No coding experience required to read the playbook. No API keys required to run the tools.
This project answers one question: "How do I actually use AI to sell better?"
It includes two things:
- A 5-chapter strategic playbook that walks you through using AI at every stage of the sales cycle — from finding prospects to closing deals
- 4 working Python tools you can run right now to see AI-assisted sales in action
Everything works out of the box. The tools run in mock mode by default, meaning you can explore every feature without signing up for anything or entering an API key.
Every tool maps to a specific stage of the sales cycle, and every playbook chapter teaches you the strategy behind it:
graph LR
A["<b>Prospecting</b><br/>Find the right accounts"] --> B["<b>Research</b><br/>Understand the prospect"]
B --> C["<b>Outreach</b><br/>Personalize at scale"]
C --> D["<b>Qualification</b><br/>Score the deal"]
D --> E["<b>Competitive Intel</b><br/>Win against rivals"]
E --> F["<b>Close</b><br/>Win the deal"]
style A fill:#4A90D9,stroke:#2C5F8A,color:#fff
style B fill:#7B68EE,stroke:#5A4FCF,color:#fff
style C fill:#E67E22,stroke:#BA6518,color:#fff
style D fill:#27AE60,stroke:#1E8449,color:#fff
style E fill:#E74C3C,stroke:#C0392B,color:#fff
style F fill:#2ECC71,stroke:#27AE60,color:#fff
graph LR
P1["Chapter 1<br/>AI Prospecting"] -.-> A1["Prospect<br/>Researcher"]
P2["Chapter 2<br/>Personalized Outreach"] -.-> A2["Outreach<br/>Personalizer"]
P3["Chapter 3<br/>Deal Qualification"] -.-> A3["Deal<br/>Scorer"]
P4["Chapter 4<br/>Competitive Intel"] -.-> A4["Battle Card<br/>Generator"]
P5["Chapter 5<br/>Implementation Guide"] -.-> A5["All Tools<br/>Together"]
style P1 fill:#f5f5f5,stroke:#ccc,color:#333
style P2 fill:#f5f5f5,stroke:#ccc,color:#333
style P3 fill:#f5f5f5,stroke:#ccc,color:#333
style P4 fill:#f5f5f5,stroke:#ccc,color:#333
style P5 fill:#f5f5f5,stroke:#ccc,color:#333
style A1 fill:#7B68EE,stroke:#5A4FCF,color:#fff
style A2 fill:#E67E22,stroke:#BA6518,color:#fff
style A3 fill:#27AE60,stroke:#1E8449,color:#fff
style A4 fill:#E74C3C,stroke:#C0392B,color:#fff
style A5 fill:#4A90D9,stroke:#2C5F8A,color:#fff
Each tool handles a different part of the sales workflow. All four run in mock mode by default — no API keys, no setup, no cost.
What it does: Builds a complete research brief on any company — overview, recent news, pain points, tech stack, key contacts, and recommended approach angles.
| Input | A company name (e.g., "Snap Inc") |
| Output | A structured research brief with talking points |
| Mock data | Pre-built briefs for Snap Inc and HubSpot; generates plausible briefs for any other company |
| Why it matters | Cuts prospect research time from 45 minutes to 5 minutes |
python run_tools.py research "Snap Inc"What it does: Takes prospect research and combines it with outreach templates to produce personalized emails scored on specificity, personalization depth, CTA clarity, and tone.
| Input | A prospect name + template type (cold_intro, warm_referral, event_followup, renewal, upsell) |
| Output | A personalized email draft with quality scores |
| Mock data | Works with any prospect; uses researcher output for personalization |
| Why it matters | Level 3 personalization at Level 1 speed — every email feels hand-written |
python run_tools.py personalize --prospect "Snap Inc" --template cold_introWhat it does: Analyzes free-text deal notes and extracts BANT signals (Budget, Authority, Need, Timeline) to produce a qualification score from 0-100.
| Input | Deal notes from your CRM, call summaries, or email threads |
| Output | A score card with BANT breakdown, risk signals, and recommended next actions |
| How it works | Deterministic regex analysis — no AI/LLM needed. Fast, free, transparent, fully offline |
| Why it matters | Turns subjective "gut feel" pipeline reviews into data-driven qualification |
python run_tools.py score --notes "Met with VP Marketing, budget of $150K approved, Q3 pilot"What it does: Produces competitive battle cards with strengths, weaknesses, differentiators, objection handling scripts, trap questions, and landmine responses.
| Input | A competitor name (e.g., "Google Ads") |
| Output | A structured battle card ready for your sales team |
| Mock data | Pre-built cards for Google Ads and Outreach.io; generates plausible cards for any competitor |
| Why it matters | Your reps walk into every competitive deal with a prepared playbook |
python run_tools.py battlecard --competitor "Google Ads"The full playbook is a strategic guide you can read without touching any code.
| # | Chapter | What You'll Learn |
|---|---|---|
| 1 | AI-Augmented Prospecting | How to reduce prospect research time by 80-90% with structured prompts |
| 2 | Personalized Outreach at Scale | How to send deeply personalized emails to hundreds of prospects, with an A/B testing framework |
| 3 | Deal Qualification & Scoring | How to replace gut-feel pipeline reviews with BANT scoring and NLP pattern matching |
| 4 | Competitive Intelligence | How to generate battle cards, run win/loss analysis, and monitor competitors automatically |
| 5 | Implementation Guide | A 90-day rollout plan with tool selection, privacy framework, and ROI measurement |
Here is real output from the Deal Scorer tool, given the input: "Met with VP Marketing, budget of $150K approved, interested in Q3 pilot"
Deal Qualification Score
========================
Overall: 45/100 (D) — Needs Work
Dimension | Score | Max
-------------|-------|-----
Budget | 24 | 25 "budget of $150K approved"
Authority | 7 | 25 "VP Marketing" (but who signs?)
Need | 7 | 25 "pilot" mentioned (but pain unclear)
Timeline | 7 | 25 "Q3" referenced (but no driving event)
Recommended Next Actions:
1. Map the buying committee — identify the economic buyer
2. Deepen discovery — quantify the business impact of the problem
3. Establish timeline — ask about driving events that create urgency
What this tells you: Budget is strong (24/25) but the deal is weak everywhere else. The scorer flags exactly where to focus your next conversation — no guesswork required.
pip install -r requirements.txt
./run.shOpens at http://localhost:8501 with four interactive tabs — one per tool.
# Prospect research
python run_tools.py research "Snap Inc"
# Personalized outreach
python run_tools.py personalize --prospect "HubSpot" --template warm_referral
# Deal qualification scoring
python run_tools.py score --notes "Budget confirmed at $200K, CEO involved, launch by Q4"
# Competitive battle cards
python run_tools.py battlecard --competitor "Google Ads"All tools support live mode using any OpenAI-compatible API:
export OPENAI_API_KEY="sk-..."
python run_tools.py research "Salesforce" --mode liveWorks with OpenAI, Anthropic (via proxy), Azure OpenAI, Ollama, and more.
This is not a concept deck or a research paper. It is a working system — a documented playbook with running tools that a sales team can adopt today.
| What | Why |
|---|---|
| Mock mode by default | Stakeholders can evaluate the system without any setup or cost |
| Deterministic scoring | The Deal Scorer uses regex, not AI — fast, free, transparent, reproducible |
| Structured output | Every tool produces data that integrates with CRM workflows |
| Minimal dependencies | Zero external dependencies in mock mode; only Streamlit for the web demo |
| Live mode ready | Flip a switch to connect to any OpenAI-compatible API for production use |
CJ Fleming — 15+ years in media sales leadership, Columbia AI certification.
This project is the bridge between operational sales expertise and the AI-enabled future of revenue organizations. The playbook reflects how pipeline reviews actually work, what makes outreach convert, and how to roll out new tools without disrupting a sales team's rhythm. The tools prove that AI enablement is not theoretical — it is working code you can run today.