Take full controlof your IoT back end

TagoIO connects to 34+ networks through 650+ ready-to-use connectors, so most devices send data without custom parsing. LoRaWAN is supported natively through network server integrations including The Things Stack, Everynet, Actility, Loriot, Helium, Senet, machineQ, and Orbiwise. Cellular devices on NB-IoT and LTE-M, Sigfox, satellite backhaul, and MQTT or HTTPS devices connect directly with a device token. TagoTiP, our open telemetry protocol, adds UDP, TCP, HTTP, and MQTT transports for constrained hardware. New connectors are added as partners ship devices, and the connector library is open source on GitHub.

No. TagoIO is the IoT application platform that sits on top of a LoRaWAN network server (LNS). The LNS handles the radio side: device joins, deduplication across gateways, and adaptive data rate. TagoIO receives the uplinks from the LNS, decodes the payload, stores the readings, runs your Analysis and Actions, and sends downlinks back through the same LNS. You keep your existing LNS, whether that is The Things Stack, Everynet, Actility, Loriot, or another provider. If you need to run your own network server, ChirpStack is available as a middleware inside TagoDeploy instances.

Yes. Any device that can reach one of the supported networks or speak MQTT or HTTPS can send data to TagoIO today. If no connector matches your hardware, write a Payload Parser in JavaScript to turn the raw bytes into named variables; the open-source decoders repository on GitHub has hundreds of parsers to start from. The Live Inspector shows each packet as it arrives, so you debug the parser against real traffic. You can also send a sample payload from the Admin with the device's own credentials before the hardware ships, and the parser, dashboards, and alerts run exactly as they will in production.

Yes, and the data is yours. Export any device's readings as CSV from the Admin, from dashboard widgets, or through the API, with filters so you only pull what you need. Actions forward new readings in real time to your own HTTP endpoint or MQTT broker, so a copy can land in your database as it arrives. Profile Backups save your devices, dashboards, Analysis scripts, Actions, connectors, users, and settings as JSON, kept for 12 months, and you can schedule them with an Action. CSV import loads historical data into Mutable and Immutable devices. If you want the database itself rather than a copy, TagoDeploy runs a dedicated database for your instance, with its machine size, read replicas, and backup schedule under your control, and with TagoIO Build we design the system around your own database. When an account closes, TagoIO deletes its data within 30 days.

TagoIO runs on Amazon Web Services. Multi-tenant accounts are hosted in the United States (us-east-1, Virginia) by default, with a European option in eu-west-1 (Ireland) for GDPR data residency. With TagoDeploy you run a dedicated, single-tenant instance in any of 12+ AWS regions, so the data stays in the country your contract requires. If the data cannot leave your own infrastructure, TagoIO Build delivers the system in your cloud (AWS, Azure, Google Cloud, or another provider) or on-premises in your datacenter or plant. Data is encrypted with TLS 1.2+ in transit and AES-256 at rest, and TagoIO has held ISO 27001 certification since 2023.

Yes. The multi-tenant platform runs on AWS, TagoDeploy gives you a dedicated instance in the AWS region you choose, and TagoIO Build goes further: we take the scope of your operational problem and deliver a complete, working IoT system, deployed where you need it. That can be a TagoDeploy instance, your own cloud account on AWS, Azure, Google Cloud, or another provider, or on-premises in your datacenter or plant. The data structure is built around the database your team specifies, so the readings land in a store you already run, query, and back up. TagoIO Build is quoted at a fixed price for a closed scope, with support as a separate annual contract. If you prefer to build and operate the system yourself, the multi-tenant platform or TagoDeploy is the better fit.

Yes. Everything you can do in the Admin is available through the REST API, with SDKs for JavaScript and Python and an MQTT broker for devices and services that prefer publish and subscribe. Actions act as outbound webhooks, calling your endpoint when data arrives or a condition is met. Analysis runs your own Node.js or Python code on the platform on a schedule or in response to data, which is where most ERP, CRM, and ticketing integrations live. The same API powers our integrations with partners such as HubSpot and Jira.

Yes. TagoAI is the AI assistant built into the TagoIO Admin. It reads your device payloads, writes and fixes Analysis code, builds and edits dashboards from a plain-language request, and answers questions grounded in the official documentation. It runs in Read Only, Edit, or Full Access mode, acts only within your own permissions, and logs every change in the Audit Log. Outside the Admin, the open-source TagoIO MCP Server connects your account to AI assistants and coding tools such as Claude, ChatGPT, Cursor, and GitHub Copilot, so you can query live device data in natural language or generate Analysis scripts with knowledge of your real devices. You can use the TagoIO default model or bring your own provider (OpenAI, Anthropic, AWS Bedrock, Google Gemini, OpenRouter). Your data is not used to train models, and TagoAI is off by default on EU profiles.

The TagoIO SLA commits to 99.9% monthly uptime, which allows no more than about 43 minutes of unavailability per month, with service credits of 10% or 25% when we miss it. The SLA covers TagoDeploy instances and accounts with SLA coverage. All customers can watch availability live at status.tago.io, which reports each region (US-E1, EU-W1) and service (API, Analysis, Actions, MQTT broker, LoRaWAN middlewares) with a 90-day uptime history and incident reports, and you can subscribe to updates by email, RSS, webhook, or Slack. When something goes wrong, you will know about it there first.

Every account gets the documentation, the community forum, and the Developer support plan with responses under 24 hours. Production accounts can add the Professional plan, which brings critical-impact response under 4 hours, or the Enterprise plan, with critical-impact response under 2 hours and escalation procedures for business-critical systems.