Automation you build and host yourself, or automation that reads your messages.
n8n is one of the most flexible automation platforms there is: a source-available, self-hostable canvas where you wire nodes together and drop into JavaScript or Python whenever you need more control. For a technical team that wants to own its infrastructure, few tools give a builder more room. this+that takes a different shape. It reads the messages your team receives, decides what needs to happen, and runs the work, described in plain language rather than wired on a node graph, and grounded in a shared brain.
Two different audiences for automation
n8n is a builder's tool. You assemble a workflow node by node, drop into code when the visual editor runs out of road, and self-host it on infrastructure you control. For a developer who already knows the pipeline she wants, that flexibility is the whole point. this+that works the other way around. It reads your inbox and chat channels, works out what each message needs, and runs the workflow, with the steps described in plain language and grounded in your brain. n8n is built for technical teams who want to build and host their own automation. this+that is built for people buried in messages who will never wire a node graph and need an agent to read them, decide, and act. These are different bets on who automation is for, and plenty of organizations will run both.
Feature by feature
How the two products compare across the things that matter most.
| Topic | n8n | this+that |
|---|---|---|
| Core philosophy | Build it yourself on a node canvas, and drop into code when you need more. You wire the triggers, nodes, and data flow, and the workflow runs the path you built. | Read messages, decide what they need, and run the work that follows. The logic is described in plain language, not wired node by node. |
| Who it is for | Technical teams and developers who want control, the option to write JavaScript or Python, and infrastructure they own. | Non-technical teams buried in messages who need an agent to read them, decide, and act, with no node graph to build. |
| Where work starts | With a trigger you configured in a workflow. n8n is event-reactive once you build and wire the trigger node. | With the messages already arriving in your channels. Nothing to assemble before each one runs. |
| How automations are built | Visual node editor with the ability to drop into a Code node for JavaScript or Python. Powerful and precise, with a real learning curve. | Plain-language workflows. 32 action types across email, the DoBox, calendar, and per-channel sends, with conditional logic, loops, and autonomous sub-agents. |
| AI decision-making | A native AI Agent node and LangChain-based AI nodes you add to a workflow, connecting many model providers. The workflow runs the path you wired around them. | AI decision-making is built into the workflow. Steps can branch on what a message says, what it is asking for, and how urgent it is. |
| Inbox awareness | Not built in. A workflow can be triggered by a new email node, but n8n does not read and triage your inbox proactively. | Core capability. this+that reads every message across email, Slack, Teams, and Google Chat and surfaces what matters. |
| Task layer | No persistent task layer. A workflow can write a row or a card wherever you wire it, but tracking lives in whatever tool you connect. | DoBox. Tasks are auto-created from messages across multiple shared lists, with assignees, priorities, due dates, comments, attachments, and recurring tasks. |
| Knowledge / grounding | Variables and data stores hold values inside a workflow. No team knowledge layer that grounds the automation in what your team knows. | The brain: a team knowledge layer with version history that grounds workflow steps and the AI. It does not yet auto-update from your messages; that is coming. |
| Integrations | A large library of integration nodes, plus HTTP and code nodes for anything with an API. Deep flexibility for a team willing to build. | Deep, native handling of your messaging channels, plus the open MCP standard for tools like GitHub, Notion, HubSpot, Asana, ClickUp, and more. |
| Channels | Trigger on an email or chat event you wire up, but not a unified inbox. No proactive cross-channel reading. | Gmail, Outlook, Slack (personal and team bot), Google Chat, and Telegram read as one stream. Teams ingested. WhatsApp, Instagram, and Messenger coming soon. |
| Hosting and data control | Source-available and self-hostable. You can run it on your own infrastructure and keep data inside your environment. A genuine strength. | A hosted team product. OAuth connections, inference inside AWS with no training on your data, and SOC 2 in progress. |
| Cost model | Self-hosting is free under its fair-code license; you run as much as your servers handle. Managed cloud plans are execution-priced. Strong cost control at scale for teams that self-host. | A hosted subscription, free during beta. Not metered per execution. |
| Maturity | A mature, widely adopted platform with a large community and deep builder features. | Newer product, actively developed. Core automation, inbox, and task features are available now. |
- Raw flexibility. The node canvas, paired with a Code node for JavaScript or Python, lets a developer build almost anything, including logic the visual editor alone cannot express.
- You can self-host it. Running n8n on your own infrastructure keeps your data inside your environment, which matters for teams with strict requirements.
- The integration library is large, and HTTP and webhook nodes reach anything else that has an API.
- At scale, the cost story is hard to beat: self-hosting is free under its fair-code license, so high-volume teams are not billed per run.
- It is a natural home for developers who want to own both the build and the infrastructure it runs on.
- AI and agent nodes are growing fast, with a native AI Agent node for teams building LLM-driven workflows themselves.
- It reads your team's messages and acts on them, rather than waiting for triggers you wire up per workflow.
- Describe a workflow in plain language, with conditional logic, loops, and autonomous sub-agents, and you are done; there is no node graph to assemble or host.
- Because a step can branch on what a message actually says and asks for, the automation is not limited to the fixed conditions you mapped in advance.
- Task extraction comes built in. Every conversation is scanned for requests, commitments, and deadlines, surfaced in DoBox automatically.
- The universal inbox, the team task manager, and the knowledge layer are bundled, not something you build a workflow to populate.
- There is nothing to host, maintain, or learn, which suits teams that do not have a developer to run a platform.
See what's waiting in your inbox
Connect your inbox and get a free analysis of the tasks, follow-ups, and open commitments this+that would surface across your channels.
Free to use · No credit card required