Skip to content

Digital transformation

The plumbing under the marketing

Replatforming · Measurement · Marketing operations

Most marketing problems that look like strategy problems are plumbing problems. The tracking is wrong, the site cannot ship a page without a sprint, and nobody agrees what a lead is. No amount of campaign work fixes that — it just runs on top of it.

01

Why the plan keeps failing at the same point

These are the four blockers I find most often. They are unglamorous, they are rarely on anyone’s roadmap, and they quietly cap everything else you do.

The website is a bottleneck
Every change needs a developer, a ticket and a sprint. Marketing cannot ship a landing page, so eventually it stops trying and the site ossifies.
The numbers do not agree
Three tools give three different answers. The monthly report gets produced, nobody trusts it enough to act on it, and decisions get made on instinct anyway.
Last-click is still deciding your budget
Attribution that credits the final touch and hides everything that did the actual work. You end up defunding the channels that are working.
The process lives in someone’s head
No planning cadence, no test log, no definition of done. Quality depends on who happens to pick the work up, and none of it survives that person leaving.
02

Who this is for

Teams where the marketing plan is broadly right and execution keeps hitting the same wall.

  • Teams blocked by their own site

    Every page change is a developer ticket. Marketing has no control of the conversion funnel it is measured on.

  • Companies on a legacy CMS

    Slow, awkward to secure, and holding back both SEO and the team’s ability to move.

  • Post-raise teams

    Building the marketing function properly for the first time, and wanting the foundations right before the headcount arrives.

  • Regulated businesses

    Where Legal and Compliance are genuinely in the loop and the process has to work for them too, not around them.

  • Teams that do not trust their data

    Reporting exists but nobody acts on it. Usually a tracking and definitions problem rather than a tooling one.

  • Teams inheriting a stack

    You did not choose any of this, the person who did has left, and nothing is documented.

03

What you get

Each workstream is designed to move a specific metric, not to pad a report.

01

Website and platform migration

Moving off a legacy CMS onto something the marketing team can actually operate — faster, more secure, better for search, and editable without a sprint. I have led this end to end, including the front-end build.

02

Measurement and attribution

Tracking standards, GA4 and Search Console configured properly, attribution that reflects how people actually buy, and dashboards that surface the handful of numbers worth looking at. This is the work that stops last-click quietly misallocating your budget.

03

CRM and lifecycle plumbing

Connecting marketing activity to the pipeline: CRM integration, lead definitions everyone agrees on, and the reporting that lets you see which activity produced revenue rather than which produced clicks.

04

Operating rhythm

Planning, testing and governance frameworks — how work gets prioritised, how a test is judged, what “done” means. Boring, and the difference between a team that compounds and one that stays busy.

04

How it runs

Four phases with real durations. Nothing here is a proprietary framework — it is the order the work has to happen in.

01

Diagnose

Week 1–2

Audit the stack, the tracking, the site and the way the team works. I map what is actually broken as opposed to merely irritating, and what each fix would unlock. You get the findings whether or not you go further.

02

Sequence

Week 2–3

A prioritised plan. Not everything is worth fixing, and the order matters — measurement usually has to come before anything else, because otherwise you cannot tell whether the rest worked.

03

Build

Month 1–3

The actual work: migration, tracking implementation, CRM integration, process. I do this hands-on rather than briefing it, which is generally faster and always cheaper than a translation layer between the strategy and the build.

04

Hand over

Ongoing

Documentation, training, and a team that can run it without me. The test of this work is whether it still functions six months after I stop invoicing.

05

What this has produced

This is the least glamorous work I do and the most reliably valuable. Two published examples, plus a platform migration I can describe but not name.

Website conversion rate, from 0.67% to about 1.5%
2.2×Website conversion rate, from 0.67% to about 1.5%
Share of sessions from organic search, taking it to the site’s number-one channel
43%Share of sessions from organic search, taking it to the site’s number-one channel
Site traffic after rebuilding measurement and moving a team off last-click attribution
+30%Site traffic after rebuilding measurement and moving a team off last-click attribution

The first two figures are from a current engagement with a regulated digital-asset firm — a legacy WordPress site replatformed onto Next.js, with the measurement and content programme rebuilt around it. The third is eBay / KnownOrigin, 2022–2024, which is written up in full. I have also integrated CRM tooling and KPI dashboards and built the planning, testing and governance frameworks at Improbable.

Read the KnownOrigin case study in full
06

Questions people actually ask

Including the ones about money.

The other four

Most engagements use two or three together. A strategy sets the direction; the rest do the work.