displacement-radar

Draft-only competitor displacement intelligence for seller teams. Built during the Apollo x Google Cloud Hackathon.

What It Does

displacement-radar helps a seller team turn competitor pain signals into reviewed outbound drafts.

• Discover or enter competitors.
• Identify recent competitor signals from public sources.
• Map signals to likely affected accounts and buyer personas.
• Use Apollo to find decision makers and contact data.
• Generate evidence-backed outreach drafts for human review.

The product boundary is intentionally draft-only: it does not send emails, create Gmail messages, or enroll contacts in sequences without external human action.

Main Workflow

1. Enter your company and product.
2. Add competitors manually or use model-assisted competitor discovery.
3. Run a scan to collect and rank competitor signals.
4. Review signal-backed opportunities and impacted accounts.
5. Select decision makers found through Apollo.
6. Generate draft outreach for selected contacts.
7. Review, edit, and approve any external action outside the app.

Preview screenshots from frontend

Using the Apollo MCP to identify decision makers at potential customer companies (from auto-discovered displacement opportunities):

Comprehensive competitive landscape report, generated from web-scraping, Claude-reviewed competitorrs, and Apollo GTM tools:

Opportunity graphs:

Competitive landscape visualizations:

Quickstart

python -m venv .venv
source .venv/bin/activate
pip install -r apollo_hackathon/requirements.txt
uvicorn apollo_hackathon.app.main:app --reload

Then open http://127.0.0.1:8000.

The canonical app lives in apollo_hackathon/.

Configuration

Copy and fill values from:

• apollo_hackathon/.env.example

Key variables:

• APOLLO_API_KEY - Apollo account/contact lookup.
• OPENAI_API_KEY - OpenAI web-search prospecting.
• CLAUDE_MCP_CONFIG - path to a Claude MCP config file.
• PREFER_CLAUDE_MCP_PROSPECTING - prefer Claude MCP prospecting when configured.
• CLAUDE_DRAFT_EMAILS - use Claude for draft email generation when configured.
• GOOGLE_CLOUD_PROJECT / GOOGLE_CLOUD_LOCATION / VERTEX_MODEL - Vertex/Gemini options.
• DATA_PATH / SEED_DATA_PATH - local app data and seed snapshot paths.

Testing

pytest

pyproject.toml is configured to run tests under apollo_hackathon/tests.