Update subgraph.md

docs/new/how-to-guides/sinks/subgraph/subgraph.md

@@ -5,13 +5,14 @@ There are two ways of making Substreams sink to a subgraph:
 The subgraph will read the `EntityChanges` object and consume the data.
 - Use the [**Substreams triggers**](./triggers.md) to consume Substreams Protobuf directly inside your subgraph.
 
-## What Option To Use
+## Which Option To Use
 It is really a matter of where you put your logic, in the subgraph or the Substreams.
 
 - [Substreams Triggers](./triggers.md): Consume from any Substreams module by importing the Protobuf model through a subgraph handler and write all your transformations using AssemblyScript. This method creates the subgraph entities directly in the subgraph.
-- [Substreams Graph-Out](./graph-out.md): By writing more of the logic into Substreams, you can consume the module's output directly into `graph-node`. You will create the subgraph entities in the Substreams and the subgraph will read them. 
+- [Substreams Graph-Out](./graph-out.md): By writing the subgraph entities into Substreams, you can consume the module's output directly into `graph-node`. You will create a [graph_out module](./graph-out.md) that emits an [EntityChanges](https://github.com/streamingfast/substreams-sink-entity-changes/blob/develop/proto/sf/substreams/sink/entity/v1/entity.proto#L11) Protobuf.
+The subgraph will read the `EntityChanges` object and consume the data directly.
 
 Having more of your logic in Substreams benefits from a parallelized model and a cursor to [never miss data](../../../references/reliability-guarantees.md), whereas triggers will be linearly consumed in `graph-node`.
 
 
-<figure><img src="../../../.gitbook/assets/consume/service-subgraph.png" width="100%" /></figure>
+<figure><img src="../../../.gitbook/assets/consume/service-subgraph.png" width="100%" /></figure>