Module 4.2 · Track 4 Data Fundamentals ★0 / 2
Track 4 · Data Fundamentals

Spreadsheets and CSVs

The plainest file on your computer quietly became important for working with AI. This module is about the humble CSV: what it is, why it matters once AI is part of your work, and how to use it well.

Get it right, and a slow manual job like turning a rough campaign plan into a populated content calendar becomes something AI can do alongside you.

◐Two parts◷About 12 minutes★2 check-ins→A file to clean and keep
By the end: you'll understand why clean data matters to AI, and be able to prepare a file yourself.
Why it mattersThe big picture
AI is only as good as the data you give it.
Give it clean, well-structured data and it works well. Give it a mess and it will still answer, just wrongly, and often without flagging the problem. This track is about that difference, and it starts with the most common data file there is.
clean, structured data→answers you can trust
messy, unstructured data→answers that look right
Why it mattersStep 1 of 4
A CSV was once background admin. It is worth understanding now.
A CSV, short for comma-separated values, is a plain text file of rows and columns: each line is one record, and a comma marks where one column ends and the next begins. It sits underneath most spreadsheets, and it is the form your data takes when it moves between tools, AI included.
.csv
Before AI:
quiet admin
then AI
.csv
→ AI tools → Asana → reports
Now:
worth knowing
Why it mattersStep 2 of 4
Give AI data with a clear shape, and it can work with it directly.
Rows and columns are a form AI reads well. Hand it a tidy file and it can sort, total and summarise, and get the figures right.
datesitevisitors
01/06Museum342
02/06Museum389
AIreads the shape
✓"Busiest day was 2 June. Total was 731."
Why it mattersStep 3 of 4
A rough list can become a working content plan, without the manual entry.
Here is the same idea as a job a destination team does often. Follow it through.
Your rough list
  • Lighthouse feature
  • Coastal food trail
  • Autumn walks guide
  • Partner spotlight
AI makes a CSV
title,channel,date,owner Lighthouse,Instagram,05/09,Sam Food trail,Blog,08/09,Priya Autumn walks,Email,12/09,Sam Partner,LinkedIn,15/09,Jo
Import to your board
To do
Lighthouse
Autumn walks
Scheduled
Food trail
Partner
Click the diagram to walk through it →
Why it mattersStep 4 of 4
For a destination team, this comes up constantly.
Once you can move clean data in and out, AI fits into the everyday work rather than sitting to one side.
▦
Visitor numbers
Exported from your booking system for AI to summarise or forecast.
▤
Campaign calendar
A plan AI drafts, then drops into Asana or Monday.
◫
Partner listings
Operator data you clean, match and keep consistent.
❝
Review exports
Feedback pulled out for AI to read for themes.
Why it mattersCheck-in
★ Quick check-in

You've a rough list of twenty content ideas for the summer campaign, and you want them in your team's Asana board with dates and owners. What's the sensible move?

Putting it to workA closer look

Now, a look inside one

That is why it matters. Next, you'll open one of these files and see what makes AI read a clean one well, and stumble on a messy one.

  1. 1Flip clean against messy
  2. 2Compare the three views
  3. 3Ask it a real question
You don't need to take it all in. We'll point out what to notice, and there is nothing you can break.
Putting it to workTry it

Clean versus messy, and what AI does with each

👉 Start here: flip File from Clean to Messy, then open How AI reads it to see what trips it up. ✕
File
View
What's actually saved in the file.
How a spreadsheet displays that same file.
What the AI tool sees when it reads the file.
You ask AI: "What was total revenue?"
Putting it to workCheck-in
★ Quick check-in

In the messy file, one revenue figure was written as £1,945, with a comma inside the number. What will most tools do with it?

Putting it to workChoose your approach

Tools that open, check and clean CSVs

Excel
The spreadsheet most teams already have, and strong for hands-on editing. Watch its two famous habits with CSVs: reformatting dates and dropping leading zeros the moment a file opens. If a file looks different after saving, ask your AI assistant to compare the two versions.
Google Sheets
Free
Free, shareable, and opens CSVs cleanly. A good default for quick checks and light cleaning, with formulas most people already half-know.
Numbers
Free
Comes with every Mac. Fine for viewing and small edits; export back to CSV when you are done, so the plain format travels on.
OpenRefine
Free
A free tool built specifically for cleaning messy data at scale, with steps you can review, save and repeat. Worth learning when tidying thousands of rows is a recurring job; for a single file, asking AI for a cleaned version is usually faster.
CSV Lint
Free
Checks a CSV for structural problems, broken rows, mismatched columns, before you rely on it. A quick pre-flight for a file that is going somewhere important.

No single right answer. Pick what fits the job and what you already have open. For most quick fixes, a spreadsheet you know plus an AI tool is plenty.

Putting it to workYour turn

Clean a file yourself

Start here · the sample

The short messy file you just saw. Good for a first go.

The prompt to paste

Here is a CSV of visitor data that I know is messy. First, list the formatting problems you can see. Then give me a cleaned version: one row per record, dates as YYYY-MM-DD, numbers with no symbols or separators, and the title and total rows removed. Explain what you changed.
Go further · a fuller, real-world file
This one is closer to a real export: three sites, a full quarter, extra columns and messier throughout. Clean it, then ask a question of it and get an answer. For example: which site earns the most revenue per visitor? A good clean-up is what makes that answer trustworthy.
How did it go?
Putting it to workGood to know

Before you use this for real

A few things the tidy example won't show you, worth a quick check on any file you're handed.

The separator isn't always a comma
In much of Europe, a CSV uses a semicolon between columns and a comma as the decimal point, so a file can open scrambled on another machine. If a whole row lands in one cell, that is usually why. In Excel, check File, Options, Advanced, and the separator settings. In Google Sheets, set the locale under File, Settings before importing.
You'll meet close cousins
A TSV (tab-separated values) is the same idea with tab characters instead of commas between the columns. When a tool offers "export to CSV", that is usually the option you want.
Glance before you act
Open the file and look: one header row, one record per line, columns lined up. Thirty seconds here saves a confused answer later.
Wrapping up

What to take with you

  • A CSV is plain text: commas separate columns, line breaks separate rows. That plainness is why AI reads it so easily.
  • The cleaner the data going in, the more you can trust what comes out. Odd dates, stray symbols and extra rows quietly break the answer.
  • Once you can move clean data in and out, everyday jobs like a campaign content plan become something you can do with AI.
Up next
4.1 Structured vs Unstructured Data
Which of your data is ready for AI, and which needs work first.
Open module →
Up next
4.3 JSON and Data Interchange
The other format AI tools pass data around in.
Open module →
Coming soon
6.1 Working with Your Own Documents
Putting cleaned files to work inside AI tools.
★0 of 2 points collected ✓ Module complete. Points saved.