Professional Certificate in Data Analysis

No. Every exercise works with the free tiers of Claude, ChatGPT, or Gemini, plus a free spreadsheet and free tools like Looker Studio and SQLite online. The dashboard module covers the paid BI-native AI landscape (Power BI Copilot, Tableau Agent, Gemini in Looker) so you can evaluate them at work, but none are required for the program.

This is Data Analysis — the business-analyst track. The rule that draws the line: you learn to interpret models and analyses (regression, forecasts, A/B tests), frame decisions, query data, build dashboards, and tell the story that drives action. The moment the job becomes building, training, and deploying predictive machine-learning models, that's Data Science — a separate craft, covered by our Professional Certificate in Data Science. Take Data Analysis if you want to answer business questions with data and AI; take Data Science if you want to build predictive models. Module 5 has a dedicated lesson on exactly where the boundary sits.

No — and deliberately so. In 2026 you describe what you need in plain English and AI writes the SQL and the analysis; your job is to frame the question and verify the output. You'll paste prompts into AI chat tools and work in spreadsheets and free BI tools. The SQL module teaches you to read and validate queries, not hand-write them from scratch — because validation, not authorship, is the skill that matters now.

Module 0 includes an honest self-assessment with a gap-to-course map. If you score low on spreadsheet fluency or basic AI prompting, it routes you to the three short prerequisite courses first. If you've ever pulled insight from a spreadsheet — even informally — you'll likely clear the bar.

Three short courses: AI for Google Sheets & Excel, Data Analysis with AI, and Tell Stories with Data and AI. Together they take about 5 hours. The Module 0 self-assessment tells you whether you can skip any of them. (AI for Databases & SQL is an optional booster before the SQL module.)

You analyze one realistic business — Northwind Outfitters, an outdoor-gear retailer with a regional-sales mystery — across Modules 2-8: framing the question, cleaning its messy data, running the statistics, querying it with SQL, building its dashboard, and delivering the recommendation. Then the capstone hands you a completely different business (FreshCrate, a meal-kit subscription) to analyze solo. You finish with two complete analyses plus a personal AI-analyst playbook.

About 6 weeks at 4 hours per week — roughly 24 hours total, split between ~15 hours of lessons and ~9 hours of hands-on analysis practice. Fully self-paced, and the capstone rewards learners who don't rush it.

The evidence points the other way. AI is absorbing the mechanical work — writing SQL, first-draft cleaning, generating charts — but it's confidently wrong a lot on real data: text-to-SQL accuracy collapses from ~86% on clean data to ~10% on real enterprise schemas, and wrong queries run silently. That shifts the analyst's value to framing, judgment, and verification — catching the errors AI can't see it's making. This program trains exactly that combination, which is why it's more durable than the tool skills alone.

The course teaches individual skills in about 2 hours — basic analysis, exploration, presentation. This program integrates everything into a working system across 40 lessons: a full analytics pipeline on a realistic business, applied statistics you interpret correctly, SQL you validate, dashboards you make honest, a recommendation that drives a decision, and a capstone you run solo. It's the difference between knowing the pieces and having run a complete project with them.

The certificate carries a verifiable credential ID. More practically, the program produces artifacts you can show — a cleaned dataset with a reconciliation, a statistical analysis, a validated SQL workflow, an honest dashboard, an executive recommendation, and a capstone scored against a professional rubric. In interviews, walking through how you caught an AI fanout join that doubled revenue lands harder than any certificate line.

No. The program teaches the statistics that matter for decisions — reading them correctly, telling significance from importance, and resisting causation-from-correlation — at an interpretation level, not a derivation level. There are no proofs or matrix calculus (that's the data-science side). If a task genuinely required calculus, you'd be on the data-science track; here, you interpret, and AI computes while you verify.

The AI layer, systematized, plus the verification discipline. Most working analysts use AI casually. This program gives you the full operating model: where AI fits each pipeline stage, its documented failure modes (the silent fanout join, the confident-but-wrong narrative, the misleading chart it picked), and the verification habit that catches each one — plus a playbook that turns scattered prompting into a repeatable system, and the storytelling frameworks (Minto, so-what, three altitudes) that get your work acted on.

Two directions. To go deeper in the analyst craft toward leadership — governing an org's metrics layer, running experimentation programs, leading an analytics function — the pathway continues to the Master Certification tier. To learn to build predictive models, the Professional Certificate in Data Science is the sibling track. Module 0 shows you the full map, and the capstone's final lesson marks exactly where you stand and both next climbs.