What is MongoDB 8? Documents and Collections

What you will learn

By the end of this lesson you will know:

• What MongoDB is and what problem it solves
• What a document and a collection are
• How MongoDB stores data compared to a spreadsheet
• Where MongoDB is used in real projects

You do not need to install anything yet. Just read along and look at the examples.

MongoDB in one sentence

MongoDB is a database that stores your data as documents that look like JSON.

A database is a place where an application keeps its data so it is still there tomorrow: user accounts, blog posts, orders, chat messages, and so on.

Most older databases (like MySQL or PostgreSQL) store data in tables made of rows and columns, a bit like an Excel sheet. MongoDB takes a different approach. It stores each record as a document, which is a set of key: value pairs, just like a JavaScript object.

Think of it like this: A table-based database is like a filing cabinet where every form must have exactly the same boxes. MongoDB is like a folder of index cards. Each card describes one thing, and cards can have slightly different details written on them.

What is a document?

A document is a single record. Here is a document that describes one user:

{
  _id: ObjectId("6718a2f1c3b4e5d6f7a8b9c0"),
  name: "Asha Roy",
  email: "asha@example.com",
  age: 24,
  isMember: true,
  hobbies: ["painting", "cycling"],
  address: {
    city: "Kolkata",
    country: "India"
  }
}

Let’s break it down:

Notice how natural this is. If you have written JavaScript before, this looks exactly like an object. There is no need to split the address into a separate table or the hobbies into another table. Everything about Asha lives together in one place.

Note: On disk MongoDB stores documents in a format called BSON (Binary JSON). BSON is faster to read and supports extra types like dates and ObjectId. You will learn about it in the BSON Types and Document Structure lesson. For now, just think “JSON”.

What is a collection?

A collection is a group of documents. If a document is one index card, a collection is the box that holds all the cards of the same kind.

For example, an online shop might have these collections:

• users – one document per customer
• products – one document per item for sale
• orders – one document per order

And a database is simply a group of collections. So the structure looks like this:

MongoDB server
└── shop            (database)
    ├── users       (collection)
    │   ├── { name: "Asha", ... }      (document)
    │   └── { name: "Rahul", ... }     (document)
    ├── products    (collection)
    └── orders      (collection)

If you know SQL, here is a quick translation:

Documents in a collection can differ

In a SQL table, every row must have the same columns. In MongoDB, two documents in the same collection can have different fields. This is called a flexible schema.

// Both documents can live in the same "products" collection
{ name: "Notebook", price: 49, pages: 200 }
{ name: "Headphones", price: 1999, color: "black", wireless: true }

A notebook has pages, headphones have color and wireless. MongoDB is happy to store both side by side.

Tip: Flexible does not mean messy. In real apps you still keep documents in a collection similar to each other. Later in the course you’ll learn schema validation, which lets you add rules when you want them.

A first taste of MongoDB code

You talk to MongoDB using commands. The most common tool is mongosh, the MongoDB Shell. Here is what saving and reading a document looks like:

// Save a new user into the "users" collection
db.users.insertOne({ name: "Asha Roy", age: 24 })

// Find every user who is 18 or older
db.users.find({ age: { $gte: 18 } })

Don’t worry about the details yet. Just notice that the commands look like JavaScript and the data looks like JavaScript objects. That is a big reason MongoDB feels friendly to web developers.

Why do developers like MongoDB?

1. It matches how you code. Your app already works with objects. MongoDB stores objects, so there is less converting back and forth.
2. It is flexible. You can add a new field to new documents without changing a table structure first.
3. It scales. MongoDB can spread data across many servers (called sharding) when your app grows.
4. It has powerful queries. You can filter, sort, group, join, and even run full-text and AI vector search.
5. It runs anywhere. On your laptop, on your own server, or in the cloud with MongoDB Atlas.

What’s new in MongoDB 8?

This course uses MongoDB 8 (the 8.x line, up to 8.3), which is what you get today when you install MongoDB Community on your own computer. MongoDB 9.0 was released in late 2026 and is rolling out on Atlas first, so you may see it mentioned. As a beginner you do not need to memorise new features, but it is good to know that version 8 brought:

• Big speed improvements for everyday reads and writes
• Query settings, a way to tell the database how to run a specific query
• A more powerful bulkWrite command that can write to many collections at once
• Better Queryable Encryption, so sensitive data can stay encrypted while you search it
• Newer releases also added hybrid search tools like $rankFusion and $scoreFusion

Everything you learn in the beginner lessons works the same way in older versions too.

Where is MongoDB used?

• Web and mobile apps – user profiles, posts, comments, shopping carts
• Content management – blogs and news sites where each article has different fields
• Real-time analytics – counting clicks, views and events
• IoT – storing readings from sensors using time series collections
• AI apps – storing text and its vector embeddings for semantic search

Common beginner mix-ups

• “MongoDB has no structure.” It does: databases → collections → documents → fields. It just doesn’t force every document to look identical.
• “MongoDB stores JSON text.” It stores BSON, a binary version of JSON with extra types.
• “NoSQL means no queries.” MongoDB has a very rich query language. “NoSQL” only means it does not use SQL tables.

Practice

Try writing these on paper or in a text editor. There is no wrong place to practise.

1. Write a document for a book with a title, an author, a year, and a list of genres.
2. Which would you call the collection that holds all book documents?

// 1. One book document
{
  title: "The Alchemist",
  author: "Paulo Coelho",
  year: 1988,
  genres: ["fiction", "adventure"]
}

// 2. A good collection name is plural and lowercase: "books"

Key takeaways

• MongoDB is a document database. It stores records as JSON-like documents.
• A document is one record, made of fields (key: value pairs).
• A collection groups similar documents. A database groups collections.
• Documents can contain arrays and other documents, so related data stays together.
• This course uses MongoDB 8, but the basics are the same across versions (including the newer 9.0).

Next lesson: SQL vs NoSQL: When to Use MongoDB