What I like best about n8n is its flexibility and the balance it strikes between a visual workflow builder and the ability to add custom logic when needed. The drag-and-drop interface makes automation easy to build, while the large library of integrations and API support lets me connect different tools without much effort. Performance has been reliable even for multi-step workflows, and the option to self-host provides excellent value and control over data. I also appreciate the growing AI capabilities, which make it easier to automate repetitive tasks and build intelligent workflows without starting from scratch. Review collected by and hosted on G2.com.

While n8n is very powerful, the learning curve can be steep for users who are new to workflow automation or APIs. Some advanced workflows require JavaScript or a deeper understanding of data transformations, which can reduce the benefits of the no-code experience. The interface can also become difficult to manage as workflows grow in size and complexity. Although the documentation is helpful, more guided onboarding, debugging tools, and native integrations for niche services would make building and maintaining complex automations even easier. Review collected by and hosted on G2.com.

The best part is that its opensource (which means its almost totally free) and the community modules built around it is massive.. Plus The mcp integration it has is great. I literally connect this with my codex and ask it to build the entire n8n workflow to me and it does amazing job.

The Ui is good but I still like the UI of make.com better than this but overall not much complain here.

Regarding Performance, I self host and use it. I hosted this on a $5 vps of netcup and it works pretty well for my use case and never had a problem. I think there can be problem when there are many tasks in queue and for parallelism but for major tasks, there is no issue at all

The good part of n8n is the community it has. I can find tons of people in reddit and discord to support me when I am stuck in any issue. Onboaridng is also much smoother. I would suggest the first task to do is create an api and connect this with codex/claude code and see the magic happen from there Review collected by and hosted on G2.com.

Make and zapier makes it much easier to connect this with facebook ads but while in case of n8n, its not that easy.

Plus make.com provides a free starter plan to try this out. I wish n8n had some very basic free plan (on cloud) so that it can be easy for beginners to get started with. Review collected by and hosted on G2.com.

The multiple options to automate tasks. And being a low-code platform for this, as code nodes open a lot of options to create more efficient workflows and connect to other apps that don't have native nodes. Review collected by and hosted on G2.com.

We use the cloud version, and I really dislike the RAM exhaustion when multiple workflows run. Also, the fact that they never clarify how much RAM your instance has. Also, customer service is only available via email, and 99% of the time, they blame you for the issues the workspace has. Review collected by and hosted on G2.com.

n8n is the 'Zapier killer' for anyone who values control and privacy. The node-based visual builder is incredibly intuitive, but the real power is in the code nodes—being able to drop in custom JavaScript exactly where you need it is a game changer. Plus, self-hosting means no 'task tax' and total data sovereignty. It’s the ultimate bridge between no-code and full-code automation. Review collected by and hosted on G2.com.

The biggest drawback of n8n is the technical barrier to entry. It's not a 'plug and play' tool; you need a solid grasp of JSON and logic to get the most out of it. Additionally, if you choose the self-hosted route to save money, be prepared for the overhead of server maintenance and the occasional headache during version updates. Review collected by and hosted on G2.com.

What I like most about n8n is its flexibility and developer-friendly approach to workflow automation. Unlike many no-code platforms, n8n provides the right balance between visual workflow building and the ability to write custom JavaScript when more advanced logic is needed. Visual drag-and-drop workflow builder that's easy to understand. Self-hosting option, giving complete control over data and infrastructure. Extensive integrations with APIs, databases, AI models, and third-party services. Support for custom JavaScript code and complex workflow logic. AI agent and LLM integrations for building intelligent automations. For me, the most valuable feature is the ability to combine no-code automation with custom code. I can quickly automate simple workflows while still having the flexibility to implement advanced business logic whenever required. The biggest benefit is increased productivity. n8n reduces repetitive manual work, simplifies system integrations, and enables me to build powerful automation workflows without developing an entire integration platform from scratch. Review collected by and hosted on G2.com.

The biggest drawback is workflow maintainability. As automations grow in size, managing, documenting, and troubleshooting large workflows becomes increasingly challenging. Breaking them into smaller reusable workflows helps, but better visualization and debugging tools would make the experience even better. Debugging complex workflows can become difficult, especially when many nodes and conditional branches are involved. Some integrations require manual API configuration because there isn't always a pre-built node available. Review collected by and hosted on G2.com.

What I like most is the “Automation as Code” philosophy. As an Azure Architect, I don’t want yet another siloed SaaS tool; I want a platform I can self-host, keep under version control, and extend with custom JavaScript when needed. n8n is the only tool I’ve found that gives me “enterprise-grade” control over my data flows, while still being just as quick to build in as any no-code alternative. Review collected by and hosted on G2.com.

While n8n is incredibly powerful, it still doesn’t have the “Enterprise-First” deployment rigor I expect as an architect. The manual effort needed to keep environments in sync, along with the “opt-in” approach to error handling, means you have to invest a significant amount of time in Operational Discipline to make it genuinely production-ready. I’d really like to see stronger native version-control integration, as well as a global retry-policy engine. Review collected by and hosted on G2.com.

I love the flexibility of the self-hosted version. It allowed our team to automate sensitive customer data workflows without sending that data to a third-party cloud, which was a dealbreaker for us on other platforms. Review collected by and hosted on G2.com.

What I find most challenging is the steep learning curve for non-developers. While the power is there, the UI can feel overwhelming when trying to map complex JSON data without a background in Javascript. Review collected by and hosted on G2.com.

This is one of the best and easiest ways to implement AI workflows without needing extensive technical knowledge or a deep understanding of AI concepts. Interestingly, it also helped me learn and apply AI concepts such as Agents and LLMs. Review collected by and hosted on G2.com.

The licensing for this application can be somewhat confusing and tends to be on the expensive side. While the software is still evolving, it has yet to fully mature for large-scale enterprise deployments, particularly in areas where governance, privacy, and security are critical—especially given the current landscape of AI. Review collected by and hosted on G2.com.

The power it gave me to build AI workflows when other tools had limitations. Self hosting the community edition is a big + for my solo projects.

There is a big of a learning curve but it is actually healthy as it helped me learn concepts like JSON, javascript, to really understand how integrations work Review collected by and hosted on G2.com.

So there are a couple of things that could be improved. For the AI integrations, we still aren't able to fully customize the settings of the AI models that we use. For example, GPT-5 has been released quite a while ago with new parameters like minimal reasoning or verbose, which are elements we still cannot edit when using OpenAI as a chat model. I must say that the cloud instances have very limited power and crash a lot when processing lots of data. Review collected by and hosted on G2.com.

What I like most about n8n is that it gives me the power of a low-code tool without boxing me in. The visual workflow builder is easy to pick up, so getting my first automations live didn’t take long. At the same time, it still lets me build fairly complex logic when I need it. Once a workflow is set up, it just runs in the background and stays out of the way, which means it’s something I actually use regularly instead of only testing once and then forgetting about it.

I also appreciate how many features and integrations are available out of the box. Connecting different tools is usually straightforward, and when something isn’t supported directly, calling custom APIs is still simple enough to do. The documentation and community support have been helpful as well, especially when I’m trying to tackle something a bit more advanced.

Overall, n8n hits a sweet spot between ease of use, flexibility, and straightforward implementation and integration—something I haven’t really found in other automation platforms. Review collected by and hosted on G2.com.

I wouldn’t say I “dislike” n8n, but it does have a few rough edges. The learning curve can feel pretty steep at the beginning, especially if you’re not already comfortable with APIs or other automation tools. Some nodes and error messages come across as quite technical, so getting more complex workflows to run exactly the way you want can take some trial and error.

Overall, performance and stability are solid, but I do occasionally run into small bugs or editor quirks that break my momentum. The documentation has improved a lot, which I appreciate, yet there are still times when I wish it included more real-world examples or clearer guidance for advanced use cases.

None of this is a deal-breaker for me, but these are the main areas where I think n8n still has room to grow. Review collected by and hosted on G2.com.