> ## Documentation Index
> Fetch the complete documentation index at: https://sequinstream.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# How to stream Postgres to Meilisearch

> Build real-time search indexes with Postgres change data capture (CDC) and Meilisearch. Learn to keep your search collections in sync with your database.

This guide shows you how to set up Postgres change data capture (CDC) and stream changes to Meilisearch using Sequin.

With Postgres data streaming to Meilisearch, you can build real-time search experiences, maintain up-to-date search indexes, and provide fast, typo-tolerant search for your users.

By the end of this how-to, you'll have a configured Meilisearch instance that stays in sync with your database.

<Tip>
  This is the how-to guide for streaming Postgres to Meilisearch. See the [quickstart](/quickstart/meilisearch) for a step-by-step walkthrough or the [reference](/reference/sinks/meilisearch) for details on all configuration options.
</Tip>

## Prerequisites

If you're self-hosting Sequin, you'll need:

1. [Sequin installed](/running-sequin)
2. [A database connected](/connect-postgres)

## Basic setup

### Prepare your Meilisearch instance

Sequin converts Postgres changes and rows into JSON documents and sends them to your Meilisearch index.

You'll need to have a Meilisearch instance running and accessible to Sequin. You can run Meilisearch locally for development or use a hosted version.

<Tip>
  If you're developing locally, you can connect Sequin to a local Meilisearch instance.
</Tip>

### Mapping Postgres tables to Meilisearch indexes

From the [Meilisearch docs](https://www.meilisearch.com/docs/learn/core_concepts/indexes/):

> In Meilisearch, each index contains documents with the same structure.

Sequin recommends creating one sink per Postgres table. Each sink will stream changes from a single Postgres table to a single Meilisearch index. This ensures schema consistency across indexed documents.

For large-scale applications, you may consider sharding indexes—e.g., sharding a `users` index by region into `users_us`, `users_eu`, `users_asia`, etc.

You can use Sequin's [filters](/reference/filters) to accomplish this. For example, to shard `users` by region, create a sink per region with filters to route the correct rows to the corresponding Meilisearch index.

### Advanced: Using routing functions

Sequin supports [routing functions](/reference/routing) for dynamically routing rows and performing advanced operations. With routing functions, you can:

* Route rows from one table to multiple Meilisearch indexes
* Perform function-based updates (increments, array modifications, etc.)
* Apply conditional logic for complex routing scenarios

See the [Meilisearch sink reference](/reference/sinks/meilisearch#routing) for detailed examples.

## Create Meilisearch sink

Navigate to the "Sinks" tab, click "Create Sink", and select **Meilisearch Sink**.

### Configure the source

<Steps>
  <Step title="Select source table or schema">
    Under "Source", select the table or schema you want to stream data from.
  </Step>

  <Step title="Specify filters">
    Choose whether to include `insert`, `update`, and/or `delete` actions. For full sync, select all three.

    For example, if you uncheck `deletes`, rows deleted from Postgres won't be removed from the Meilisearch index.

    You can also use [filters](/reference/filters) to only include certain rows—e.g., `WHERE in_stock = true` for product availability or for sharding by tenant.
  </Step>

  <Step title="Specify backfill">
    Optionally trigger a [backfill](/reference/backfills) to populate your Meilisearch index with existing data from your table. This is optional at sink creation and can be done later.
  </Step>

  <Step title="Transforms">
    Meilisearch requires a defined schema with:

    1. A unique document primary key (string or integer).
    2. Searchable and filterable fields.

    Use a [functional transform](/reference/transforms#function-transform) to convert Postgres rows to Meilisearch documents:

    ```elixir theme={null}
    def transform(action, record, changes, metadata) do
      %{
        id: record["id"],
        name: record["name"],
        description: record["description"],
        price: record["price"],
        in_stock: record["in_stock"]
      }
    end
    ```

    For dynamic schemas or minimal transformation:

    ```elixir theme={null}
    def transform(action, record, changes, metadata) do
      record
    end
    ```

    Or explicitly assign the `id`:

    ```elixir theme={null}
    def transform(action, record, changes, metadata) do
      Map.put(record, "id", record["product_id"])
    end
    ```
  </Step>

  <Step title="Specify message grouping">
    Keep the default setting to ensure updates to the same row are processed in order. This helps maintain consistent state in Meilisearch.
  </Step>
</Steps>

### Configure delivery

<Steps>
  <Step title="Specify batch size">
    The right value for "Batch size" depends on your data and your requirements.

    Meilisearch supports bulk document addition via the `/documents` endpoint. A good starting point is 100–1000 documents per batch. Very large batches (10k+) may impact memory or throughput depending on your instance size.
  </Step>

  <Step title="Select the Meilisearch index you created">
    Under "Meilisearch Index", select the index you want to stream data to.
  </Step>

  <Step title="Create the sink">
    Give your sink a name, then click "Create Meilisearch Sink".
  </Step>
</Steps>

## Verify & debug

To verify that your Meilisearch sink is working:

1. Make some changes in your source table.
2. Verify that the count of messages for your sink increases in the Sequin web console.
3. Search in your Meilisearch index to confirm the document changes. (You can use the [meilisearch-ui](https://github.com/riccox/meilisearch-ui) for a simple UI to interact with your Meilisearch indexes/tasks)

If documents don't seem to be flowing:

1. Click the "Messages" tab to view the state of messages for your sink.
2. Click any failed message.
3. Check the delivery logs for error details, including the response from Meilisearch.

## Advanced use cases

### Using function updates for counters and aggregations

Meilisearch's function update feature, combined with Sequin's routing functions, enables powerful use cases like maintaining counters or aggregated values without re-indexing entire documents.

For example, if you have a `products` index and want to update view counts or inventory levels:

```elixir theme={null}
def route(action, record, changes, metadata) do
  case metadata.table_name do
    # When a product is viewed, increment its view counter
    "product_views" ->
      %{
        action: :function,
        index_name: "products",
        filter: "id = #{record["product_id"]}",
        function: "doc.view_count = doc.view_count + 1"
      }

    # When inventory changes, update the stock quantity
    "inventory_transactions" ->
      %{
        action: :function,
        index_name: "products",
        filter: "sku = '#{record["sku"]}'",
        function: "doc.stock_quantity = doc.stock_quantity + context.quantity_change",
        context: %{quantity_change: record["quantity_change"]}
      }

    # Default behavior for products table
    "products" ->
      %{index_name: "products", action: if(record["deleted_at"], do: :delete, else: :index)}
  end
end
```

This approach is much more efficient than re-indexing the entire product document for simple counter updates.

## Next steps

Assuming you've followed the steps above for your local environment, "[How to deploy to production](/how-to/deploy-to-production)" will show you how to deploy your implementation to production.