We were trailing it and wanted to essentially switch our entire backend to it - and technically it seemed to be able to do the job, but their licensing turned out to not be a fit.
For a moderately used app we very quickly burned through their “executions” that were allotted by our license - and that’s where we host it ourselves, configuring and paying for the servers, load balancers, key value store and database, with its failovers and backups.
So the license was to use it on top of all that, and even their highest enterprise license was cutting it close, and if you “run out” of these executions, the service just stops working …
And all of that would have been fair if it was hosted, but sounds ludicrous to me for something we self host.
I think it is an incredible piece of tech, but just not suited for a dynamic startup, and once we spent the time to code up the alternative paths for our use cases, it no longer made sense to use n8n at all, as we mostly solved all the problems it was helping us with.
I mean my current role doesn't put me into these kinds of considerations anymore, but I didn't hear they've changed their ways - so I just thought that's still the way you pay for Oracle DB
Many moons ago I was a sysadmin at a company and first heard about their licensing strategy when sizing up some new multi-socket Opteron servers. First time I learned about per-CPU licensing.
There are alternatives to n8n depending on your stack of what is being orchestrated. Node-red, and others have quietly existed for a very long time, similar to how n8n existed for a good while before being discovered by the AI world.
I think that's a critical argument that n8n and others will have to overcome - why should users decide to over-specialise one aspect of their stack when they can do so much more and have so much more control elsewhere?
One for the many, and maybe one for a different audience.
Node-red didn't have a login at the time, and was well geared towards direct IoT type flowing. I loved how I can just jump into the logic and flows.
N8N has a lot of pre-configured connectors.
If Node-red had the pre-configured connectors to different systems as easy as n8n, I think the point here would be even stronger.
It's why I use both. I start with the low level, and if orchestration beyond it is needed, I can tie in another tool if needed.
We used n8n for two things mostly - AI agents and process automation.
For AI we just built own MCP servers, and then the agents are quite easy to use as the major frameworks kinda help you with it. N8N’s AI is kinda just a UI layer for langchain - though we just used google’s adk.
For process automation - well there is so much options it’s not even funny.
Like you could make a car like a truck, but why not just buy a truck in the first place?
> then you start getting into additional costs, controlling how much of your metrics are surfaced vs. kept internal, etc
I don't know what this means. In my experience with anything sold B2B, all the terms are negotiable. If you want unlimited everything, you can ask for that.
They are open core I think (MIT+enterprise features model)
Question to folks who’ve used n8n extensively, I’m curious, what are your experiences with n8n, and how much does it end up being a web of verbose “visual python” in practice?
I’m very much biased here and have a vested interest, because I’ve been working on a new product not far from this space, but much more oriented at technical users (platform engineers, primarily, see [0] and [1] for a shameless plug, not released yet), but really, I’m curious about what experiences folks have had here, and what your main issues with it were, esp. if you used it in a platform/devops engineering role, or maybe why you decided not to use it.
Oh, it's open (core) source. And while certain (just a couple of) enterprise features should have been made open to qualify as being called open source, it's very close to that. Most powerful features are open, ready to self host, modify and make your own.
Does it end up driving webs of python partials forming apps. Absolutely. Does it scale ? It does. Do complex flow remain maintainable? As a coder I prefer to maintain a repo of code than visual elements made of snippets. But, the critical advantage is productivity, for simple flows the community intelligence solves everything so you can get an operational set of valuable solutions within hours, even minutes once proficient with the interface. Another factor is, you can deploy pilot flow acting as applications, test them with production data, and make that live with the press of a button once pilot testing is done. With a code project you would need a robust and well polished cicd pipeline to get that.
The limit or cons to me is a logic and compute heavy solution just isn't a fit to run on an n8n platform, scaling n8n just isn't as intuitive as scaling pure application component that do one thing.
An example you may have a cpu heavy node, and a memory heavy node. It makes scaling the whole instance very inefficient. Scaling memory of a dedicated memory intensive application and scaling compute for the compute intensive component simply is far more optimal.
If resource cost is not significant relative to the value of your flows then just scale a self hosted n8n and you only need to digest having to maintain, following your analogy, a "nest of pythons".
Note: n8n sadly only supports python or JavaScript for custom code nodes, would have been nicer had they built a polyglot runtime instead. That's however more than what every other flow platforms let users do.
It is absolutely not open source.
The "Fair Source" license that n8n invented has two related qualifications that make it not open source:
> You may use or modify the software only for your own internal business purposes or for non-commercial or personal use.
It's not open source if you can't use it professionally or sell work derived from it [ed: comments have correctly called out that this is not the deal, thank you]. There's no chance this license or anything like it is ever going to be an OSI approved open source license. https://opensource.org/licenses
I also find it weird how little use it gets. Possibly a side-effect of true open source having been more popular to the point of source available being historically unknown.
Does anyone _really_ use these low/no-code platforms to create products? I was always under the impression that you'd primarily use something like this for "internal business purposes" i.e. little internal utilities that you can't justify spending serious development time on. Which the license lets you do.
Apparently there is a total market of Ableton addons[1] (for example) sold on separate markets. I would call such addons (or packs) "low code".
So there is definitely a potential market for "add ons". But does n8n a) support that and b) encourage such markets for money?
Further nitpick: Their Python implementation is based on WSAM so libraries that require C compilation won't work.
However if this funding let's them integrate a Claude-Code like tool, they'll have an amazing product.
I've been evaluating n8n as a way to build things quickly for clients, but I do wonder about what happens when they want to turn the automation into a full app. I wish there was a first-party way to export an n8n workflow as a plain Python script or set of scripts.
Have you ever had to migrate a project from n8n to code?
Most of my workflows remained in n8n, those that are unimportant or turned out unnecessary to the application. It saved me days for each, not having to build an app backend and cicd for those.
But what I explained is that contrary to many open core projects, n8n public sources form a (generous) comprehensive solution, aside a couple of features that should have been public sources, the solution stands as a grea, with little to no limits, platform. Unlimited users, no cap on workflows, no cap on number of nodes etc etc.
Also, their licensing is good, you can pay and get the extra features and do what you want, including modifications if I'm not mistaken. That's free software without the free beer aspect, I never claimed it was free software or pure open source.
I'm also all for free software, but this is the sort of solution that doesn't fit well with the open source philosophy , making it rare to see open sourced. That's all.
Now that they raised significant money, the situation will slowly change to prioritize monetization, I guess.
It’s not open at all. It’s a nonfree license.
It’s neither free as in freedom nor free as in beer.
You’re not “all for free software” if you carry water like this for a proprietary nonfree source-available package such as you are here.
That's a terrible reimagining, subtlying implying that open-source is old and past it.
It's a pre-open source license. They've not quite made the bar.
It's really not that deep; U2 and Mogwai exist in the same timeline, in the same shared canon of contemporary music.
> It's a pre-open source license
This statement is strictly ahistorical; the earliest software licenses which made source code available to everyone and included restrictions on redistribution and/or use date back to the late 1990s[1][2].
You can certainly _try_ to make the case that these are the same as the Xerox license, but I don't think it would be a very strong one.
This is basically just allowing self-hosting of a third-party's cloud, which is an improvement over traditional SaaS, but shouldn't dilute the FOSS label.
This is such a strange thing to post in response to a link which states:
> Although n8n's source code is available under the Sustainable Use License, according to the Open Source Initiative (OSI), open source licenses can't include limitations on use, so we do not call ourselves open source.
It's as if you don't want source code to be available _at all_ unless it's under a FOSS license
The existence and growth of FOSS is something that has happened as a result of considerable advocacy, and while its broad success has become somewhat self-sustaining, there will never not be the risk of a slide into more single-corporation-friendly "source available" realms.
It's not a bad thing to push for "source available" to be considered as not going far enough, and to not let it supplant FOSS through purely pragmatic concerns.
There is something off about this to me in a world where FOSS exists in it's present form primarily to the outsized benefit of hyperscalers and entrenched incumbents
There was a post on another forum earlier this week on this same broad topic which resonated deeply with me, as someone (who like most of the US population) is a layoff and a medical emergency away from ruin:
> When I started getting interested in open source, I had problems like unreliable software, the inability to inspect or improve it, limited experience with collaborating. Open source solved those, but now my most pressing problem is that the excellent software I use is undermaintained and outright abandoned because the creators can't afford to keep donating time to it. Open source has been a process for solving problems, not the end goal. If it's not capable of solving problems, it's time for new approaches.[1]
Not exactly the fault of n8n, but the confusion is there to clear up. That is all I'm reading into it.
They did start out by incorrectly calling themselves open-source, but to their credit they stopped doing that and have been very clear ever since.
> Nowhere in any common open license does anyone promise to keep working on their project, much less on particular terms. Any contributor to a permissively-licensed project can license their next contribution however they want. Any steward of a copyleft project with rights to all contributions can, too. Much as you could pick an Apache-licensed project, fork it, and sell your enhanced version under proprietary terms, a project steward can share new work under new terms, as well.
>
> None of this changes the license terms for old releases. Prior versions with MIT or Apache 2 or MPLv2 or what have you in the LICENSE file remain available to use, share, and change under those terms. That includes forking. We see that every time a going-forward relicensing spawns a new one. The reason the new license terms matter for new releases is that those new license terms apply to the diff between the old release and the new one.[1]
[1]: https://writing.kemitchell.com/2023/09/23/Two-Kinds-Relicens...
Are these projects comparable?
First off, Node-RED handles real-time event data much, much better in my experience. Because of where Node-RED came from, there's much better support for IoT, MQTT, Modbus, OPC UA, edge protocols, etc. n8n is much more limited in this regard, and the fact that the Node-RED and FlowFuse community has literally thousands of custom nodes makes the calculus pretty clear.
I also think that FlowFuse/Node-RED has better integration of AI workloads. In theory n8n is designed around AI, but it treats it the same way OpenAI's AgentKit does - as sort of opaque connections. FlowFuse/Node-RED instead treats it as an actual message payload (both in terms of how you connect to the APIs and how you interact with what's generated), so instead of throwing your request into the void and hoping for the best, you can control every minute part of the flow.
That also makes for much more transparent debugging and visual data flow - the whole idea of these low-code environments is to give you the same control as high-code without the headache. Abstracting that away too much gives you less control, which is sort of the antithesis of this approach.
Like I said though, SUPER biased here.
No-code: "I don't need code, this is so easy!" 2 weeks later "I wish I had access to literally any code system to make this work."
Low-code: "I don't need code, this is so easy!" 2 weeks later "Oh awesome I can actually use code here!"
With extra restrictions, n8n is at most "source available".
(And BTW I like OSI’s definition!)
Instead of claiming the words “open source” always mean OSI’s definition, it’d be better to clarify up front that you’re talking about OSI’s “Open Source Definition”, which is (effectively) a trademark or term of art and does not preempt all possible definitions of the phrase. (Special note that the words “open source” do not always refer to code.)
The turf battle over the words free and open seems silly to me, with both sides arguing that the other side’s word literally means something other than they intend. Stallman has argued that open means source available, and OSI has argued that free means price. Both sides are right, and both sides are stubborn too.
* Free Software - meets FSF's four freedoms
* Open Source - meets OSI's definition
* Source Available - you can read the source code
Rather than clarify confusion & raise a real and important distinction, this post carries water for obfuscation and confusion: if people aren't clear the answer isn't to loosen the definition, it's to make the distinction clearer.
This is not an argument in favor of n8n.
If you enable the enterprise feature flag or use the Docker image, the result is source available.
I think it’s fair to call Windmill open source. It’s using the open core model for commercialisation. Just because you publish open source it doesn’t obligate you to make all the code you write open source.
Hiring 50 fairly well paid developers is roughly $15M/year, maybe more if one insists on SV compensation which always seem a bit absurd.
$240M total funding is a lot of money. They’ve only been around for about 5 years and probably didn’t start out fully staffed.
So they’re basically covered for the next 10-15 years even if they had zero sales ?
Having 500 employees won’t speed things up and would actually slow down development - so why so much funding?
Or who actually waits that long? The first version of Windows 10 was released about 10 years ago and soon will be EOL.
I feel software investment is like some oil ETFs — there is more investment money than the thing to invest in…
Of course that valuation makes sense if you've seen the insane prices they charge.
Specially if they go the PaaS/SaaS AI route.
For their Business (self-hosted) plan, they have essentially 0 cost per customer.
> The focus for the fresh funding will be on expanding its engineering capabilities and hiring.
https://pitchbook.com/news/articles/ai-agent-startup-n8n-lan...
It's an okay product I appreciate that it's selfhosted with good documentation but they absolutely destroyed their brand with excessive affiliate marketing and now nothing of substance is left if you search for it anywhere.
I like n8n. It feels a little less rough around the edges for visual coding than something like huggin or nodered. The documentation is good, but finding examples and things like that offsite is impossible.
Alternative would be writing custom code, deploying it somewhere, setting it to run automatically on schedule somehow, and modifying it and redeploying through a dozen steps every time.
Of course there is docker and cron and deployment scripts - but all of that is not needed with n8n for these kinds of use-cases.
For me, that's the primary value of n8n - nodes themselves are nice-to-have shortcuts, some of the time. Maybe I'm not familiar with tools that make it easy to "just write code" and have everything else (deployment, orchestration etc) covered?
I kid you not, we use another no code solution at work and it was originally meant for PM to create workflows. It came to us the devs to make it and we resent daily working on it.
Our life would have been much similar if our workflows had been written in code.
Of course this is a standalone page written in some language that I forget. I think Cursor mentioned some animal name... anyway. Can you please put this into our product please?"
They get a readme, compose.yaml and git repo.
This has worked for set-it-and-forget-it experience.
I tried n8n but for some automation needs, it just wasn't flexible and/or I'd have to build a custom module. My choice was custom code in n8n or custom code in whatever.
The problem, in general is not about “unpure” OSS.
The problem is “free riding” by slapping “open source” marketing without any real or meaningful open source contribution, nor any intent.
We should be happy when companies do this.
stallman is from the free software movement which doesn’t agree with open source since FSF is concerned with moral freedoms.
$40M rev makes this a 62.5x rev multiple. AI has been around 40x lately so it’s a bit high but it sounds like there was competition to lead and those are March numbers so it’s probably about on par.
Personally I find these multiples absurd but big VC needs to put money somewhere and AI is the new SaaS so here we are.
> The Series C comes less than a year after a $60 million Series B, which valued the German group at a reported $350 million.
That’s a huge step up so maybe their growth numbers are that good.
https://pitchbook.com/news/articles/ai-agent-startup-n8n-lan...
I hope they spend a good bit of the $180M on building out their input connectors.
Or will this become a white elephant too large too sell like Zalando?
Whatever happened to IFTTT?
Edit: It’s a valuation of 2.5 Billion - hence my question. There is snowballs chance that they will ever be worth that much. They are SaaS and not consumer products. They have no side gig like amazon or google - they have a single product in a tight market.
Like Databricks with no consumer products, no side gig, single product, but 100 billion valuation[1]?
[1] https://www.forbes.com/sites/sasirekhasubramanian/2025/10/07...
Geniunely asking since I do wonder what their USP is and why it's worth 2.5B.
n8n offers an orchestration, integration, ETL platform as SaaS/PaaS (plus self-hosted).
For many medium sized and enterprise companies such a service is as important as the data platform from Databricks. From my personal experience in a small EU-wide enterprise in finance sector, the budget for the data platform is in the lower 7digits/year. For automation and workflows multiple solutions in place. Some are outdated for years (on mainframes) or with an EoL in the next years (due to retirements in-house or at the datacenter provider).
Since decades such companies externalize their own business logic – formally running on homegrown software – to SaaS vendors. Established workflows are quite often not able to integrate these kind of "modern" integration. REST? GraphQL? WebSocket? We speak SOAP and FTP!
So, from my point of view n8n is a valid solution in a growing market.
I don't know but their sudden heavy adaption of AI will have detrimental affects on their other business such as integration of less "sexy" basic services such as FTP and SOAP.
I guess it all comes down to how well Claude and Co. work with integrating legacy systems based on priority protocols (in part). Because I assume that's the kind of systems that are coming up for retirement.
I think the pie is big enough for everyone to benefit.
I haven’t tried these agent-and-connector-based approaches yet — where should someone start to get a good grasp of this kind of automation?
My question to non-tech folks who used n8n, especially marketers: what has been your experience with n8n? Did it help you automate creative things like blogs, newsletters, white papers, etc? What tips would you give about n8n?
https://github.com/n8n-io/n8n?tab=readme-ov-file#what-does-n...
I asked an LLM if there's ways to detect suspicious starring activity (e.g. if stars were purchased). It suggested checking the project's star history [2] (doesn't appear suspicious).
It also suggested the stars to issues ratio. n8n has 147k:6k (about 25:1) compared to, say, rails with 57k stars and 18k issues (about 3:1).
I haven't looked deeply into n8n (is it 'no-code' for building agents?). I just see hype and am default skeptical.
[0] https://github.com/n8n-io/n8n
[1] https://github.com/EvanLi/Github-Ranking/blob/master/Top100/...
I wouldn't call it "nocode". You need to get pretty techincal to implement useful functionality. You need to write SQL, you need to extract data from XML or JSON, you need to describe HTTP queries and parse responses. You're doing it in a GUI editor, connecting nodes, so it looks like a block diagram with ordinary nodes, conditional splits, loops and so on.
For me, personally, it looks very weird and I wouldn't use this product. It's much easier to just write code. But some people are afraid of code and will jump over all kind of hoops to pretend they're not programming.
https://platform.openai.com/docs/guides/agent-builder
The openai agent builder launched 2 days ago is basically inspired by n8n. n8n when launched wasn't an AI tool, it was inspired from numerous enterprise integration tools like Mulesoft, which were inspired by dozens of other enterprise tools, some launched even decades ago.
If you haven't tried you should check it out. Its an amazing way for no-coders to build something substantial in a relatively quick manner.