Nova
Log inGet started
AI training · Tech and data

AI training for data analysts

AI writes the query, cleans the export, and draws the first chart. The hardest part is still yours: asking the right question, checking the numbers, and saying what they mean for the business.

Get startedTry the interactive demo

Free, no account needed. You try the real product yourself.

In brief

AI for data analysts, in brief

AI helps data analysts with tasks such as “Write and check a SQL query,” “Clean an export before analysis,” and “Take a first look at an unfamiliar file.” Start with a single task, in Claude: allow 10 minutes the first time, and always review the result before you use it.

  • 10concrete use cases
  • 6recommended tools, 5 of them free to start
  • 10 to 20 minper use case, the first time

The tasks where AI helps most

  • Data cleaning and preparation
  • SQL queries
  • Charts and dashboards
  • Ad hoc analyses for business teams
  • Summary memos and presentations
  • Metric documentation
Use cases

10 concrete AI use cases for data analysts

For each task: how it goes today, then with the tool. And the time it takes the first time.

Claude

Write and check a SQL query

TodayFour joins, a poorly documented business rule, and a total that doesn’t match the finance team’s.

With ClaudeA commented query, its pitfalls explained, and two control queries.

  • 10 min
  • every day

Julius · Clean up a file and get it back

Clean an export before analysis

TodayDates in three formats, amounts stored as text, duplicates: half the time of an analysis goes into preparing the file.

With JuliusA clean file, the list of fixes applied, and the excluded rows kept separately.

  • 15 min
  • every week

ChatGPT · “Chat” mode, with an attached file

Take a first look at an unfamiliar file

TodaySomeone sends you a file with no documentation, and you have to guess what each column measures before you can answer the question.

With ChatGPTAn overview of the file, its anomalies, and five leads for analysis.

  • 10 min
  • every week

Also

Microsoft Copilot · Copilot in Excel

Answer a business request in Excel

A calculated column, a pivot table, and a chart, with formulas you can see and check in the workbook.

  • 10 min
  • every day

Julius · Charts on request

Build the chart for the committee

A chart that carries a message, with the table of values to check it.

  • 10 min
  • every week

Claude · With an attached file

Explain a change in a metric

A quantified breakdown of the change and three hypotheses, each with the data that would confirm it.

  • 15 min
  • every week

ChatGPT · “Chat” mode, with an attached file

Check whether a gap is significant

The right test, its calculation in detail, and a cautious conclusion a non-specialist can read.

  • 15 min
  • every month

Gemini Notebook · One notebook per data domain

Query the data dictionary

The definition of a metric, its tables and its filters, each point linked to the passage in the document.

  • 15 min
  • as needed

Claude · Projects

Write the summary memo for an analysis

A one-page memo, conclusion first, in the style of your earlier memos.

  • 15 min
  • every week

Gamma · “Paste”

Prepare the results presentation

A clear presentation, one message per slide, exported to PowerPoint.

  • 20 min
  • every month

All 10 use cases on this page are in Nova, each with its steps and its prompt.

Try the interactive demo
One use case, start to finish

Claude for data analysts: write and check a SQL query, step by step

  • 10 min
  • every day

What to check

Give the table structure, not the data. A query that runs without errors can still count rows twice: always check against a known total.

Write and check a SQL queryClaude
  1. Copy the structure of the tables involved: table names, columns, types, keys. Not the data.
  2. In Claude, paste the structure, then the prompt below.
  3. Read the explanation join by join, then run the query on a short period.
  4. Compare the result with a total you know before you use it.

The prompt to copy

You are a data analyst, comfortable with SQL [PostgreSQL, BigQuery, SQL Server]. Here is the structure of the tables: [paste the table names, columns, types, and keys]. I want to get [the metric: for example, monthly revenue by region, split between new and existing customers] over [period]. Write the query, with comments, then explain each join and say where rows could be counted twice or lost. Then give two control queries to verify the result. If a business rule is ambiguous, ask me before you write.

My path

What you’ll be able to do, one step after another

In Nova, “My path” starts from your level: one next step at a time.

  1. Talk to AI like a colleague

    The first moves, with free tools.

  2. Write faster, without mistakes

    In your jobWrite and check a SQL query

  3. Get answers from your documents

    In your jobClean an export before analysis

  4. Find an answer you can trust

    In your jobPrepare the results presentation

  5. Never retype anything again

    In your jobTake a first look at an unfamiliar file

For a team: everyone moves forward in their own job. You see the team’s progress, never one person’s activity.

The tools

AI tools for data analysts, and what they cost

Claude

Free to start

It writes and explains a SQL query, breaks down a change in a metric from a file, and writes the summary memo in your style.

Recommended planFree to start. Pro, $20 per month ($17 per month if billed yearly). At a company: Team, $25 per user per month ($20 if billed yearly).

Julius

Free to start

You upload an export and ask your question. It writes the calculation, runs it, draws the chart, and gives you back a cleaned file.

Recommended planFree, $0: 25 credits per day. To export charts and reports: Plus, $20 per month, or $16 per month billed yearly.

ChatGPT

Free to start

It reads an Excel or CSV file, explores it, and draws charts. Handy for a first look at an unfamiliar file or for checking a test.

Recommended planFree, with a limit on files per day. Plus, $20 per month, if you use it every day.

Gemini Notebook

Free to start

A notebook with the data dictionary, the business rules, and the scoping notes: it answers from those documents alone and shows the passage.

Recommended planFree: 100 notebooks, 50 sources per notebook, 50 questions per day.

Microsoft Copilot

If the company is on Microsoft 365: in Excel, it writes the formula, does the calculation, and creates the chart, showing the cells it changed.

Recommended planAt work: Microsoft 365 Copilot Business, €15.60 excl. tax per user per month, annual commitment. For personal use: Microsoft 365 Personal, €10 per month, or €99 per year.

Gamma

Free to start

You paste your analysis memo and get a clean presentation for the committee, one message per slide, exportable to PowerPoint.

Recommended planFree, $0, to try it: 400 credits that do not refill. Then: Plus, €8 per user per month, billed yearly (€96).

Prices taken from the vendors’ official pages, dated on each tool page. These subscriptions are not included in Nova. Brands are named to identify the tools. No affiliation.

Ground rules

The rules to follow with AI in this job

Always do this

  • Every figure an assistant produces gets recalculated before it is shared. Check at least one total you already know.
  • Personal data: remove names, emails, IDs, and salaries before you upload a file. Otherwise, use the tool approved by your IT team and your data protection officer.
  • Give the table structure rather than the data when that is enough. For a query, it almost always is.
  • The interpretation carries your name. The assistant knows nothing about what happens on the ground, the changes in scope, or the history of the metric.

Never hand this to AI

  • Uploading a customer database, a payroll file, or health data to a consumer tool.
  • Repeating a cause-and-effect conclusion the assistant suggested without testing it.
  • Publishing a dashboard through a public link when it contains confidential figures.
Questions

Frequently asked questions about AI for data analysts

Which AI tools should data analysts use?

Nova recommends 6 to start with: Claude, Julius, ChatGPT, Gemini Notebook, Microsoft Copilot, and Gamma. 5 of them can be used for free to get started.

Can data analysts use ChatGPT?

Yes. It reads an Excel or CSV file, explores it, and draws charts. Handy for a first look at an unfamiliar file or for checking a test. The plan to choose: Free, with a limit on files per day. Plus, $20 per month, if you use it every day.

Where should data analysts start with AI?

With “Write and check a SQL query,” in Claude. Allow 10 minutes the first time. A commented query, its pitfalls explained, and two control queries.

Is this an accredited course that leads to a certificate?

No. Nova is software, not an accredited training provider: it issues no diploma or certification. It costs €19 per month for one person, and €129 per month for a team of up to ten people.

Go further

AI training: related jobs