Insights
Practical writing on measurement
Practical guides to marketing mix modelling, data foundations, tracking and testing.
50 articles

Measurement strategy
A 90-day plan to fix marketing measurement
What to do in the first three months to move from disputed numbers to decisions with evidence.

Conversion optimisation
Checkout optimisation: where to look first
Find the step with the biggest drop, understand why, then test.

Analytics implementation
Choosing a consent management platform
What to check before you choose, and how to implement it so data and compliance both hold up.

Data and analytics engineering
Joining CRM and marketing data without double counting
Capture the right IDs, join on them, and count each sale once.

Marketing mix modelling
Mix modelling for lead generation and subscription businesses
Choosing the right outcome, handling time lags, and linking leads to value.

Measurement strategy
How to brief a measurement agency
What to include so proposals are comparable and the work answers the right question.

Conversion optimisation
Research methods for conversion optimisation
Which research method answers which question, and how to combine them.

Analytics implementation
A tracking QA routine for every release
A short, repeatable routine that stops releases from silently breaking tracking.

Data and analytics engineering
Data contracts for marketing pipelines
Agreements between the teams that produce marketing data and the teams that use it.

Marketing mix modelling
Refreshing a mix model: how often, and what changes
A refresh cadence that keeps results current without rewriting history each time.

Measurement strategy
Measuring TV and BVOD alongside digital
Getting the data right, modelling them separately, and testing where you can.

Conversion optimisation
A/B/n tests: why more variants need more traffic
Each extra variant needs its own sample and raises the risk of false wins. Here's how to plan for it.

Analytics implementation
Enhanced conversions and offline conversion imports
How to send ad platforms the outcomes that matter, and what each approach needs.

Data and analytics engineering
Designing a marketing data model at weekly grain
A simple star schema that serves dashboards, forecasts and mix models from one source.

Marketing mix modelling
How to validate a mix model before you trust it
Six checks that separate a model you can act on from one that only fits the past.

Measurement strategy
KPIs marketing and finance both accept
A KPI tree that connects marketing activity to business outcomes, with definitions both teams sign.

Conversion optimisation
Prioritising tests: why evidence beats gut feel
A simple scoring approach that puts the most promising tests first.

Analytics implementation
Cross-domain tracking without broken sessions
Keep one visit as one session when people move between your domains.

Data and analytics engineering
Pulling spend from every ad platform reliably
APIs, connectors, currencies, time zones and the re-pull window that keeps spend accurate.

Marketing mix modelling
Priors in Bayesian mix modelling, without the maths
What priors are, why every model has assumptions, and how to set them sensibly.

Measurement strategy
Measuring brand channels that work over months
Why short windows undervalue brand activity, and what evidence captures its longer effects.

Conversion optimisation
Reading test results: lift, ranges and when to call it
How to interpret an A/B test result without over- or under-claiming.

Analytics implementation
GA4 key events: choosing what to count
Which actions to mark as key events, which to track without marking, and which to leave out.

Data and analytics engineering
Mapping campaigns to channels: the taxonomy that makes reporting work
A single mapping from platform campaigns to channels, used by every report and model.

Marketing mix modelling
Response curves: finding where the next pound goes furthest
How saturation works, why average return misleads, and how to use curves to plan budget.

Measurement strategy
Lift studies versus geo tests on Meta and Google
What platform lift studies measure, how they differ from geo holdout tests, and when to use each.

Conversion optimisation
Writing a test hypothesis that teaches you something
A simple format that makes every test worth running, whether it wins or not.

Analytics implementation
Writing an event specification developers will follow
One document that defines every event, parameter, trigger and owner, and keeps tracking consistent.

Data and analytics engineering
Data quality checks every marketing pipeline needs
Freshness, completeness, validity and reconciliation, so problems are caught before a report goes out.

Marketing mix modelling
Adstock and carryover explained for marketing teams
Why advertising keeps working after the spend stops, and how mix models account for it.

Measurement strategy
Building a measurement framework: which method answers which question
Match each marketing question to the method that can actually answer it.

Conversion optimisation
Conversion optimisation for B2B sites with low traffic
What to do when there isn't enough traffic for classic A/B tests.

Analytics implementation
UTM naming conventions that survive a reorganisation
A standard that stays readable when teams, agencies and tools change.

Data and analytics engineering
BigQuery or Snowflake for marketing measurement?
Both work. The deciding factors are usually where your data already is and who will run it.

Marketing mix modelling
What drives the cost of a marketing mix model
Five things that make a project bigger or smaller, and where to save.

Measurement strategy
White-label measurement: what agencies should look for in a partner
Criteria, red flags and how to protect the client relationship.

Conversion optimisation
Optimise the site or buy more traffic? Deciding with evidence
A framework for comparing the two, and the measurement each option needs.

Analytics implementation
Server-side tagging: when it's worth it
Benefits, costs and prerequisites, so you can decide with your eyes open.

Data and analytics engineering
The control variables your mix model is missing
Price, promotions, distribution and competitors: why leaving them out biases everything.

Marketing mix modelling
How much data do you need for marketing mix modelling?
History, variation, controls and granularity: what matters and what's negotiable.

Measurement strategy
The CFO's guide to marketing measurement
What finance should ask for, and what good answers look like.

Conversion optimisation
Why most A/B tests can't detect the effect they're looking for
Underpowered tests, peeking and what to do when traffic is limited.

Analytics implementation
Consent Mode v2: what it changes in your data
What the new signals do, basic versus advanced implementation, and how to read reports now.

Data and analytics engineering
Building a weekly marketing dataset in dbt
A layered structure for spend, outcomes and controls that stays trustworthy.

Marketing mix modelling
Mix modelling or attribution: which answers which question?
They're not rivals. They answer different questions, and mixing them up is expensive.

Measurement strategy
Geo holdout tests: a practical guide
From candidate regions to readout, step by step.

Conversion optimisation
A/B test sample size without the maths anxiety
How big a test needs to be, explained with one chart and one rule of thumb.

Analytics implementation
GA4 audit: twelve settings to check this week
A practical checklist, with why each setting matters.

Data and analytics engineering
The MMM data readiness checklist
Twenty checks to run before you commission a mix model.

Marketing mix modelling
Marketing mix modelling: a practical guide for UK teams
What a mix model is, what it needs from you, and what you should get back.