Free GenAI Tutorial — prompting for analysts in 12 days
How to get useful work out of AI for Excel, SQL and reporting - and how to catch it when it is confidently wrong.
A free, hands-on GenAI course for analysts
Twelve short lessons on using AI properly for analytics work - writing formulas and queries, cleaning data, drafting commentary, and catching the moments it is confidently wrong.
Start with Day 1.
Who this is for
Anyone in MIS, finance, operations or analytics who has tried ChatGPT or Copilot, got a mediocre answer, and wondered whether the problem was the tool or the question. It is almost always the question.
The 12 days
Day 1 — What a language model actually is
Not magic, not a search engine, not a calculator. Knowing what it is explains everything it …Day 2 — The anatomy of a prompt
Seven parts. Most bad prompts are missing five of them.…Day 3 — Being specific - the biggest lever
Every vague word in your prompt is a decision you have handed to a machine that cannot ask y…Day 4 — Giving it your data
The model cannot see your file. Everything it knows about your data is what you paste.…Day 5 — Controlling the output format
The difference between an answer you paste into your work and one you have to rewrite.…Day 6 — Examples and few-shot prompting
Showing beats telling. One worked example does more than three paragraphs of instruction.…Day 7 — Getting it to think step by step
For anything with more than one step, asking for the working improves the answer.…Day 8 — AI for Excel
Where AI genuinely saves an analyst time every single day.…Day 9 — AI for SQL
Give it the schema and it writes good SQL. Skip the schema and it invents your tables.…Day 10 — AI for Power BI, Tableau and reporting
Measures, calculations, documentation and the commentary nobody wants to write.…Day 11 — Verifying the output
The single skill that separates people who use AI safely from people who get caught out by i…Day 12 — Data privacy, ethics and 12 practice prompts
What must never leave your building, and twelve prompts to practise everything from this cou…The prompt workbench
Every lesson has a live prompt box. It scores the structure of what you write against the seven parts of an effective prompt and tells you exactly what is missing.
What this tool is, precisely
It is a structure checker, not an AI. Nothing you type leaves this page - no model is called, no data is sent anywhere, and it works with your internet disconnected. It cannot tell you whether an answer would be correct, only whether you have given a model enough to work with.
That distinction matters, and this course would be dishonest if it blurred it. Practising prompt structure is genuinely useful; a web page pretending to be ChatGPT would not be.
That distinction matters, and this course would be dishonest if it blurred it. Practising prompt structure is genuinely useful; a web page pretending to be ChatGPT would not be.
What this course will not do
It will not teach you to build AI models, fine-tune anything, or write Python for machine learning. It will not tell you which AI tool to buy. And it will not suggest AI can replace knowing your tools - every single prompt here assumes you can judge whether the answer is right.
The honest claim
On the work covered here, AI makes a competent analyst noticeably faster. It does not make a non-analyst into one, because you still have to spot the wrong answer. That is why the Excel, SQL, Power BI and Tableau tutorials sit alongside this one rather than being replaced by it.
How to get the most out of it
- Rewrite the weak prompts yourself before reading the strong version.
- Take the score as a hint, not a grade. A 70 that gets you what you needed beats a 100 that does not.
- Try each prompt in a real AI tool too - the workbench checks structure, the real tool checks whether it worked.
- Do Day 11 twice. Verification is the lesson that keeps you out of trouble.