MongoDB $unset Stage

What Is $unset?

$unset is a MongoDB aggregation stage that removes one or more fields from every document in the pipeline. Everything else stays—so it is the easy way to drop passwords, temps, or nested paths before you return results.

$unset = drop these fields, keep the rest

{ $unset: "password" }
{ $unset: [ "password", "tempScore" ] }

nested → "profile.temp"
missing field → no error

vs $project → allow-list / reshape
vs $set → add/overwrite fields
pipeline only → collection unchanged

How to Use $unset in MongoDB

Name the fields to drop, then place the stage where you want them gone.

1. List the field names as a string or an array of strings.
2. Use dot notation for nested paths (for example "profile.temp").
3. Put $unset after you finish using those fields (often after $set / $sort).
4. Remember: this cleans the pipeline output, not the stored collection.

Pseudocode

fieldsToDrop = ["password", "tempScore"]

for each document:
    for each name in fieldsToDrop:
        delete document[name]   // or nested path
    emit document               // other fields unchanged

Cheat Sheet

Keep these facts handy while cleaning pipeline documents.

Live Preview

Start from a student document and drop fields the way $unset would.

$unset: "password"

before → { name, email, password, tempScore, profile }
after  → { name, email, tempScore, profile }

Worked Walkthrough — Hide Password

Asha’s document includes a password. Before we return her profile, we drop that field with $unset.

Document:
  {
    name: "Asha",
    email: "[email protected]",
    password: "secret",
    tempScore: 88
  }

$unset: "password"

Result:
  {
    name: "Asha",
    email: "[email protected]",
    tempScore: 88
  }

password gone · other fields kept

MongoDB Examples

We use one student document. First insert it, then remove password, then remove several fields at once. Run these in mongosh after inserting the sample data once.

db.students.insertOne({
  _id: 1,
  name: "Asha",
  email: "[email protected]",
  password: "secret",
  tempScore: 88,
  profile: { city: "Chennai", temp: true }
});

db.students.find()

{
  _id: 1,
  name: "Asha",
  email: "[email protected]",
  password: "secret",
  tempScore: 88,
  profile: { city: "Chennai", temp: true }
}

{ name, email, password, tempScore, profile }
                 ↓ $unset "password"
{ name, email, tempScore, profile }

db.students.aggregate([
  { $unset: "password" }
]);

{
  _id: 1,
  name: "Asha",
  email: "[email protected]",
  tempScore: 88,
  profile: { city: "Chennai", temp: true }
}

db.students.aggregate([
  {
    $unset: ["password", "tempScore", "profile.temp"]
  }
]);

{
  _id: 1,
  name: "Asha",
  email: "[email protected]",
  profile: { city: "Chennai" }
}

Performance Notes

Dropping fields is inexpensive. Prefer removing large unused fields before heavy stages when documents are wide and the pipeline is long.

Key Takeaways