Utility PM Skill Iterate: Storevine Iterate Campaign Analytics

Scenario

After running /pm-skill-validate on the campaign-analytics skill, the contributor received two warnings: (1) EXAMPLE.md is 142 lines, below the 150-line recommendation, and (2) SKILL.md is missing a “When NOT to Use” section. The contributor uses /pm-skill-iterate to apply both fixes based on the validation report. [fictional]

---

Prompt

/pm-skill-iterate

Skill: campaign-analytics
Path: skills/measure-campaign-analytics/

Apply fixes from validation report:
1. Q-01: Expand EXAMPLE.md Attribution Model section to reach 150+ lines
2. Q-04: Add "When NOT to Use" section to SKILL.md

---

Output

Iteration Change Summary: campaign-analytics

Intended Changes

1. Target: skills/measure-campaign-analytics/references/EXAMPLE.md -> Attribution Model Change: Expand section with a worked first-touch vs. multi-touch attribution comparison using fictional campaign data Source: validation report Q-01

2. Target: skills/measure-campaign-analytics/SKILL.md -> When NOT to Use Change: Add new section after “When to Use” listing three scenarios where other skills are more appropriate Source: validation report Q-04

Proposed Changes

skills/measure-campaign-analytics/references/EXAMPLE.md

Attribution Model — before:

Attribution Model

Model: First-touch attribution Rationale: For Storevine’s initial Campaigns launch, first-touch is sufficient because each merchant typically runs one campaign at a time. Multi-touch attribution will be revisited when automated drip sequences ship in Q3. [fictional]

Attribution Model — after:

Attribution Model

Model: First-touch attribution (with multi-touch comparison) Rationale: For Storevine’s initial Campaigns launch, first-touch is the primary model because each merchant typically runs one campaign at a time. A multi-touch comparison is included below to establish the baseline for when automated drip sequences ship in Q3. [fictional]

First-Touch Example

Campaign	Sends	Revenue (first-touch)
Spring Sale	4,200	$18,400 [fictional]
New Arrivals	2,800	$9,200 [fictional]
Win-Back	1,500	$6,100 [fictional]

Multi-Touch Comparison

Campaign	First-Touch Revenue	Multi-Touch Revenue	Delta
Spring Sale	$18,400	$14,200	-$4,200 [fictional]
New Arrivals	$9,200	$11,800	+$2,600 [fictional]
Win-Back	$6,100	$7,700	+$1,600 [fictional]

Insight: First-touch over-credits the Spring Sale campaign because many of those buyers also received the New Arrivals email before converting. Multi-touch redistributes $4,200 of attributed revenue to downstream touchpoints. This gap will widen as drip sequences add more touchpoints per buyer journey. [fictional]

skills/measure-campaign-analytics/SKILL.md

When NOT to Use — before:

(Section did not exist)

When NOT to Use — after:

When NOT to Use

• A/B test design: If you need to design a controlled experiment comparing two campaign variants, use /experiment-design instead. This skill defines the measurement framework, not the experimental protocol.
• Event tracking schemas: If you need to specify the exact event payloads and tracking implementation for engineering, use /instrumentation-spec. This skill defines what to measure, not how to instrument it.
• Dashboard layout: If you already know your KPIs and need to specify the dashboard UI, use /dashboard-requirements. This skill produces the analytical inputs that feed into a dashboard spec.

Applied Changes

File	Change
skills/measure-campaign-analytics/references/EXAMPLE.md	Expanded Attribution Model with first-touch vs. multi-touch worked comparison (+22 lines, now 164 lines total) [fictional]
skills/measure-campaign-analytics/SKILL.md	Added “When NOT to Use” section with three differentiation scenarios

Version

Current: 1.0.0 Suggested bump: patch (documentation improvement, no output contract change) New version: 1.0.1 [fictional]

HISTORY.md

Created

# campaign-analytics Change History

## 1.0.1 (2026-04-11)

- Expanded EXAMPLE.md Attribution Model section with multi-touch comparison
- Added "When NOT to Use" section to SKILL.md for clearer skill selection