Building your own playbook
Prospecting 9 min
0/28
Business problem
After 3 months of prospecting you have knowledge, but it's in your head and scattered across messages. A new employee (or your future 'you') starts from zero. A playbook turns knowledge into an asset: the answer to 'what works for you in prospecting' takes the form of a document, not a memory.
Proven method
Playbook structure:
- ICP — the ideal customer card (current, with lessons: who doesn't work),
- CHANNELS — what and when (email/LinkedIn/phone, sequences),
- PROMPTS — all tested prompts with a 'what worked' annotation,
- SEQUENCES — email templates (opener, value, CTA, follow-ups 2-
- ,
- METRICS — the pipeline with benchmarks (reply 3-8%, booking 20-30%),
- LESSONS — what won (subjects, openers, send times) and what didn't. AI workflow: every week AI updates the playbook from data (which opener got the most replies) and your notes. The rule: a playbook isn't a document written once — it's a report from experiments. Every month review 3 sections: ICP, prompts, lessons.
Market example
- BAD
- Unstructured prospecting without ICP filters and generic outreach to random accounts.
- GOOD
- After 6 weeks the playbook contains: 'L4 opener (trigger) works 2x better than L3 (problem) — 9% vs 4% reply' (a lesson from data). 'ICP update: 5-15 person companies reply, 1-person ones don't — they have 0 budget' (a lesson). 'Research prompt changed: + signal strength column' (a prompt). A new salesperson reads the playbook and knows in week 1 what works — instead of testing everything from scratch.
Ready-to-use prompt
You are a sales playbook editor. Context: I have [X] weeks of prospecting experience. My data and notes: [DATA: metrics, what worked, what didn't] Task: Build/update my playbook: 1. ICP: update the card from data (who replies, who doesn't) 2. Channels: what works, what I dropped and why 3. Prompts: a list of tested ones + what worked (annotations) 4. Sequences: an email 1-5 template with lessons 5. Metrics: the pipeline with benchmarks 6. Lessons: the 5 most important (what wins, what to avoid) Format: 6 sections, ready to paste into a document.
Model rekomendowany: Claude 3.5 Sonnet / GPT-4o Copy and paste into ChatGPT, Claude or Perplexity
ALTERNATYWNE SCENARIUSZE
Variations & alternative scenarios
New salesperson version: the playbook as onboarding (what to read in week 1).
Manager version: the playbook as a team standard + control checklists.
Practical exercise
Create a playbook skeleton: 6 sections with headings. Write into each section what you already know (even if it's 3 sentences). Schedule the weekly update in your calendar.
Zapisano w przeglądarce