Every function here — growth, operations, product, client success — runs on the same warehouse. When a

model breaks, a definition drifts, or a dashboard goes stale, decisions across the company degrade quietly.

We need an analyst who owns that layer: builds it in dbt, keeps it trustworthy as the logic underneath

changes, and turns it into visibility people actually use.

This is not a report-pulling job. It is equal parts data engineering discipline and business judgement — you

will be asked what the number means, not just what it is.

• Build and extend dbt models across staging, intermediate and mart layers — tested, documented, and

version-controlled.

• Own the warehouse structures for your domains: schema design, incremental strategies, and the

cost/performance trade-offs behind them.

• Maintain essential models as business logic shifts — refactor definitions without breaking downstream

consumers, and communicate the change when you do.

• Monitor the KPIs across the funnel — lead flow, connectivity, conversion, CAC — and build a deep

understanding of what moves each one.

• Investigate movements before anyone asks: find the drop, isolate the cause, name the growth

opportunity.

• Convert a business question into a data problem, and have an intuition for when the answer should be

productised rather than answered once.

• Measure how our AI agents perform in the wild — call outcomes, containment, transfer rates, language-

level and cohort-level differences.

• Build the datasets that let product and ML teams compare agent versions and prompt changes against

real conversion outcomes.

• Surface failure patterns early and quantify their revenue impact.

• Design dashboards teams open daily — clear visual representations, sensible defaults, no ambiguity

about what a metric means.

• Make data self-serve: stable, well-named marts that stakeholders can query without a ticket.

• Maintain what you ship. Dashboards are a product, not a deliverable.

• Automate recurring reporting and set up alerts so problems reach people at the moment they matter.

• Work directly with internal stakeholders across growth, ops and product — scope the real question,

agree a closure deadline, and respond in a timely fashion.

• Remove your own toil: every hour of manual pulling should become a pipeline.

• Strong SQL — highly optimised queries on very large relational databases, not just joins and

group-bys.

• 1+ years in an analytics, BI or data-engineering role.

• Ability to rapidly learn BI platforms, database schemas, ETLs and business logic, and apply

critical thinking to them.

• Excellent communication — a simple, impactful message built out of a complex analysis.

• Analytical skill paired with business judgement and the ability to influence a decision.

• Hands-on dbt — models, tests, macros, exposures.

• Python for analysis and pipeline work.

• Cloud warehouse experience (Snowflake, BigQuery, Redshift).

• Git and code-review habits.

• Exposure to experimentation or A/B testing.

• Interest in conversational AI, speech data, or contact-centre metrics