Digital transformation after the transformation era
Why the next digital agenda is less about isolated programs and more about architecture, platforms, governance and measurable enterprise value.
Read articleDigital analytics should convert activity into decision-grade evidence
Visits, clicks and impressions describe system activity; they do not explain customer progress or business value. Analytics begins with a decision: what changes if evidence crosses a threshold? That defines the outcome, signals, population and comparison. Instrumentation without a decision model produces abundant reporting and little organisational learning.
A measurement plan should connect events to a journey and a data contract. Names, parameters, identity rules, consent state and validation tests need owners. Track failures and missingness because clean dashboards can conceal broken collection. Reconcile key events�orders, qualified leads, renewals�with operational systems rather than assuming browser data is complete.
Analysis should use cohorts, sequences and distributions. Averages hide whether new users, mobile devices or one acquisition source face different conditions. Funnel and path tools reveal patterns, but interpretation needs context from research, support and product operations. Check releases, seasonality and channel mix before interpreting correlation.
Attribution is not causality. Google Analytics describes models that allocate credit across touchpoints, including data-driven attribution; allocation remains an analytical convention, not proof that a channel caused the outcome. Incrementality requires a credible counterfactual through randomised experiments, holdouts, geographic tests or quasi-experimental designs, with assumptions and uncertainty reported.
Governance should make metrics reproducible and restrained. Maintain definitions, lineage, access controls, retention and change logs; minimise collection that lacks a clear use. Scorecards should combine customer outcomes, behaviour, economics and guardrails. Digital analytics creates advantage when teams can challenge an interpretation, reproduce the evidence and make a better decision�not when more events are counted.
Related macro
Articles
Why the next digital agenda is less about isolated programs and more about architecture, platforms, governance and measurable enterprise value.
Read articleWhy cybersecurity, identity and information integrity increasingly shape whether companies can scale digital channels, AI and connected ecosystems.
Read articleFocus
Automation is useful when reliable data and decision rules support repeatable customer interactions across channels.
Suppliers, platforms and service providers create dependencies that can transmit disruption across enterprise boundaries.
Strategic challenges
The challenge is separating legitimate operating value from use cases driven mainly by technology narratives or speculative interest.
The challenge is distinguishing critical relationships from a large vendor population with very different risk profiles.
POV
Product strategy requires exclusion: what not to build matters as much as what receives engineering capacity.
Global consistency matters, but local language and market behavior should override internal vocabulary when customers differ.
Strategic impact
Connecting threats with business consequence helps leadership prioritize capabilities, investments and control maturity.
Assessing workflow, ergonomics, acceptance and integration helps identify contexts where advanced interfaces support better execution.
What we observe
Teams can become dependent on services that add coordination cost without materially reducing engineering effort.
Slow loads, inconsistent states and inaccessible interactions can undermine even well-designed customer journeys.