Why log in tickets?
AI effects are invisible unless you capture them where work happens. Logging Δ-time and Δ-quality in tickets creates an audit trail that survives team changes and proves value to leadership.
The Ticket Fields
Add these fields to your ticket template (Jira, Linear, Asana, etc.):
AI Assist Used
- [ ] Yes / [ ] No
If Yes:
Δ-time: [estimated time saved or lost, in minutes] Δ-quality: [+improved / = same / -degraded] AI tool: [which tool used] Notes: [what worked, what didn't]
Example Ticket Entries
Positive Delta
AI Assist Used: Yes
Δ-time: +45 min saved
Δ-quality: = same
AI tool: Copilot
Notes: Generated boilerplate tests; required 10 min of assertion fixes
Negative Delta
AI Assist Used: Yes
Δ-time: -30 min lost
Δ-quality: -degraded
AI tool: ChatGPT
Notes: Generated API design that missed auth requirements; rewrote manually
No AI Used
AI Assist Used: No
Notes: Task required institutional knowledge AI doesn't have
Sprint Review: AI Effects Section
Add a 5-minute section to your sprint review:
Agenda
- Total Δ-time this sprint: Sum of all ticket deltas
- Quality impact: Count of +/=/- quality entries
- Top win: Best AI assist story (share the prompt!)
- Top fail: Worst AI assist story (what went wrong?)
- Tool usage: Which tools used most?
Sample Slide
SPRINT 47: AI EFFECTS
Δ-time: +6.5 hours saved (net)
Quality: 12 improved | 28 same | 3 degraded
Top win: Test generation for payments module
→ 90 min saved, 0 quality issues
→ Prompt shared in #eng-prompts
Top fail: Architecture doc for auth refactor
→ 45 min lost, missed compliance requirements
→ Lesson: AI lacks institutional context
Tool breakdown:
- Copilot: 23 uses, +4.2 hr
- ChatGPT: 8 uses, +1.8 hr
- Claude: 5 uses, +0.5 hr
- ✓AI Assist field added to ticket template
- ✓Δ-time logged in minutes (positive = saved, negative = lost)
- ✓Δ-quality logged as +/=/-
- ✓Sprint review includes 5-min AI Effects section
- ✓Top win and top fail shared with prompts/context
Making It Stick
- Start small: Log AI effects on new tickets only; don't backfill
- Make it easy: Use dropdowns, not free text where possible
- Celebrate wins: Share best prompts in team channel
- Learn from fails: No blame; focus on "what did we learn?"
- Review monthly: Trend Δ-time over sprints to show ROI
“"We couldn't justify our AI tool spend until we started logging deltas. Two months of data showed 12 hours saved per sprint—budget approved."”
Related Resources
- Productivity Pack — structured measurement exercises
- Fair Trial Methodology — controlled AI comparison
Apply this now
Practice prompt
Add the AI Effects fields to one ticket template today.
Try this now
Log AI effects on your current ticket and prepare a slide for next sprint review.
Common pitfall
Logging only wins—negative deltas are where the learning happens.
Key takeaways
- •Add AI Assist, Δ-time, and Δ-quality fields to your ticket template
- •Run a 5-minute AI Effects section in sprint review
- •Share wins and fails with context so the whole team learns
See it in action
Drop this into a measured run—demo it, then tie it back to your methodology.
See also
Pair this play with related resources, methodology notes, or quickstarts.
Further reading
Next Steps
Ready to measure your AI impact? Start with a quick demo to see your Overestimation Δ and cognitive load metrics.