diff --git a/docs/integrate/kafka/docker-python.md b/docs/integrate/kafka/docker-python.md
index e9d251c2..d909438c 100644
--- a/docs/integrate/kafka/docker-python.md
+++ b/docs/integrate/kafka/docker-python.md
@@ -50,7 +50,7 @@ docker compose up -d
 * CrateDB Admin UI: `http://localhost:4200`
 * Kafka broker (inside-compose hostname): kafka:9092
 
-### Create a demo table in CrateDB
+### Create a CrateDB table
 
 The easiest way to do this is through the CrateDB Admin UI at `http://localhost:4200` and execute this using the console:
 
@@ -69,26 +69,30 @@ But this can also be done using `curl`:
 curl -sS -H 'Content-Type: application/json' -X POST http://localhost:4200/_sql -d '{"stmt":"CREATE TABLE IF NOT EXISTS sensor_readings (device_id TEXT, ts TIMESTAMPTZ, temperature DOUBLE PRECISION, humidity DOUBLE PRECISION, PRIMARY KEY (device_id, ts))"}'
 ```
 
-### Create a Kafka topic and send a couple of messages
+### Create a Kafka topic
 
 Creating a Kafka topic can be done in several ways, we are selecting to use
 `docker exec` in this way:
-
 ```bash
 docker exec -it kafka kafka-topics.sh --create --topic sensors --bootstrap-server kafka:9092 --partitions 3 --replication-factor 1
+```
+
+## Process events
 
+### Submit events to Kafka
+```bash
 docker exec -it kafka kafka-console-producer.sh --bootstrap-server kafka:9092 --topic sensors <<'EOF'
 {"device_id":"alpha","ts":"2025-08-19T12:00:00Z","temperature":21.4,"humidity":48.0}
 {"device_id":"alpha","ts":"2025-08-19T12:01:00Z","temperature":21.5,"humidity":47.6}
 {"device_id":"beta","ts":"2025-08-19T12:00:00Z","temperature":19.8,"humidity":55.1}
 EOF
 ```
+Events (messages) are newline-delimited JSON for simplicity.
 
-Messages are newline-delimited JSON for simplicity.
-
-## Data loading
+### Consume events into CrateDB
 
-Create a simple consumer using Python.
+Create a simple consumer application using Python. It consumes events from the
+Kafka topic and inserts them into the CrateDB database table.
 
 ```python
 # quick_consumer.py
diff --git a/docs/integrate/node-red/mqtt-tutorial.md b/docs/integrate/node-red/mqtt-tutorial.md
index d250ee95..f6134921 100644
--- a/docs/integrate/node-red/mqtt-tutorial.md
+++ b/docs/integrate/node-red/mqtt-tutorial.md
@@ -1,5 +1,5 @@
 (node-red-tutorial)=
-# Ingesting MQTT messages into CrateDB using Node-RED
+# Load MQTT messages into CrateDB using Node-RED
 
 :::{article-info}
 ---
@@ -23,19 +23,7 @@ You need:
 2. The [node-red-contrib-postgresql](https://github.com/alexandrainst/node-red-contrib-postgresql) module installed.
 3. A running MQTT broker. This tutorial uses [HiveMQ Cloud](https://www.hivemq.com/).
 
-## Producing data
-
-First, generate data to populate the MQTT topic with Node-RED. If you already
-have an MQTT topic with regular messages, you can skip this part.
-![Screenshot 2021-09-13 at 14.58.42|690x134, 50%](https://us1.discourse-cdn.com/flex020/uploads/crate/original/1X/5722946039148ca6ce69702d963f9f842c4f972c.png){width=480px}
-
-The `inject` node creates a JSON payload with three attributes:
-![Screenshot 2021-09-13 at 14.56.42|690x293, 50%](https://us1.discourse-cdn.com/flex020/uploads/crate/original/1X/8084a53e544d681e79f85d780c621a340a7d0d30.png){width=480px}
-
-In this example, two fields are static; only the timestamp changes.
-Download the full workflow definition: [flows-producer.json](https://community.cratedb.com/uploads/short-url/eOvAk3XzDkRbNZjcZV0pZ0SnGu4.json) (1.3 KB)
-
-## Consuming and ingesting data
+## Provision CrateDB
 
 First of all, we create the target table in CrateDB:
 ```sql
@@ -49,6 +37,20 @@ Store the payload as CrateDB’s {ref}`OBJECT data type
 <crate-reference:type-object>` to accommodate an evolving schema.
 For production, also consider the {ref}`partitioning and sharding guide <sharding-partitioning>`.
 
+## Publish messages to MQTT
+
+First, generate data to populate the MQTT topic with Node-RED. If you already
+have an MQTT topic with regular messages, you can skip this part.
+![Screenshot 2021-09-13 at 14.58.42|690x134, 50%](https://us1.discourse-cdn.com/flex020/uploads/crate/original/1X/5722946039148ca6ce69702d963f9f842c4f972c.png){width=480px}
+
+The `inject` node creates a JSON payload with three attributes:
+![Screenshot 2021-09-13 at 14.56.42|690x293, 50%](https://us1.discourse-cdn.com/flex020/uploads/crate/original/1X/8084a53e544d681e79f85d780c621a340a7d0d30.png){width=480px}
+
+In this example, two fields are static; only the timestamp changes.
+Download the full workflow definition: [flows-producer.json](https://community.cratedb.com/uploads/short-url/eOvAk3XzDkRbNZjcZV0pZ0SnGu4.json) (1.3 KB)
+
+## Consume messages into CrateDB
+
 To ingest efficiently, group messages into batches and use
 {ref}`multi-value INSERT statements <inserts-multiple-values>`
 to avoid generating one INSERT per message:
diff --git a/docs/performance/inserts/tuning.md b/docs/performance/inserts/tuning.md
index d24e6d5e..7dcf5df4 100644
--- a/docs/performance/inserts/tuning.md
+++ b/docs/performance/inserts/tuning.md
@@ -117,7 +117,7 @@ value.
 
 ### Calculating statistics
 
-After loading larger amounts of data into new or existing tables, it is
+After inserting larger amounts of data into new or existing tables, it is
 recommended to re-calculate the statistics by executing the `ANALYZE`
 command. The statistics will be used by the query optimizer to generate
 better execution plans.