$match and $group

Filter documents early in a pipeline with $match, and summarize them with $group and its accumulator operators.

$match: Filtering in a Pipeline

$match filters documents using the exact same query syntax as find(), and is almost always the first stage in a pipeline — filtering early reduces how many documents later, more expensive stages need to process.

db.orders.aggregate([
  { $match: { status: "completed", total: { $gte: 50 } } },
]);

$group: Summarizing Documents

$group collapses many documents into one summary document per distinct value of a chosen field, specified via _id.

db.orders.aggregate([
  { $group: { _id: "$status", count: { $sum: 1 } } },
]);

Result

Click Run to see what this code prints.

Accumulator Operators

Operator	Computes
$sum	A running total (use $sum: 1 to simply count documents)
$avg	The average of a numeric field
$min / $max	The smallest / largest value
$push	Collects values into an array
$first / $last	The first / last value encountered in the group (order-dependent)

db.orders.aggregate([
  {
    $group: {
      _id: "$customerId",
      totalSpent: { $sum: "$total" },
      averageOrder: { $avg: "$total" },
      orderIds: { $push: "$_id" },
    },
  },
]);

Grouping by Multiple Fields

Passing an object as _id groups by the combination of multiple fields at once.

db.orders.aggregate([
  { $group: { _id: { customerId: "$customerId", status: "$status" }, count: { $sum: 1 } } },
]);

Common Beginner Mistakes

Putting $match after expensive stages instead of first

Filtering as early as possible reduces the number of documents every subsequent stage has to process — always $match first when the filter doesn't depend on a later computed value.

Forgetting the $ prefix when referencing a field

{ $sum: total } is invalid — field references inside aggregation expressions need a leading $: { $sum: "$total" }.

FAQs

Can I use $match more than once in a pipeline?

Yes — a second $match after a $group is common, filtering on the newly computed summary fields (the equivalent of SQL's HAVING clause).

What happens if I group by a field every document shares the same value for?

You get a single summary document covering the entire collection — a common pattern for computing an overall total or average.

Summary

$match filters early, and $group with accumulator operators summarizes documents into computed results. Next, you'll learn $project, $sort, and $limit for reshaping and ordering pipeline output.

Next Lesson →