Data Analyst - SQL, dbt, BigQuery, Reporting & BI - Contract-to-Hire

10–30 hrs/wk · Contract-to-hire intended · We screen on evidence, not rate; strong candidates at any point in this band are equally welcome, and the final rate is agreed after the paid pilot · Hard requirement: 4+ hours daily overlap with Toronto (EST). Americas ideal; Asia/Western Europe/Africa workable; please don't apply from +8h or worse unless you already work EST hours and can prove it with work history. This is an analyst role, not a pipeline-engineering role: you'll work inside dbt models that already exist, owning reporting, BI-development, and AI-assisted analysis where your verification is the product. Real scope from week one; contract-to-hire for the right person. Clearoute Inc. is a growth partner with 25+ years across branding, marketing, web development, and consulting, running large-scale paid and owned media for nonprofit and commercial clients. Campaign decisions run on our data platform: dbt-modeled warehouse on GCP, person-level identity resolution, five attribution models, automated validation and alerting. Our clients' biggest weeks of the year leave no room for wrong numbers, some of our reporting is compliance-grade. The live projects you'll plug into (day one): Recurring client reporting lane — performance overviews and campaign reporting on a fixed calendar, run end to end against gold-layer models. BI consolidation — validate legacy Looker Studio dashboards (a fleet with known metric-definition conflicts, e.g. competing "first-time donor" definitions) against dbt gold models; rebuild the keepers on Evidence/Streamlit/Looker. Attribution reporting — surface the warehouse's five attribution models into decision-ready views for Google/Meta budget planning. Owned-media analytics — email/SMS/WhatsApp performance measurement on person-level cohorts (retention, first-time vs. returning, revenue per send). Ramadan 2027 readiness — the reporting engine you'll own must run flawlessly through peak season (early February 2027). How we work: you'll use AI agents daily to accelerate analysis — and you'll verify everything they produce. Control totals, row counts, definition checks. Plausible-but-wrong does not leave your desk. You work inside our canonical metric definitions; when dashboards disagree, you diagnose and escalate, you never quietly ship a third number. To be direct about fit: if your happiest work is building ingestion pipelines and architecting warehouses, this role will frustrate you — the models exist; the value you add is trusted numbers and clear, decision-first communication about them. Milestone 1 (paid pilot, ~1–2 weeks, fixed price): on a test donations schema we provide: (1) validate dashboard views and document every definition mismatch; (2) build a first-time vs. returning donor cohort retention analysis with a verification appendix proving the numbers; (3) review an agent-produced analysis and verify them. Graded against a published rubric; strong delivery converts to the hourly engagement immediately, at a rate we agree then, informed by the work. Must-have: 3+ years analytics with strong SQL — window functions, CTEs, dedup patterns, grain discipline · hands-on work in a dbt-modeled (or equivalent) warehouse · demonstrated agent-in-the-loop practice with a real verification method · a recurring reporting deliverable you ran 6+ months without a miss, including upstream breakage you handled · decision-first written and spoken communication — this role presents numbers to people who act on them. Nice-to-have: dbt model contributions with tests · BigQuery · Evidence, Streamlit, or Looker · Python · fundraising/donor analytics or B2C retention analytics · cohort/experimentation methods. Data security: pilots run entirely on test data. Ongoing work requires NDA and IP assignment; access is scoped, PII is hashed at the platform boundary, and receipting-grade deliverables carry mandatory dual review. Skill tags: SQL · dbt · BigQuery · Data Analysis · Data Visualization · Looker Studio · Dashboard Development · Data Validation · Marketing Analytics · Reporting