
Bahlsen Group: Taking Over a Data & AI Estate Nobody Had Mapped
Inherited a data platform with no estate-level documentation. Mapped it, audited it in three lenses, sequenced the modernization, and shipped the two internal products that were missing.
A mapped estate, a severity-ranked audit, a sequenced modernization roadmap, and two internal products shipped.
- Security
Programme results
- Estate audit
- AI Compass
- Design system

An estate with no map
I took over as Interim Head of Data & AI in July 2026 and inherited a platform I had not built. Years of work by internal staff and external partners sat underneath me: a cloud landing zone, a warehouse on BigQuery, a transformation layer in dbt, a BI layer on top, a self-hosted runtime executing every scheduled job, and a drawer full of internal applications.
The engineering was mostly sound. What was missing was any document that explained how it fit together. The naming conventions were good and written down nowhere. Onboarding meant reverse-engineering the flow from source systems to dashboards out of individual repository READMEs.
So my first delivery was an orientation document. I pulled every repository in the estate into one read-only mirror and wrote what did not exist: the layer map, the conventions, the deploy contract, the end-to-end data flow, and an honest inventory of which repositories were load-bearing versus stubs and duplicate copies.
I want to be exact about attribution. I did not build this platform. My job was to be able to hold it in one head, and then decide what to do with it.
The audit, run in three lenses
I ran the estate through three separate reviews and ranked everything by severity: a strategy lens on bets and concentration risk, an architecture lens on failure modes and security, and a developer-experience lens on what it is like to ship a change here.
The findings clustered.
- One keystone, mid-replacement. Every scheduled workload ran on a single self-hosted runtime, and the organisation was halfway through migrating it to a second deployment. I ranked that as the top strategic risk.
- No cross-layer observability. Failure visibility was per-repository webhooks. Nothing enforced that ingestion had finished before transformation ran. Stale dashboards were caught by an analyst who happened to look.
- Uneven security and operational hygiene across the estate. The specifics stay internal; the severity ranking put the worst of them at the top of the remediation queue.
- No estate-level orientation document at all. The cheapest artefact the team was missing, and the one the audit itself produced.
One finding I pulled out of the roadmap and handled as an incident.
I describe the assessment method in How I assess an inherited data platform before planning its modernisation.
Cheap fixes first, orchestration second
The modernization assessment covered the whole chain: ingestion, transformation, orchestration, serving. Every layer scored low, and the temptation with a scorecard like that is to propose a rebuild.
I proposed an order of operations. The ingestion pattern was full-load and append-only with no merge key, and most transformation models did not deduplicate on read. Re-runs propagated duplicate rows into downstream metrics, which meant wrong numbers on live dashboards, with no orchestrator involved. Test coverage across the model layer sat far below where a platform this load-bearing needs it.
So: close idempotency and observability first, because neither needs new infrastructure. Build the orchestration backbone second. Modernize ingestion third. A multi-month orchestrator programme must not gate the deduplication fix that stops wrong numbers this week.
Writing that order down mattered more than the target architecture did.


AI Compass: turning a demand list into a decision tool
The AI use-case portfolio arrived as a document. Every department had submitted candidates, every candidate had been scored across a dozen dimensions, and the whole thing was unusable in a steering call because answering "what is unowned" meant reading it end to end.
I built it into an internal web application, access-controlled end to end, because the underlying data is sensitive.
- A command centre with the two numbers a manager acts on, how many cases have nobody's name on them and how many are quick wins, plus a feed of what moved in the last seven days.
- A delivery map, a sortable table and an impact/effort matrix with configurable axes, all reading one filter state out of the URL. Drilling into a tile keeps your filters and narrows the view. Any filtered view is a link you can send to a colleague.
- A department scorecard: every team on one page, with demand, horizon split, evidence coverage, quick wins, ownership gaps and outstanding follow-ups, every number clickable through to the cases behind it.
- An action board collecting every open follow-up across the portfolio into one list, so nobody opens hundreds of detail pages to find them.
- Shortlists, exports and presentation mode. Pick cases across screens, export the view or the selection as CSV at three levels of detail, or turn either into a slide deck or a one-page brief that prints straight to PDF.
The part I am most pleased with is the authorization model, because it will still be correct after I leave. Access and role are two separate questions, and both are answered in reviewed code. A membership change ships as a small merge request the app itself renders for you; the app writes nothing, and the merge is the authority. Permissions resolve per request, so a change takes effect immediately, including for sessions already open.
The steering method is in How I turn an AI use-case list into a portfolio leaders can act on.
A design system, because every internal app was inventing one
The second gap was visual. Every internal application had grown its own palette and buttons, and its own idea of what a chart should look like.
I built a framework-neutral design system as a monorepo: layered design tokens, accessible controls as native custom elements, and thin typed wrappers so the same components are comfortable in React, Svelte and the e-learning framework the training courses use. One Storybook catalog, one fixed semantic version across every published package, changesets on every consumer-visible change.
The accessibility work is the part worth describing. Tested colour pairings, contrast thresholds and chart-encoding metadata ship as data alongside the tokens, so a consumer can assert against them. Every series in a chart carries a shape as well as a colour, and the legend repeats the shape next to the name, which is how the charts survive greyscale printing, colour-blindness and forced-contrast mode. The one contrast pair I could not fix inside a design-system release is written up as an open item with the measured ratios.
Making the undocumented searchable
Three smaller pieces of work, all aimed at the same problem: institutional knowledge that existed but could not be queried.
I archived internal wikis into auditable, machine-readable form with per-document access lists attached, then built a conflict register that pairs contradictory statements from different sources side by side with their evidence, an accountable owner, a verification test and acceptance criteria. Each contradiction is now a ticket with a name on it.
And I built a knowledge graph over meeting transcripts as a navigation aid, with communities, cross-cutting hub nodes and the questions the graph is well placed to answer. It shipped with its own coverage warning at the top, because extraction was partial and readers need to know where the map runs out.
How I work here
Infrastructure and application live in separate repositories with their own pipelines. Changes ship as reviewed pull requests, including my own. Findings get written down with a severity, a file path and a proposed fix, so somebody else can act on them without me in the room.
An interim mandate ends. Everything I have built here, the estate map, the audit, the roadmap, the portfolio app and the design system, is an artefact somebody else can pick up and keep running.