The Measurement Gap
The metrics that personalisation systems generate most naturally — click-through rate, session engagement, content interaction depth — are easy to collect, easy to report, and poor indicators of commercial value. A personalisation system can produce measurably higher click rates on recommended content while contributing nothing to revenue, pipeline, or customer retention. The measurement framework that reports clicks as success is not measuring personalisation effectiveness — it is measuring the engagement trap.
Effective personalisation measurement connects the content layer to commercial outcomes — through attribution models that account for the influence of content across multi-touch journeys, experimentation designs that isolate the personalisation effect, and outcome metrics that reflect what the business is actually optimising for.
The Four-Level Measurement Framework
Level 1 — Engagement metrics: Content interaction rates, recommendation click-through, variant engagement depth. These metrics are necessary for operational monitoring but insufficient for commercial evaluation. They tell you what the audience is doing, not whether personalisation is creating value.
Level 2 — Journey metrics: Progression rates through defined journey stages, influence of personalised content on stage transitions, audience retention across sessions. These metrics connect content behaviour to journey outcomes and are more closely correlated with commercial value than engagement metrics alone.
Level 3 — Commercial metrics: Pipeline influenced, conversion rate by personalisation cohort, customer acquisition cost for personalised vs. non-personalised audiences, revenue attributable to personalised content journeys. These metrics connect personalisation directly to commercial outcomes.
Level 4 — System health metrics: Segment coverage, content variant freshness, decisioning rule performance, audience model accuracy. These metrics measure the operational health of the personalisation system and predict whether Levels 1–3 metrics will remain valid over time.
Key Takeaways
1. Click rates and session engagement are insufficient measures of personalisation effectiveness — they measure activity, not value.
2. The four-level measurement framework — engagement, journey, commercial, and system health metrics — provides the complete picture needed to evaluate personalisation investment and guide improvement.
3. Personalisation attribution requires controlled experimentation — holdout groups, A/B testing, and multi-touch attribution models — to isolate the personalisation effect from other influences on commercial outcomes.