From file to useful result
How to ask with this skill installed
Installing the skill gives your AI a method. Your request still has to provide the case-specific facts, constraints and expected output.
1Name the real task
Describe the decision or deliverable, not just the topic.
2Provide the evidence
Add source material, audience, limits and known facts.
3Define done
Set format, quality criteria and checks.
Source example
Replace the brackets with your case
When to use it
- Setting up a new CRM from scratch (Salesforce, HubSpot, Pipedrive, Zoho, etc.)
- Your current pipeline doesn't reflect how your sales team actually sells
- Deal stages are too vague to give managers meaningful data
- Forecasting accuracy is poor (stages don't correlate with actual close probability)
- Different teams sell differently (SMB vs. enterprise, inbound vs. outbound, product vs. service)
- Your deal sizes vary so much that a single pipeline doesn't capture the process
- You need a customer success pipeline (post-sale) separate from sales pipeline
Mistakes to avoid
- Building too many stages — if a rep has to click through 12 stages to close a $5,000 deal, they'll game the system or stop using the CRM; 5-7 stages is the practical maximum for most SMB/Mid-market processes
- Stage names that are vague ("Engaged," "Working," "Active") — these mean different things to different reps and produce inconsistent data; be specific: "Discovery Call Completed," "ROI Calculator Sent"
- Not enforcing required fields at stage transitions — if reps can advance a deal without updating the "Next Steps" or "Close Date" field, your forecasting will be meaningless
- Forgetting the "Closed Lost" stage — deals that are left in a late stage without being marked lost distort the pipeline and create false hope in forecasts; require a loss reason to exit properly
- Building a pipeline once and never updating it — stage probabilities change as your product, market, and sales motion evolve; review and recalibrate quarterly