Learning guide
AI skills for Excel users: what to learn before moving into Python
If you already use Excel, start learning AI through a spreadsheet task you understand. You can practise asking better questions, reviewing suggested formulas and checking conclusions before deciding whether Python belongs in your workflow.
Your spreadsheet knowledge gives you a useful foundation: you already work with data, calculations and business questions. The next step is learning how to assess AI assistance.
Start with a question you can check
Choose a familiar task, such as comparing monthly sales or identifying overdue invoices.
Write down what the result should measure before asking for help. For example:
“Compare revenue across complete calendar months, exclude cancelled orders and show refunds separately.”
That gives you a clearer basis for reviewing a proposed calculation than “analyse my sales”.
Practise with fictional or approved data, following your organisation’s rules for external tools.
Learn to review suggested formulas
When an AI tool proposes a formula, ask it to explain the inputs and assumptions. Then test the formula yourself.
Check what happens with blank cells, duplicate records, text in numeric fields and dates stored in different formats.
Consider a revenue calculation that multiplies quantity by unit price. A plausible formula can still be unsuitable if the dataset includes returns, discounts or tax that require separate treatment.
The important skill is recognising what the formula assumes about your data.
Check data quality before interpreting trends
Before asking why a number changed, establish whether the comparison is valid.
Look for incomplete reporting periods, missing records, inconsistent category names and duplicate transactions. Check whether the definition of a metric changed between periods.
AI can suggest checks, but you remain responsible for deciding whether those checks address the dataset in front of you.
A convincing explanation cannot repair an invalid comparison.
Practise reviewing a written conclusion
Give an AI tool a small table of verified results and ask for a short management summary.
Read it against the numbers. Does it confuse percentage growth with percentage points? Does it describe a relationship as a cause? Does it imply that an incomplete month represents a full period?
Edit the summary until every factual statement can be traced to the supplied evidence.
This connects spreadsheet work with a skill that matters in any analysis tool: communicating what the results support.
When should you move into Python?
Consider Python when your work repeatedly involves processing many files, reproducing the same transformations or maintaining a sequence of steps that has become difficult to manage manually.
That decision should follow a recurring need.
Before moving, make sure you can explain the process you want to automate: its inputs, transformations, checks and expected output. Those requirements will still matter after the tool changes.
A project to try this week
Build a monthly performance report from a fictional dataset. Document the metric definitions, data issues, calculations and final recommendation.
Use AI for suggestions, then record at least one suggestion you changed or rejected and explain why.
That gives you evidence of judgement as well as tool use.
Where does a course fit?
If you want structured practice applying AI to analysis, DataCamp’s AI for Data Analysts is one option. Its published curriculum includes data quality, prompting, insights and output verification. It should not be treated as an Excel-specific course.
Explore DataCamp AI for Data Analysts, or compare options in our AI courses for data analysts guide.
If you are still deciding between practical AI use and programming, take the course finder quiz.
Course information checked 26 October 2026. Sources: DataCamp curriculum