A pharmacy owner sends you a file with three years of sales. Dates are typed five different ways, product names are spelled differently each week, and some rows are just "bal". She asks, "How much will you charge to fix it?" If you answer with a random number, you'll either lose the job or spend a week on a ₦10,000 task.
This guide gives you a way to price spreadsheet cleanup so you get paid fairly, the client knows what they're buying, and you don't get trapped in endless "small changes". All the naira figures below are hypothetical. Set your own based on your city, your experience and what the client can afford.
Why cleanup jobs are easy to underprice
Clients see a spreadsheet and think "just tidy it". You see hidden work: finding duplicates, guessing what abbreviations mean, chasing the owner for answers, and checking that totals still match after you change things.
The biggest cost is almost never typing. It's the unknowns. A file with 500 tidy rows and a file with 5,000 chaotic rows can look the same in a screenshot, which is why you should never quote before seeing the real file.
Step 1: Inspect the file before you quote
Ask for a copy of the file, or a sample of at least 200 rows. If the data is sensitive (customer names, phone numbers, staff salaries), ask the client to remove or replace those columns for the sample. Don't take more personal data than you need.
Then run this quick audit, which takes 15 to 30 minutes:
- Count rows and columns. More rows rarely means much more work. More columns and more sheets usually do.
- Check dates. Are they real dates, or text like "3rd Oct" and "10/3/26"? Mixed formats are slow to fix. Also note that 10/3 means different things depending on day-first or month-first habits.
- Check names and categories. Sort a column such as product or customer. Count how many spellings refer to the same thing.
- Look for blanks and notes in cells. Text like "paid half" inside an amount column means you'll need rules and client decisions.
- Look for formulas. Existing formulas that point to wrong ranges or other files can break when you clean.
- Ask what the file is for. A file for monthly profit tracking needs a different finish from one for a bank loan application or tax records.
Step 2: Sort the job into a tier
Most small-business cleanups fit one of three levels. Naming them helps the client choose and stops "tidy up" from meaning everything.
| Tier | What's included | Typical file |
|---|---|---|
| Basic tidy | Consistent dates and number formats, removed blank rows and obvious duplicates, clear column headings, one clean sheet | One sheet, under 1,000 rows, mostly consistent |
| Standard cleanup | Basic tidy plus standardised names and categories, a short list of rules you applied, a flagged list of rows you couldn't resolve, simple totals by month or product | One to three sheets, a few thousand rows, mixed spellings |
| Rebuild | Standard cleanup plus a new structure: a data entry sheet with dropdowns, a summary sheet, and a short how-to for the owner | Many sheets, years of data, or data copied from WhatsApp, paper or several staff |
If your client's messages are the source of the mess, see how to turn a messy WhatsApp order list into a clean sales report. That type of job usually lands in standard or rebuild.
Step 3: Estimate your hours, then add a buffer
Break the work into pieces and guess hours for each. Here's a hypothetical standard cleanup for a pharmacy file with 4,000 rows:
- Audit and call with the owner: 1.5 hours
- Fixing dates and number formats: 1 hour
- Standardising about 300 product names into about 180 real products: 3 hours
- Resolving unclear rows and getting answers from the owner: 1.5 hours
- Monthly summary and checks that totals match the original: 2 hours
- Short notes on what you changed: 1 hour
That's 10 hours. Add a 25 to 30 percent buffer for delays and surprises and you get about 13 hours. If you value your time at ₦3,000 an hour at your current level, the price is ₦39,000. You might round it to ₦40,000.
Beginners often hate the idea of an hourly rate because they're slow. That's fine. Your first jobs may take longer than the estimate, so use the hourly number as a floor. If you've never done this, you can still charge less for the first one or two jobs in exchange for a testimonial, but say the normal price in your quote so the discount is visible.
Step 4: Pick a pricing method
Fixed price per tier
Best for most people. The client knows the cost up front, and you keep the reward if you work faster. Use it once you've audited the file.
Hourly with a cap
Good when the file is unclear or the owner is slow to respond. Say "₦3,000 per hour, estimated 10 to 13 hours, I'll tell you before passing 13." The cap protects the client's trust.
Per row or per sheet
Risky. A per-row price punishes you when rows are messy and rewards you for nothing when they're clean. Use it only for repeated, uniform work, such as monthly cleanup of the same file format.
Monthly retainer
If the owner keeps sending new messy data, offer a monthly fee for cleaning and a basic report. This is often the better business, because one client can give steady income. Start only after one successful one-off job.
Step 5: Write the quote so it prevents arguments
A short message is enough. Include:
- What you will deliver (for example: one cleaned sheet, a monthly summary, a list of rows needing the owner's decision).
- What you won't do (for example: re-entering missing data from paper, building a full app, fixing data entered after the delivery date).
- The price and what's included in revisions, such as one round of fixes within seven days.
- The timeline, counted from when you receive the full file and the answers to your questions.
- Payment terms.
For a first-time client with no track record, ask for 50 percent upfront and 50 percent on delivery. For a very small job, you can ask for full payment before sending the final file while sharing a preview of the first 50 rows. If you work through a freelance platform, follow its payment rules instead; our guide to getting started on freelance platforms from Nigeria covers the basics.
Where AI tools help and where they don't
General AI chat tools can suggest formulas, explain error messages, and draft rules for standardising names. Other types of tools, such as spreadsheet add-ins or data-cleaning software, can group similar text entries. These can cut the time on step 3 substantially.
But three cautions apply:
- Customer data. Don't paste real names, phone numbers or account details into a tool without the client's permission and a check of how that tool handles data. Use fake sample rows when asking for formula help.
- Silent errors. An automated tool can merge two different products into one. Always compare totals before and after, and spot-check at least 50 random rows.
- Pricing. If AI saves you five hours, you still deliver a result, not hours. Price by the value and risk of the work, not only by your time. Dropping your price to match your speed means you earn less as you improve.
For the skills that make these jobs easier, read spreadsheet skills that get small-business clients paying.
Common mistakes to avoid
- Quoting from a description. "It's just 2,000 rows" tells you nothing. Look at the file.
- Overwriting the original. Always keep the untouched file and work on a copy. Name versions clearly, like "sales_original" and "sales_clean_v1".
- Guessing what unclear entries mean. Flag them in a separate list for the owner. A wrong guess in financial records is worse than a blank.
- Skipping a change log. A short list of rules ("all dates converted to day-month-year; 'Paracetamol 500', 'paracetamol 500mg' merged") lets the client trust the result.
- Unlimited revisions. Define one round of fixes, then charge an hourly rate for extra requests.
- Ignoring power and data costs. If you have long outages, build the buffer into your timeline, and save work often. Large files in cloud spreadsheets need steady data; a desktop file you can work on offline may be safer.
How the price changes by situation
- Corner shop or market trader: Budgets are small. Offer the basic tidy tier at a low fixed price, then pitch a monthly update.
- Registered company, school or clinic: They often care more about accuracy and confidentiality. Price higher, offer a short written agreement, and be ready to sign an NDA.
- Clients abroad: They may quote in dollars and compare you with local freelancers. Keep your price based on your effort and skills, and be clear about currency and payment fees before you start.
- Urgent jobs: A rush request, such as a file needed for a loan or audit meeting in two days, can carry a clear surcharge, for example 25 to 50 percent. State it openly.
Frequently asked questions
Should I put my prices on my portfolio or social page?
A "starting from" price per tier works well, because it filters out clients who can't afford you. Keep the final number dependent on the file audit.
What if the client says the file is simple and I'm overcharging?
Offer the basic tier with a smaller scope, or propose a paid audit. Many freelancers charge a small fee for the audit and deduct it from the project price if the client goes ahead.
Can I do this job with no formal training?
Yes, if you can already use sorting, filtering, text-to-columns style functions, lookup formulas and pivot tables, and you're honest about your level. Practise on a messy public sample or your own invented data first.
How many jobs before I raise my price?
A practical rule is to raise the rate after every three to five completed jobs, or whenever you have more requests than you can handle.
Next step: Make up a messy 200-row sample file of your own, time yourself cleaning it, and write the quote you'd send for it. That gives you a real hours figure to work from on your first live enquiry.
Results from any side hustle or online work vary widely, and nothing here is a promise of income. Be wary of anyone who asks you to pay upfront to start earning.