How to Offer AI Data Entry and Spreadsheet Cleanup

• Typing data from PDFs, scans, and forms into spreadsheets
• Removing duplicates and fixing formatting
• Standardising names, dates, phone numbers, and addresses
• Merging lists from different systems
• Preparing data for a CRM or accounting import

Clean data saves clients hours every week. That is what they pay for.

Excel and Google Sheets

Formulas, filters, and built-in AI features such as Copilot and Gemini

AI assistants

ChatGPT, Claude, or Gemini to write formulas and scripts

Python

For large files and repeatable cleaning, written with AI help

• Use AI to read PDFs, invoices, and images into tables
• Ask for a fixed set of columns
• Spot-check extracted rows against the originals

Extract these fields from each invoice into a table: invoice number, date, supplier, total, tax. Leave a cell blank and flag it if a value is unclear. Do not guess.

• Remove exact and near-duplicate rows
• Trim spaces and fix capitalisation
• Convert dates and numbers to one format
• Split or combine columns as needed

• One format for phone numbers and postcodes
• Consistent company and country names
• Clear categories from a fixed list

Write a Google Sheets formula that converts the phone numbers in column C into the format +44 7123 456789, and explain how it works.

• Match records with a unique key, such as an email or ID
• Use XLOOKUP or a script for large files
• Flag records that do not match

• Compare row counts before and after
• Check totals still add up
• Spot-check a random sample by hand
• Use data validation rules to prevent new errors

• Keep the original file untouched
• Write a short change log of every step
• Hand over formulas or scripts so the client can repeat it

Per file

A fixed price based on size and mess

Per batch

For ongoing data entry

Monthly

Regular cleanup and reporting

Ask for a sample of the data before quoting.

• Small businesses moving to a new CRM
• Accountants and bookkeepers with messy client files
• Online stores with product catalogues
• Freelance platforms such as Upwork

• Sign an NDA for sensitive data
• Remove or mask personal data before using AI tools
• Use business AI plans with proper data protections
• Delete client files when the job is done
• Follow data protection laws such as GDPR

• Trusting AI extraction without checking
• Overwriting the original data
• Pasting personal data into consumer AI chats