Pipeline: from list to meeting
Prospecting 10 min
0/28
Business problem
Measuring 'how many emails I sent' is satisfying but says nothing about effectiveness. Without tracking conversion at every stage you don't know where you lose: whether the research is bad (few companies in the list), the content is weak (few replies), or the pipeline is sick (few meetings from replies).
Proven method
The metrics pipeline (count weekly):
- COMPANIES IN THE LIST (added this week to the system) — measures research,
- SENT EMAILS (distinguishing: unique vs follow-ups) — measures sending,
- REPLIES (reply rate = replies / unique sent; cold email benchmark: 3-8%) — measures content and list,
- MEETINGS (booking rate = meetings / replies; target 20-30%) — measures conversation,
- DEALS (conversion = deals / meetings),
- WIN RATE (won / all deals). Diagnostic indicators: a reply without a meeting (weak offer/questions), sending without replies (content or list), a meeting without a deal (conversation). AI workflow: every week AI collects the numbers from your spreadsheet/CRM and writes a report: 'Replies 4% (vs 3.8% before), bookings 25% — the list is better, but openers are weaker; test L4 variants'. The rule: don't count something 'because it's customary' — count what lets you answer 'where do we lose and what to change'.
Market example
- BAD
- Unstructured prospecting without ICP filters and generic outreach to random accounts.
- GOOD
- Week 1: 50 companies in the list, 25 sent, 1 reply (4%), 0 meetings. Conclusion: research OK (50 companies), content weak (4%). Week 2: you test L4-L5 openers (trigger instead of problem) → 40 sent, 4 replies (10%), 1 meeting. Week 3: you raise the list to 60 companies, because sending and replies are growing, but meetings are few — you work on the conversation. That's a pipeline: you know at every stage what to improve.
Ready-to-use prompt
You are a sales pipeline analyst. Context: I sell [PRODUCT]. My data for the week: Companies in the list: [X] Unique sent: [X] Replies: [X] Meetings: [X] Deals: [X] Won: [X] (Previous week: [DATA]) Task: Write a weekly report: 1. Calculate the metrics: reply rate, booking rate, conversion, win rate 2. Trend vs the previous week (up/down) 3. Diagnosis: where we lose (research/content/conversation/offer) — based on the proportions 4. 3 concrete actions for next week Format: a report.
Model rekomendowany: Claude 3.5 Sonnet / GPT-4o Copy and paste into ChatGPT, Claude or Perplexity
ALTERNATYWNE SCENARIUSZE
Variations & alternative scenarios
Beginner version: only 3 metrics (sent, replies, meetings).
Advanced version: metrics per channel (email vs LinkedIn vs cold call) and per segment.
Practical exercise
Fill your metrics pipeline with last week's numbers (or assume realistic ones). Calculate reply rate, booking rate, win rate. Point out the 1 stage where you lose the most and 1 corrective action.
Zapisano w przeglądarce