Embedded Data Governance for Water & Wastewater Utilities.
We take governance frameworks off the page and into delivery for regulated water and wastewater companies: critical-data-element standards agreed where regulatory returns are built, federated stewardship backed by central enablement, and quality and lineage controls that run inside pipelines — so evidence for economic and environmental regulators, including price-review submissions, is produced by the work rather than assembled after it.
Fragmented Silos
Legacy Utility systems and disconnected feeds.
Governance Inside Delivery, Not Beside It
Federated Stewards, Central Enablement
Production Reality
Critical-Data-Element Standards First
Critical-Data-Element Standards First
Federated Stewardship, Central Enablement
AI Governance Without Delivery Drag
Industry-Specific Friction Points
A Framework the Domains Outgrew Unevenly
The enterprise governance framework reads well, but maturity on the ground varies domain by domain — clean-water production data is stewarded one way, wastewater event data another, developer services barely at all. A single written standard is masking several very different operating realities.
Definitions Nobody Signed For
Ownership and stewardship exist where an individual cared, not where the framework assigned them. Business terms and critical data elements lack agreed definitions, so the figure called leakage in one team's report is not the leakage another team submits — and metadata and lineage records stop exactly where they are most needed.
Reactive Quality, Accelerating AI
Data-quality issues surface when a regulatory return is being compiled, not when the data was created — every reporting cycle becomes a cleanup exercise. Meanwhile AI use is spreading across the business faster than any ethical or governance control has been agreed, and the framework is silent on it.
How the Utility delivery flow works
This technical flow diagram reveals how Unolabs treats Utility data to deliver governed, production-ready outputs.
Source Layer
Maturity Baseline
Assessing each domain's actual governance maturity and building the register of critical data elements behind regulatory and price-review reporting.
Industry Logic
Standards & Stewardship
Agreeing definitions for the critical element set and appointing domain stewards with a central enablement function behind them.
Embedded Controls
Moving quality rules into delivery pipelines and capturing lineage automatically, starting with the flows that feed regulatory returns.
Activation
AI Gate & Sustain
Standing up a proportionate approval path for AI use cases and transferring the operating rhythm to your teams.
How the work is engineered for Utility
Governance Inside Delivery, Not Beside It
Controls are embedded where work already happens: definitions agreed when a data product is built, quality rules written into the pipeline that carries the data, lineage captured by the tooling teams already use. Nobody fills in governance paperwork after the fact — the delivery itself leaves the governed trail.
Federated Stewards, Central Enablement
Each domain keeps stewards who know its data and its operational context; a small central function supplies standards, tooling, coaching, and escalation. Accountability sits close to the water, while consistency comes from the centre — the opposite of a governance office policing from a distance.
Critical Data Elements Before Everything Else
Standards start where scrutiny lands: the data elements behind returns to the economic and environmental regulators, price-review evidence, and serviceability measures. Getting definitions, ownership, and lineage right for that critical set first buys credibility for extending governance everywhere else.
Where the Real Work Is
One Framework, Seven Starting Points
Rolling a uniform programme across domains at different maturity levels punishes both ends: mature domains resent re-doing what already works, immature ones drown in requirements they cannot yet meet. We sequence by risk and readiness — each domain gets a target state proportionate to the scrutiny its data attracts, and the roadmap is honest about which domains move first and why.
The Tooling You Own Is the Tooling We Use
A governance design that assumes a new catalogue, a new quality platform, and a new workflow tool has already failed procurement. Water companies run layered estates — telemetry historians, GIS, works and asset management, billing systems holding personal data under data-protection law — and governance has to land on that estate as it stands, extending what is owned before proposing anything new.
Ethical AI Controls That Do Not Become the Bottleneck
The fastest way to create ungoverned AI is an approval process nobody can get through. We design the AI gate as a time-boxed, criteria-based review — proportionate to the risk of the use case, with personal-data and customer-impact checks built in — so a safe use case clears quickly and the difficult conversations are reserved for the genuinely difficult cases.
Visible work products, not vague advice
Each deliverable is designed to be used by Utility architects, engineers, data owners, and operations teams after the engagement ends.
How We Measure Governance Uptake
A governance programme is working when delivery behaves differently, not when documents multiply. These measures start from wherever your domains genuinely are today, and movement is reviewed with your stewards in the open.
Stewardship coverage of critical data elements
Share of registered critical data elements with a named, active steward and an agreed definition, reported by domain so the uneven starting points stay visible instead of averaging out.
Embedded quality-rule pass rates
Proportion of prioritized data flows carrying automated quality rules inside the pipeline, and the pass-rate trend of those rules — quality caught at creation, not at return-compilation time.
Lineage coverage for regulatory reporting
Percentage of figures in scoped regulatory returns traceable end-to-end through captured lineage, measured per return so gaps are specific and fixable.
Time to approve a safe AI use case
Elapsed days from an AI use-case proposal to a governed decision, tracked alongside the share cleared at first review — the gate must protect without becoming the queue everyone routes around.
Frequently Asked Questions
Will embedded governance slow our delivery teams down?
It is designed not to — that constraint shapes every choice. Controls ride inside the work delivery teams already do: definitions agreed during build, quality rules in the pipeline, lineage captured by tooling. What disappears is the after-the-fact cleanup before every regulatory return, which is where the real delivery time was going.
Our domains are at completely different maturity levels. Where do you start?
With an honest baseline, then risk-based sequencing. Domains whose data carries the most regulatory scrutiny move first, each toward a target state proportionate to that scrutiny — not a uniform standard imposed everywhere at once. Mature domains keep what works; the framework bends to reality rather than the other way round.
What stops governance regressing after the engagement ends?
It was never ours to run. Stewards are appointed from your domains and coached in post from the first phase, the central enablement function is staffed by your people, and controls live in your pipelines and tooling rather than in our heads. Exit criteria are defined at the start — we leave when the rhythm runs without us, and there is no dependency to unwind.
How an engagement starts
Share your governance framework and one recent regulatory return ahead of a discovery session — we respond with a candid read on where the framework and day-to-day delivery diverge, and which critical data elements would anchor the first phase.
Interested in the full industry blueprint?
We have deeper technical documentation for Data Strategy & Governance for Utility in the Utility sector.
Bring your governance framework. We'll find where delivery quietly routes around it.