Data

Data Architect Job Description

A Data Architect decides how data is structured, named, governed and shared across an organisation, then holds that design steady while other people build against it. This is a design and standards role rather than a build or an operations one: a Data Engineer implements the pipelines, a Database Administrator keeps a given database healthy, and the Architect decides which system is the source of truth for a customer record, what a canonical entity definition looks like, and on what terms a new source is allowed into the estate. Because these decisions have a long shelf life, the useful evidence is not which tools a candidate knows but which trade-offs they have owned and what happened two years later. Employers get the most signal by asking a candidate to critique a design they inherited, including the parts they chose to defend.

Key skills

Conceptual, logical, and physical data modellingWarehouse, lakehouse, and operational store trade-offsMaster and reference data design, including canonical entity definitionsGovernance frameworks: ownership, stewardship, classification, retentionData contracts and interface design between producing and consuming systemsMetadata, lineage, and catalogue practicePrivacy and residency obligations treated as architectural constraintsMigration sequencing from legacy estates toward a target-state design

Responsibilities

  • Define the target-state data architecture and a roadmap the business can realistically absorb
  • Own canonical definitions for core entities so teams stop maintaining conflicting versions
  • Set modelling, naming, and layering standards that delivery teams build against
  • Review proposed new sources and systems against the architecture before they are adopted
  • Design how sensitive information is classified, masked, retained, and deleted
  • Sequence migrations off legacy stores so nothing critical is stranded mid-move
  • Advise leadership on build-versus-buy choices for platform components
  • Keep lineage and design documentation current so the architecture is discoverable, not tribal

Requirements

  • Extensive experience designing data models that multiple teams then built on
  • A track record of owning an architecture decision through implementation and its consequences
  • Fluency in the trade-offs between warehouse, lakehouse, and transactional stores
  • Working knowledge of governance and privacy obligations in your jurisdictions
  • Ability to write a design document a non-specialist executive can act on
  • Experience influencing delivery teams without direct authority over them

Nice to have

  • Led a migration from an on-premise estate to a cloud platform
  • Earlier career spent hands-on building pipelines or running databases
  • Exposure to a regulated sector with formal retention and audit obligations
  • Familiarity with domain-oriented ownership models such as data mesh
  • Introduced a catalogue or glossary that colleagues genuinely kept using

What to look for in a great Data Architect

Longevity of decisions is the thing to probe. Anyone can draw a diagram; the question is whether their designs survived contact with reality. Ask for a decision they made three or more years ago and what it looks like now, including what they would change. Strong architects describe constraints, not just ideals, and can explain why they accepted a particular compromise. Look also for someone who writes well, because this role produces standards, design notes, and definitions other people must read and follow. A candidate who cannot express a design clearly on paper will not win adoption however sound the design is.

Interview questions to ask a Data Architect

Ask how they would resolve two teams disagreeing on what counts as an active customer when both have a dashboard proving their own number. You are testing whether they reach for ownership, definition, and governance rather than a purely technical fix. Then ask them to describe an architecture they inherited and disagreed with: what they changed first, what they left alone, and why. Leaving things alone is a maturity signal. Close with what would make you tell us we do not need a Data Architect yet. Honest candidates will answer it, and that answer is worth more than enthusiasm.

Where to source Data Architects

Senior data engineers who have drifted into design work informally are the most common route, and they arrive knowing what is actually buildable. Consultants from analytics practices bring breadth across many estates but should be tested on whether they ever lived with their own decisions. Enterprise architects from larger organisations can be excellent, though some are used to a level of authority a smaller company will not grant. Look in governance communities, dimensional-modelling user groups, and among practitioners who publish designs openly, since this is a field where people share their reasoning in public.

Red flags when hiring a Data Architect

The clearest warning sign is a candidate whose answer to every situation is the same reference architecture regardless of company size or constraint. Be wary of people who have only produced diagrams and never watched an implementation team try to follow them. Someone who dismisses your existing systems wholesale in a first interview will likely alienate the teams they need on side. Watch for governance framed purely as restriction, with no account of how people get their work done, because that architect gets routed around within months. An inability to name any decision they regret suggests limited exposure to consequences.

How an ATS speeds up hiring a Data Architect

This hire shapes years of downstream work, so the process deserves structure rather than a chain of unstructured chats. With Pitch N Hire's ATS you can run every candidate through the same design-focused interview kit, capture written scorecards from data, engineering, and business stakeholders, and compare people on the same criteria instead of on who interviewed most persuasively. Because architects are often assessed through a written design exercise, keeping submissions and reviewer notes attached to the candidate record makes the debrief quicker and the decision defensible later. Pipeline visibility also shows you when a strong candidate has been waiting a week on one stakeholder's feedback.

Hiring a Data Architect? See Pitch N Hire on your roles.

FAQ

Hiring a Data Architect — FAQs

What does a Data Architect do? +
A Data Architect defines how data is modelled, named, governed, and moved across an organisation, and sets the standards that delivery and analytics teams build against. Typical outputs are a target-state design, canonical definitions for core entities, modelling and naming conventions, contracts between systems, classification and retention rules, and a migration roadmap. The role is accountable for coherence over time rather than for shipping any individual pipeline or report.
What is the difference between a Data Architect and a Data Engineer? +
A Data Engineer builds and operates pipelines, transformations, and platform components. A Data Architect decides how the whole estate should fit together and sets the rules those pipelines follow. In a small company one person may do both, and that is often sensible. The split becomes worth making when several teams are building independently and starting to contradict each other, because that is a design and governance problem rather than a capacity problem.
When does a company need a Data Architect? +
The trigger is usually symptoms rather than headcount: multiple conflicting definitions of the same metric, nobody able to say which system owns a record, repeated bespoke integration work for every new tool, or a legacy platform everyone agrees must be replaced but nobody can sequence. Before that point a senior engineer with design instincts is often enough. Hiring an architect too early tends to produce elegant documents with no implementation behind them, which damages the function's credibility.
How do we evaluate a Data Architect with no data leader in-house? +
Use a written exercise grounded in your own situation. Give the candidate a short description of your systems, two or three real business questions you struggle to answer, and ask for a one-page proposal with the trade-offs made explicit. Have your engineers read it for buildability and a business stakeholder read it for comprehensibility. If neither group understands it, that is information. You can also ask an independent senior practitioner to review the final two submissions, which is a modest cost against a decision this durable.
What should a Data Architect job description include? +
Describe the estate honestly, including the legacy parts, because architects want to know what they are inheriting. State the first outcome you need, such as a single customer definition or a migration plan, rather than listing platforms. Be explicit about authority: can they set standards teams must follow, or will they advise and persuade? That one line filters candidates more effectively than any technology list. Also say who they report to and which stakeholders they must win over.
Pitch N Hire ATS

Post this job description and track every applicant in one place

Pitch N Hire is an applicant tracking system. Publish this role to your careers page and job boards, then follow every applicant through screening, interviews, and offer without a spreadsheet.

  • Publish to your careers page and job boards from one job record
  • Screen and shortlist applicants against the criteria in this description
  • Move candidates through stages with the whole panel seeing the same view

Free for 1 user · No credit card · Talk to a real hiring expert

Built for recruiters & hiring teams

Ready to hire a Data Architect?

Post this role to multiple job boards and screen, interview and decide — all in one AI-native platform.

Prefer to talk? Book a demo · Talk to sales · View pricing

Free 1-user plan · No credit card · Talk to a real hiring expert

One Hiring Infrastructure.
Zero Tool Chaos.

Demos are consultative. We respect privacy and enterprise
governance. No lock-ins.

Start free Book demo