comms-measurement

Skill: Comms Measurement

Translates business and communications objectives into a structured measurement framework — KPIs by funnel stage and PESO channel, baselines, realistic targets calibrated against sector benchmarks, and a clear methodology for attribution. Ensures comms is accountable to business outcomes, not just activity metrics.

Core Principle

Activity ≠ Results. Most agencies report Level 1 metrics (impressions, press hits, followers) and call it measurement. This skill insists on all four levels, with explicit methodology for attribution and sector-calibrated targets.

---

The 4-Level Framework

Level 4: BUSINESS IMPACT
└── Revenue, market share, NPS, CAC, retention rate
    Attribution: partial — comms claims contribution with caution

Level 3: AUDIENCE BEHAVIOUR
└── Website visits, sign-ups, trial, purchase, search lift, referral
    Attribution: direct — UTM, promo codes, brand search volume

Level 2: AUDIENCE PERCEPTION
└── Brand awareness (aided/unaided), consideration, preference, sentiment, recall
    Attribution: survey pre/post + media weight correlation

Level 1: COMMS ACTIVITY
└── Reach, impressions, press hits, SOV, engagement, followers
    Attribution: direct — platforms and monitoring tools

Rule: Every campaign must have at least one KPI at each level. If Level 2 measurement (perception) is missing because "we don't have budget for a brand tracker," flag this as a measurement gap that invalidates the campaign's learning value.

---

Step 1: Map Objectives to Metrics

Comms Objective	Level	KPI	Method	Target	Baseline
[Objective from Platform]	[1–4]	[Specific metric]	[Tool/methodology]	[Number]	[Current state]

---

Step 2: Sector Benchmarks

Never set targets without referencing sector norms. Use these calibration tables:

Brand Awareness Benchmarks (France, annual campaign)

Category	Typical unaided awareness gain per €100k media	Notes
FMCG mass market	+1–3 pp	Large existing base, heavy competition
FMCG premium / bio	+2–4 pp	More targeted, less noise
B2C tech / app	+3–6 pp	Higher shareability, younger audience
B2B enterprise	+2–5 pp among target audience	Very narrow target, longer cycles
Luxury	+0.5–2 pp	Slow burn, exclusivity limits reach
NGO / associatif	+4–8 pp	Earned media leverage, cause resonance

Example calibration: A FMCG bio brand spending 400k€ over 6 months targeting 25-40 ans can realistically expect +2–4 pp unaided awareness. A target of +6 pp requires either higher spend, exceptional earned media, or a viral moment — flag if set without justification.

Social Media Benchmarks (organic, 2026)

Platform	Avg engagement rate	Good rate	Exceptional
LinkedIn (brand)	1.5–2.5%	>3%	>6%
Instagram (brand)	0.5–1.5%	>2%	>4%
TikTok (brand)	3–6%	>8%	>15%
Twitter/X (brand)	0.3–0.8%	>1%	>2%
Facebook (brand)	0.1–0.5%	>1%	>2%

Earned Media Benchmarks

KPI	Realistic for 6-month campaign, 60k€ PR budget
Tier 1 press articles (Les Echos, Le Monde)	3–8
Tier 2 press articles (trade press)	15–30
Total earned reach (all press)	2M–15M potential impressions
Key message penetration	>50% of articles carry ≥1 KM
Positive/neutral sentiment	>80%

Email Marketing Benchmarks (France, B2B vs B2C)

Metric	B2B	B2C
Open rate	22–35%	18–28%
Click rate	3–6%	2–4%
Unsubscribe rate	<0.5%	<0.8%

---

Step 3: Vanity Metrics Identification

Flag these in every measurement framework as not sufficient for evaluation:

Metric	Why it's a vanity metric	Better alternative
Total impressions	Includes accidental views, no intent signal	Unique reach + frequency
Follower count	Bought, inherited, inactive — means nothing	Engagement rate, % active followers
"Viral" post (one spike)	Doesn't persist in brand memory	SOV trend over 3 months
AVE (Advertising Value Equivalent)	Discredited, arbitrary, misleading	Audience reach + message penetration + sentiment
Number of press releases sent	Activity, not impact	Press articles published / articles pitched
App downloads (without retention)	Zombie users inflate numbers	D7 retention, MAU/DAU ratio

AVE policy: AVE is formally deprecated. Replace with: (Audience reach of publication × estimated % who read the article) + sentiment score + key message penetration rate.

---

Step 4: Attribution Methodology

The hardest problem in comms measurement. Choose the right method for each campaign:

Method	Best for	Limitation
UTM + conversion tracking	Digital campaigns, direct response	Only tracks clicks — misses brand impact
Pre/post survey	Awareness and perception shifts	Expensive; other variables confound
Brand search lift	Brand campaigns with significant spend	Only works with reach scale; search intent varies
Geo-split test	Large brand campaigns (national vs control region)	Requires regional media buying
Media mix modelling (MMM)	Long-term brand investment	Requires 2+ years of data, expensive
Self-reported (survey)	Qualitative validation only	Recall bias, social desirability

For most mid-size campaigns: Combine UTM tracking (Level 3) + pre/post awareness survey (Level 2) + media monitoring for message penetration (Level 1). This triangulates sufficiently.

Attribution claim language:

• ✓ "The campaign contributed to a +X pp increase in awareness, measured via pre/post survey. Other factors (seasonal, product) also in play."
• ✗ "Our campaign generated +X pp awareness." (overclaims causality)

---

Step 5: Reporting Cadence

Report	Frequency	Audience	Contents
Live monitoring	Daily (campaign period)	Campaign team	SOV, social performance, press alerts
Weekly dashboard	Weekly	Agency + client comms director	L1–L2 KPIs, anomalies
Mid-campaign review	Mid-point	Client marketing director	Full scorecard, optimisation recommendations
End-of-campaign	End	Full stakeholder group	All 4 levels, attribution claims, learnings
Annual comms review	Yearly	CMO / CEO	Long-term brand metrics, strategic re-assessment

---

Dashboard Template

# Campaign Measurement Dashboard: [Campaign Name]
Reporting period: [Dates] | Prepared by: [Agency]

## Executive Summary
Status: [On track / Needs attention / Exceeded targets]
Key win this period: [1 sentence]
Key concern this period: [1 sentence]

## KPI Scorecard

| KPI | Level | Target | Current | vs Baseline | Status |
|-----|-------|--------|---------|-------------|--------|
| [KPI 1] | L1 | | | | 🟢/🟡/🔴 |
| [KPI 2] | L2 | | | | 🟢/🟡/🔴 |
| [KPI 3] | L3 | | | | 🟢/🟡/🔴 |

## Earned Media
Press hits: [N] | Top placement: [Title + reach]
Key message penetration: [%] | Sentiment: [%P/%N/%Nt]

## Paid & Social
Reach: [N] | Engagement rate: [%] | Best content: [Description]

## Owned
Traffic: [+/-% vs baseline] | Newsletter: OR [%] / CTR [%]

## Vanity Metrics (for reference only — not evaluated)
Total impressions: [N] | Follower growth: [N] — not used for success assessment

## Attribution Note
[Describe what was measured how, and what causal claims are supportable]

## Optimisation Recommendations
1. [Action on underperforming area]
2. [Amplify overperforming area]

---

Output

Save to data/1-Projets/campaigns/[campaign]/measurement-framework.md

Structure:

1. Measurement Architecture — 4-level KPI set
2. PESO KPI Dashboard — Targets with sector-benchmarked baselines
3. Vanity Metrics Watchlist — What NOT to optimise for
4. Attribution Methodology — How causal claims are structured
5. Reporting Templates — Dashboard per report type
6. Data Sources Map — Where every number comes from

---

Integration Points

• Receives from: brief-analyzer (KPIs), campaign-strategy (channels), audience-intelligence (segments)
• Feeds into: Campaign reports, annual reviews, new business pitches
• Triggered by: /comms:measure, /comms:strategy, /comms:campaign