How Much Does AWS IoT Cost for 1,000 Devices?
A practical way to estimate AWS IoT cost for 1,000 devices by looking at message frequency, payload size, rules, storage, logs, dashboards and operating behaviour.
Vishnu Patekar
Founder & CEO, Delighteck - Product Engineering · Reviewed by Delighteck Engineering
The cost of AWS IoT for 1,000 devices is not decided by the number 1,000.
It is decided by what those devices do every day.
A thousand devices sending one small reading per hour are cheap to operate. A thousand devices sending data every few seconds, triggering rules, writing to databases and producing logs can become a very different bill.
The calculation starts at the device
Before opening a pricing calculator, write down:
- How often does each device publish?
- How large is each message?
- Does the device stay connected all day?
- Does every message trigger one rule or many?
- Are commands sent back to the device?
- Are shadows used for state or configuration?
- How long is raw data stored?
- What dashboards, alerts and logs are required?
This is where cost control begins.
Three example patterns
| Device behaviour | Messages per device | What it usually means |
|---|---|---|
| Hourly reporting | 24 per day | Low-duty telemetry, meters, periodic sensors |
| Every 5 minutes | 288 per day | Monitoring where trend matters |
| Every 10 seconds | 8,640 per day | Near-real-time telemetry, much higher cloud load |
For 1,000 devices, those become 24,000, 288,000 or 8.64 million device-to-cloud messages per day before retries, commands, rules, shadows or logs.
This is why "1,000 devices" is not enough information.
Message size matters
Payload design affects cost and downstream performance.
A compact JSON or binary message with only useful fields is different from a verbose payload that repeats metadata every time. It is common to start with readable JSON during prototyping, then tighten the message format before scale.
Do not optimise too early. But do not ignore payload design until after deployment.
Rules, storage and dashboards
An AWS IoT architecture rarely stops at IoT Core.
Messages often flow into rules, Lambda, DynamoDB, Timestream, S3, OpenSearch, alerts or dashboards. Each downstream service has its own cost model.
The expensive surprise is often not the MQTT connection. It is high-frequency data multiplied by rules, writes, logs and dashboards.
Device Shadow and command patterns
Device Shadow can be useful for configuration and last-known state. It can also be overused.
If every telemetry update becomes a shadow update, cost and complexity can increase. For many products, telemetry, configuration and command topics should be designed separately.
The official AWS docs also note additional metering details for rules and shadow-related behaviour on the AWS IoT additional metering details page.
A practical estimation worksheet
| Question | Example answer |
|---|---|
| Number of devices | 1,000 |
| Reporting interval | Every 5 minutes |
| Payload size | 1 KB |
| Messages per day | 288,000 |
| Rules per message | 1 ingestion rule |
| Storage | 12 months aggregate, 30 days raw |
| Commands | Rare configuration updates |
| Dashboards | Internal operations dashboard |
| Logs | Error logs only after pilot |
Once this is clear, pricing becomes a calculation instead of a guess.
How to keep AWS IoT cost under control
- Send only useful data.
- Aggregate at the edge where it makes engineering sense.
- Separate telemetry, state and commands.
- Avoid triggering multiple rules for every message unless needed.
- Store raw data only as long as it has value.
- Use logs deliberately; full debug logs from every device can become noisy.
- Revisit the message interval after pilot data.
Related reads
Relevant Delighteck pages: AWS IoT Engineering, Cloud and IoT, Device Cloud.
Bring the engineering problem.
You do not need every requirement figured out before speaking with us. Tell us what you are building, where you are today and what is proving difficult.
Discuss your product