Data quality, metadata, governance, migration, reconciliation and cross-functional delivery in complex enterprise environments, with financial services as the strongest recent commercial context.
Trusted Data for Trusted Decisions
Steven Zhang — Trusted Data & Transformation Consultant
Make fragmented enterprise data reliable for transformation, analytics and AI.
I help regulated and data-intensive organisations resolve the gaps between data, meaning, ownership and decision-making. My established commercial core combines enterprise data governance, data quality, metadata, migration, reconciliation and practical transformation delivery. Financial services is my strongest recent commercial context.
Reliable decisions depend on data that remains connected to reality, meaning, evidence and accountability. That is what makes complex change measurable, trustworthy and actionable.
The Trusted Data Framework is my current synthesis of that recurring problem: a practical approach for finding where trust breaks down and deciding what to do next.
Discuss a role or advisory needView my experienceExplore the Trusted Data Framework
A first conversation can start with a role brief, a data-quality or governance challenge, a migration problem, or a specific example of a decision that depends on data you are not yet confident you can trust.
Email: steven.zhang@trusted-data.tech
LinkedIn: Steven Zhang on LinkedIn
What actually goes wrong
One customer. Three records. One confident wrong answer.
Nothing here is a broken system. Each one is doing its job. The trust is lost where nobody decided what counts as one customer.
- CRM
ACME LTD
London · active
- Billing
Acme Limited
SW1 · dormant
- Risk register
ACME (UK) LTD
UK · high risk
One company, three representations, no agreed identity.
You ask the AI
How many customers do we have?
It answers
“Three.” Confidently, in half a second.
The missing work is agreeing what counts as one customer — and making that definition usable in every system. That is the work I do: keeping identity, meaning and context intact as information moves between systems, teams and uses.
Three strengths behind the work
Make trust practical through delivery.
Independent assurance and research-led work on spatial-temporal representation developed a lasting focus on what is real, how records refer to it and whether evidence can withstand challenge.
Independent AI-ready data, semantic and controlled prototype work extends the same trust principles into source authority, permissions, provenance, conflicting evidence, accountability and human review.
From problem to decision
Three questions move the work forward.
Identify who is making the decision, what outcome matters, and what could go wrong if the data is misunderstood.
Find the critical sources, definitions, ownership, quality rules and unresolved conflicts behind that decision.
Test one practical intervention, learn from the evidence, then adapt and scale only when it works.
Explore when useful
Use the depth only when it helps the decision.
Start with the problem in front of you. The Framework, delivery methods and supporting concepts are there when more structure helps — not because every situation needs the whole system.