6. Inserting documents

So far you have only queried the imported data. It's time to learn how to create your own documents.

Inserting a document

To insert a single document use insertOne:

db.listings.insertOne({
    name: "My apartment in Barcelona",
    neighbourhood_group_cleansed: "EIXAMPLE",
    room_type: "Entire home/apt",
    accommodates: 4,
    price: "$75.00"
})

MongoDB returns an object with the automatically generated _id:

{
    acknowledged: true,
    insertedId: ObjectId("507f1f77bcf86cd799439011")
}

Inserting multiple documents

To insert several documents at once use insertMany:

db.listings.insertMany([
    {
        name: "Private room in Gràcia",
        neighbourhood_group_cleansed: "EXTRAMURS",
        room_type: "Private room",
        accommodates: 2,
        price: "$40.00"
    },
    {
        name: "Studio in Ciutat Vella",
        neighbourhood_group_cleansed: "CIUTAT VELLA",
        room_type: "Entire home/apt",
        accommodates: 2,
        price: "$65.00"
    }
])

MongoDB returns the IDs of all the inserted documents.

Automatic collections

If the collection does not exist, it will be created automatically when you insert the first document:

// This collection does not exist yet
db.mis_propiedades.insertOne({ name: "My first property" })
// Now it exists

The _id field

Each document must have a unique _id. If you don't provide it, MongoDB automatically generates an ObjectId:

// MongoDB generates the _id automatically
db.listings.insertOne({ name: "Beach house" })

// You can provide your own _id
db.listings.insertOne({
    _id: "mi-id-personalizado",
    name: "Downtown apartment"
})

Important: The _id must be unique. If you try to insert a document with an _id that already exists, the operation will fail.

Data validation

MongoDB does not automatically validate your data. You can insert documents with different structures in the same collection:

// All valid in the same collection
db.pruebas.insertMany([
    { name: "Ana", age: 28 },
    { name: "Carlos", city: "Madrid" },  // No age field
    { email: "laura@example.com" }        // No name or age fields
])

This is flexible, but you must be careful with the consistency of your data.

Solutions