dependency tracking Notion isn't built for."
    }
  ],
  "landminesToSet": [
    "Ask how long it took to set up and maintain their project structure in Notion.",
    "Ask what happens to performance once a database has thousands of rows."
  ],
  "battlecardMarkdown": "# Battlecard — Notion\n**Domain:** notion.so\n\n> An all-in-one connected workspace...\n\n## How they position themselves\nPositions itself as the single flexible workspace...\n\n## Strengths\n- Extremely flexible building blocks\n...\n## How to win against them\n- Lead with purpose-built engineering workflows\n...",
  "sourcesUsed": ["ai-company-enrichment", "b2b-tech-stack-detector"]
}

techStackSummary / pricingSummary are null when that organ returns no data; recentMoves is [] when no news is found. sourcesUsed always reflects which organs actually contributed.

---

Pay-per-event pricing

Event	Price	Charged when
competitor-teardown	$0.30	Once per competitor successfully analyzed into a battlecard.

One charge bundles up to four orchestrated sub-actor runs (firmographics, tech stack, pricing, recent news) plus the OpenAI synthesis into a finished one-page battlecard. The Actor logs its OpenAI token usage and estimated AI cost per competitor at the end of every run, so the event price always stays safely above the model cost.

---

Setup

The AI layer calls the OpenAI API. Set the environment variable / secret OPENAI_API_KEY on the Actor. It's required — without it the run stops before synthesis. The orchestrated organs run under your Apify account.

---

Use it from an AI agent (Apify MCP server)

This Actor is available through the Apify MCP server, so AI agents (Claude, ChatGPT, or anything that speaks MCP) can call it as a tool — hand it domains and yourOffering, and read back a full battlecard per competitor — with no custom integration code.

---

Run it locally

npm install
# put your input in storage/key_value_stores/default/INPUT.json
export OPENAI_API_KEY=sk-...   # required for the AI battlecard
apify run -p

---

Related actors

• B2B Pricing Extractor — Pull a competitor's published pricing, plans and tiers from their site.
• B2B Competitor Monitor — Watch competitors for changes to pricing, messaging and product over time.
• B2B Tech Stack Detector — Detect the technology stack and buying signals behind any domain.
• B2B Intent Radar — Surface in-market buying intent signals for your target accounts.

---

Built by the No Niner team.