What is a database? Database vs spreadsheet explained

A database is an organised collection of data stored on a computer so that software can search it, change it and protect it from mistakes quickly. Unlike a spreadsheet, a database splits data into linked tables, lets many people work at once, rejects invalid input and remembers who changed what.

Quick answer

• A database isn’t a program you open - it’s the store that applications use: a web shop, a CRM, an ERP or your internal tool.
• A relational database keeps data in several tables linked by a key, e.g. customers, orders and order lines.
• A database enforces rules: required fields, correct format, no duplicates.
• Excel is great for calculations and analysis; a database is for shared, everyday work on the same data.

What is a database, in plain words?

Picture a filing cabinet: one drawer for customers, one for orders, one for products. Every card has the same layout and its own number, and an order doesn’t copy the customer’s address - it just notes the number of their card.

That is a relational database. Drawers are tables, cards are rows, and the card number is the key that links tables. Through a database management system (e.g. PostgreSQL, MySQL, Microsoft SQL Server or, for small desktop needs, Microsoft Access), an application asks questions like “all orders from customer 42 this year” and gets the answer in a fraction of a second.

A worked example: a small firm’s orders

An illustrative example, not client data. A firm with three staff keeps orders in Excel. Each year it has 120 customers, 600 orders and about 2,000 order lines (one row = one product on an order). Every row also carries the customer name, VAT number and address.

One flat sheet: customer details are typed into each of the 2,000 rows, so 3 × 2,000 = 6,000 cells a year repeating what you already know. The largest customer, with 25 orders, has 80 rows - the same address 80 times.

Three linked tables: customers (120 rows), orders (600 rows, each holding a customer number) and order lines (2,000 rows, each holding an order number). Customer details are entered once: 3 × 120 = 360 cells, about 94% less retyping.

The customer moves. In the sheet you find and fix 80 rows. At 30 seconds a row that’s 40 minutes, and one missed row means an invoice to the old address. In the database you change one cell, and every order shows the new address straight away.

Typos. Assume one manual entry in fifty is slightly different (“12 High St” vs “12 High St.”). In the sheet that’s 2,000 / 50 = 40 rows a year that a customer filter misses, so the customer’s sales report is wrong. In the database you pick the customer from a list, so the mistake can only happen at first entry and is fixed once.

Database vs spreadsheet: a comparison

	Excel spreadsheet	Database
Structure	one flat sheet, data repeats	several linked tables, each fact once
Several users	anyone can overwrite any cell	per-user rights, no colliding edits
Input rules	optional, easy to bypass	required fields, format, no duplicates
History and backups	file versions, if someone keeps them	change log and automatic backups
Size	up to 1,048,576 rows per sheet	practically unlimited for a small firm
Best for	calculations, analysis, one-off lists	shared, everyday work inside applications

The 1,048,576-row limit comes from Microsoft’s Excel specifications and limits page. The firm in the example would hit it in more than 500 years. The problem isn’t size - it’s repetition and several people working on the same data.

How does a database stop wrong input?

A database doesn’t accept whatever you type. You can require that an order can’t exist without a customer, that a VAT number has the right length, that a quantity can’t be negative and that a customer with orders can’t be deleted. When two people sell the last item in stock at the same moment, the database lets only one sale through.

You also get history and backups: the database records who changed a price, and backups run automatically. According to Eurostat, in 2025 45.52% of EU enterprises that pay for cloud services also hosted their database in the cloud (enterprises with 10 or more employees).

When is a spreadsheet no longer enough?

A spreadsheet is enough while one person maintains the data and the sheet is used for calculation or overview. You need a database when the same fact is retyped in several places, when several people work in the same file, or when a wrong entry costs money - like the invoice to the old address in the example.

How to spot that moment is covered in internal tools beyond spreadsheets, and how to move existing data in from Excel to a custom application. A database sits behind an application, not a plain website - see website vs web application. We build applications like this as part of custom software development.

Frequently Asked Questions

Is Excel a database? An Excel sheet can work as a simple flat database: a list where every row has the same columns. It has no linked tables, input rules or access rights, so it isn’t a database in the full sense.

What is Microsoft Access? Access is Microsoft’s program for building small relational databases on a desktop, with forms and reports. It suits one or a few users; web applications and larger teams usually use a server database.

Do I need to know SQL to use a database? No. SQL is the language the application uses to talk to the database. Staff work in the application through forms and buttons, and a developer writes the SQL.

Related Articles

• Internal tools: when spreadsheets and SaaS stop being enough
• From Excel to a custom application: what about the data
• Website vs web application: what is the difference?

Thinking about an application instead of a spreadsheet?

We help you decide whether you need a database, shape the tables around how your business works and build an application your team can use without retyping.

Reach out at [email protected] or via the form on our homepage.