High-quality web analytics is not simply a matter of deploying a tool, collecting events or producing regular reports. It needs to connect the organisation’s objectives, decision-making processes, data model, technical implementation and ongoing quality assurance.
My goal is to create measurement that helps people make better decisions and remains useful as websites, technologies, teams and privacy requirements change.
What high-quality web analytics means to me
Good measurement must answer real questions. It is not enough for data to be sent correctly and displayed in a tool. The data also needs to be understandable, sufficiently accurate, traceable and useful to the people who make decisions with it.
I therefore assess quality across several connected layers:
- purpose: it is clear why something is measured and which decision it should support,
- definition: people understand what a metric includes, excludes and how it is produced,
- implementation: the technical solution matches the design and works in important edge cases,
- quality assurance: errors and changes can be detected, assessed and corrected in time,
- use: the data reaches the people and processes where it can create value,
- sustainability: the solution can evolve together with the product and organisation.
These layers cannot be separated for long. Precise implementation without a useful objective creates expensive data collection. A good strategy without delivery remains a document. A report based on untrustworthy inputs can lead to worse decisions than having no report at all.
The analytics problems I work on
I am most useful when analytics goes beyond the routine configuration of a single tool. This can include:
- translating product and business objectives into a measurement strategy,
- designing a data model and measurement architecture for a large or technically complex project,
- preparing measurement for a website or platform migration while preserving data continuity,
- investigating differences between web analytics, advertising platforms and business data,
- assessing an existing implementation and determining which data can be trusted,
- connecting analytics data with other systems and putting it to practical use,
- designing controls that prevent, detect and communicate measurement problems,
- measuring emerging areas where an established approach does not yet exist.
In a complex project, the underlying problem rarely sits within the analytics tool alone. Its cause may lie in the objective, technical architecture, development process, use of data, allocation of responsibilities or conflicting expectations across teams.
Understanding the context and the real cause first
I begin by comparing the available data sources and looking for obvious errors, inconsistencies and missing context. I avoid drawing large conclusions too early. First I need to understand how the data is produced, who uses it, which decisions the measurement should support and why the current solution looks the way it does.
This means speaking with the project owner, engineers, marketing teams and other people who build the system or act on its data. An apparent error may have a legitimate historical or operational reason. Conversely, an implementation that looks technically correct may be measuring something different from what people assume.
Only by combining data, technical, business and historical context can we identify the real problem, understand its importance and decide whether solving it will create enough value.
From strategy and data models to a working solution
Depending on the scope of the problem, I can contribute throughout the analytics solution’s full lifecycle:
- clarifying objectives, decisions and the needs of different roles,
- designing the measurement strategy and data model,
- creating implementation documentation and technical architecture,
- working with engineers and other teams during delivery,
- implementing parts of the measurement, tooling or quality controls myself,
- testing data and validating important scenarios,
- evaluating the result and the impact of the change,
- developing the solution further as new needs and findings emerge.
I do not automatically stop at an audit or a set of recommendations. When it helps the project, I connect strategy with practical delivery and build original scripts, methods or tools when established options are not sufficient.
Analytics as part of a wider system
Web measurement does not operate in isolation. It is part of the product, technical platform, marketing, decision-making processes and legal framework. Depending on the project, I therefore connect analytics with:
- website and platform engineering and architecture,
- SEO, PPC and other marketing disciplines,
- UX, CX and product decision-making,
- business data and the needs of leadership,
- privacy, consent, GDPR and ePrivacy in collaboration with lawyers and data protection officers,
- AI, measurement of its effects and changes in search and digital visibility.
My role often involves translating the same problem between leadership, domain specialists and technical teams. Each group needs a different level of detail, but the resulting solution must remain consistent in both meaning and implementation.
Trustworthy data requires continuous work
Analytics changes together with the website. A new feature, a change in consent, an update to the data layer, a platform migration or a new business model can affect both the meaning and quality of measurement. An implementation that is correct today is not guaranteed to remain correct a year from now.
A strong solution therefore includes preventive controls, the ability to detect problems and clear communication of their impact. When an error occurs, it needs to be acknowledged, its consequences addressed where possible and the process improved so that the same failure does not recur in the same way.
The long-term outcome should not be dependence on a single analyst. The people who use the data should understand its meaning, limitations and how to work with it safely.
What I need from the client and their team
Good analytics cannot be created without the people who understand the product, technical environment and business reality. To produce a strong result, I particularly need:
- access to the relevant people, systems and available data,
- open communication about objectives, history, constraints and known problems,
- clear responsibilities and someone with the authority to make decisions,
- a willingness to allocate resources to changes that require other teams,
- a genuine intention to use the data, not merely collect it.
I can prepare a professional recommendation, explain it and help deliver it. Final priorities and business decisions, however, must be made by the person who holds that responsibility.
Related information
Services and engagement fit
Types of projects, the boundaries of my offer, pricing and how to assess whether working together makes sense.
Ethics and responsibility
My commitments concerning truthful data, confidentiality, privacy, respect for people and responsible use of technology.
Contact
Contact and billing details are available on a separate page.