I run an automation agency at orbiscr.com, and we use n8n for our projects. We create systems for others to use in their business automation rather than for our own use, and we use n8n to build these systems.

We primarily use n8n to create scrapers, website crawling tools, data collection systems from websites, and email automation, all integrated with different applications. I do not use it solo; I use it in combination with other technologies. We sometimes use it with a Python backend and partially with n8n. We call the backend through our architecture, for example, when we have a Python backend and a front-end with React, and then we have another backend with n8n, combining it with different usage patterns.

The best features of n8n are reading documents like Google Sheets and Google Drive, and especially the AI agent functionality, the Agent Mode and AI agent nodes, which I find very amazing, along with webhooks and HTTP nodes.

I am integrating n8n with Google products. For example, I created an application where the client has huge data, and we integrated n8n there with Google Sheets. The Google Sheets already contained Google Drive links and Google Drive folder links, which we then list and send to AI. We summarize it through AI and then send it back to Google Sheets.

n8n currently has Python in the beta version, and if it can be fully released, it would be more amazing because JS is working very well, but if Python is added as a standard feature, the results will be exceptional. Python is very reliable and a very diverse programming language that helps tremendously with complex tasks like machine learning. It has different libraries like Selenium, so web scraping and data collection from websites are very effective with Python compared to JavaScript.

The setup of n8n is sometimes very complicated, especially with the HTTP node, which is complex to set up. The local credential and environmental credential configurations are somewhat complex and could be simplified.

I have been working with n8n for almost one year.

I would rate the scalability of n8n an eight, which is acceptable.

I have never needed to communicate with n8n's technical support.

The setup of n8n is sometimes very complicated, especially with the HTTP node, which is complex to set up. The local credential and environmental credential configurations are somewhat complex and could be simplified.

Once n8n has passed all the tests, it works fine, which is why people use it on a VPS instead of a local server. Using the paid version is somewhat expensive.

The difference between the free version and paid version of n8n is that the paid version, if hosted locally, is almost the same. However, with the latest update from using the website, you get the AI assistant that can set up the flow for you automatically with prompts, which is a good feature. Cloudy also does the same, so people use Cloudy and they use a localhost of n8n. For example, I normally create a JSON file for the complete flow through my AI coder and then import it to the n8n canvas.

My customers mostly prefer deployment on the cloud. We are using AWS, Hostinger, and there are a few more cloud providers. I would rate this review a 9.5.

One standout project I built with n8n was for a distribution client using Odoo ERP. They had a fully manual order fulfillment process where orders from their Shopify and WooCommerce stores landed in Odoo, but the warehouse staff had to print pick lists, pack slips, and perform extensive manual work to create shipping labels in ShipStation and then re-enter all those details into Odoo. This process was very time-consuming and error-prone. I built an n8n workflow that acts as the central system, triggering every 10 minutes on new confirmed sales orders in Odoo using the Odoo node. The workflow validates inventory levels, applies custom business logic for bundling and formats, creates the shipment request, receives the tracking number and label URL, and sends a personalized SMS to the customer via Twilio.

I have also implemented a wide range of workflows across different business functions using n8n. For example, in retail, I automated the synchronization of online orders from Shopify to Odoo ERP, which automatically creates sales orders, checks inventory levels, sends confirmation emails to customers, and triggers notifications. I built complex multi-step workflows integrating CRM, marketing, and customer support systems. For instance, when a lead is marked as high priority in Salesforce, n8n triggers a workflow that notifies the account manager on Slack, adds the lead to a marketing nurture campaign, and schedules a follow-up call in Google Calendar.

My client absolutely wanted to get hands-on participation in the project, but had no technical background. Instead of coding the task itself, I walked him through some of n8n's features and gave him workshops and tutorials. Now he is able to configure some of the flows himself without relying on an external consultant.

n8n has positively impacted my organization, particularly in terms of saved time, reduced errors, and increased productivity. My recommendation is to start with n8n if you have any need for automation or integration, especially if you value control, cost predictability, and technical transparency. For technical users, developers, and IT professionals, n8n is more transformative. Unlike some black-box automation platforms, n8n empowers you and does not hide complexity; it visualizes it. You own the workflows and your data, and the cost structure is a major advantage. For a self-hosted scenario, n8n is free forever, which is phenomenal for small businesses or for prototyping. Even the very reasonable price in n8n Cloud plans is transparent, and in my team, it has made an impact on junior people because it gives them more hands-on experience with implementations, allowing them to transition to custom coding later.

n8n's strongest feature is its transparent execution model. Every step of the workflow execution is very transparent, and you can see the exact input and output data of every single node. There is also a debugger function, which is very valuable when building complex integrations or troubleshooting APIs. Additionally, the expressive power of the JavaScript and SQL nodes is another superpower. It allows me to drop into code for complex transformation while using pre-built nodes for the heavy lifting of API connections.

I use the JavaScript and SQL nodes in my workflows, particularly when importing data into a database. Data that comes from very old legacy systems or from a massive master sheet in Excel requires mastering SQL because mapping the fields sometimes cannot be done manually. Using the SQL function allows me to automatically map the columns and save considerable time for the customer.

I also appreciate that n8n is open source. This means I can inspect the code, customize it if necessary, and even contribute back to the community. This level of transparency is critical when working with clients who have strict IT policies. Additionally, the community around n8n is very active. I can usually find pre-built nodes, example workflows, or guidance on complex integrations. This accelerates the development time and reduces the need for custom coding while still giving me the flexibility to build very sophisticated workflows.

Regarding improvements for n8n, I would highlight two areas. First, while there is a growing community, the ecosystem of pre-built nodes for niche or proprietary systems still lags behind some commercial competitors. I often have to build custom HTTP request nodes or create my own node packages, which is doable but adds to project timelines. Second, features for team collaboration and lifecycle management are still evolving. Moving workflows from development to staging to production can be a manual export-import process, and native Git integration and more robust user roles for larger client teams would be a game-changer.

Regarding needed improvements, I believe much of the existing documentation is not peer-reviewed. There is definitely a big community, but if the documentation or tutorials were approved or at least had a badge from n8n, customers would feel that this meets the standards of n8n editors themselves, which would give more confidence in the flows.

I have been using n8n since it became popular in 2023.

n8n is stable in my experience if you are an experienced implementer who knows how to fine-tune configurations for production and scalability. However, if you are a junior developer or preparing for a prototype, you might face challenges that could lead to failures in production.

My experience with n8n's scalability indicates that while you can scale up workflows or deployments, many of the prototypes and pre-built flows are not resilient enough for production when there is extensive manual error or manual processes. There is considerable fine-tuning that needs to be considered, and it often requires changing some features and flows. Therefore, it is crucial to have a technical team to support you with real experience in n8n and large-scale implementations.

Through n8n Cloud support, I have had a good experience. I received dedicated support via email with generally good and faster response times than the free tier. I describe the support team as knowledgeable, helpful, and responsive. They often address questions relevant to my scope, and at times they also recommend searching for technical documentation or specific forums. Ultimately, they always point me to the relevant solutions for any problem.

Previously, I used Zapier and Make.com. The reason I switched is that those tools are still not complex enough to manage multi-flow processes and are better suited for simple use cases with basic automations and triggers, particularly for small companies that want to automate one or two processes.

The pricing experience was a bit of a learning curve. Working in an on-premise environment with no cost is great. For cloud environments, I noticed that depending on the number of nodes used and the number of executions, the basic plan might not be enough, and this is something I learned through experience. These days, when I recommend n8n, I always go with the tier one pricing system, opting for the premium option.

I have not seen a return on investment from my side because the impact is primarily on the customer side, where non-technical people can go through a learning curve and take over some of these tools, allowing them to save costs because they can reduce the hiring process for external developers for this type of automation that they can do themselves.

I have not seen a return on investment from my side because the impact is primarily on the customer side, where non-technical people can go through a learning curve and take over some of these tools, allowing them to save costs because they can reduce the hiring process for external developers for this type of automation that they can do themselves.

The pricing experience was a bit of a learning curve. Working in an on-premise environment with no cost is great. For cloud environments, I noticed that depending on the number of nodes used and the number of executions, the basic plan might not be enough, and this is something I learned through experience. These days, when I recommend n8n, I always go with the tier one pricing system, opting for the premium option.

Before choosing n8n, I evaluated Zapier, Make.com, and Workato. Among the three, I felt that n8n provides the most visual experience, the most powerful tool for execution and for production, and a very reasonable pricing system.

For those looking into using n8n, I advise that collaboration is key. I typically use Git or version control repositories to maintain workflow definitions for larger projects. I also document each workflow extensively within n8n using node descriptions and comments, so anyone on the team can understand the logic. For cross-functional teams, I use screen share and workflow walkthroughs along with peer reviews, which I recommend to other users and junior implementers starting with n8n.

I advise testing first in staging. I suggest starting simple and then scaling. Begin with a simple workflow connecting Slack with Google Sheets before adding complexity. It is important to master the fundamentals, such as data structure and JSON data between the nodes, and then gradually work on optimization, rate limits, monitoring execution, and deploying additional nodes along the way.

I rate n8n an eight on a scale of one to ten because it is a very powerful tool. However, I feel that for very complex production environments, it still requires considerable fine-tuning and knowledge, and sometimes it might not be fully ready to support large-scale implementations in production. For prototyping and small use cases, as well as medium and small enterprises, n8n can be a great option. For larger implementations, it might still lag behind.

n8n is actually used for many workflows, specific to healthcare and the supply chain industry, though I cannot say anything specific and must remain abstract.

I have used n8n to connect diverse applications across different operating systems and different API calls. However, I am yet to see direct access to queues and Kafka queues.

The best features of n8n are its low-code capabilities with everything at your disposal. The connectors are getting enriched, which is beneficial. It's all about these low-code tools. I don't see any difference between this and classical EAI; it's a kind of integration where we are doing extensive integration with different sources and targets, and here we are interacting with different sources and targets to apply the enriched data to the models and everything so that it makes better decisions.

They haven't fully helped tailor solutions for our needs. We need to make adjustments, enhancements, and create custom workflows. I'm not saying it is completely ready at this time, but it solved most of our purposes. You cannot ask for everything as these products are getting to a matured stage, so we have to give some time for them to reach maturity.

Regarding integrations, if they can make it more seamless and beef up their workflow constructs, that would be great.

Another improvement would be in making n8n more popular through a similar approach to Kafka's documentation. Kafka has created ten-minute capsules of every use case, covering basics, foundational concepts, and each construct in detail. If a newbie spends at least a week studying the documentation, they'll be equally proficient. Any experienced person, senior developer, or architect who goes through Kafka's documentation and videos will have a good jumpstart. If n8n applies a similar methodology, that would benefit the entire community and help them grow faster.

I have experience with n8n for approximately one plus years.

The initial setup was not complex; it was straightforward.

I did not get much opportunity to deal with their customer service or technical support for n8n.

I deployed it alone.

As a chief principal architect, I maintain a hands-on approach to implementation.

The documentation in general was adequate, though it could definitely be better. However, it was decent enough for what I needed.

On a scale of 1-10, I rate n8n a seven out of ten.

Our customer's main use cases for n8n are extensive, as they are heavily utilizing it. n8n is clubbed with MCP, MCP protocols in terms of tracking, alerting, and automating some of the jobs, specifically in terms of billing, invoice generation, infrastructure monitoring, and downtime management; these are the areas where this platform is used.

The most valuable features in n8n that I can highlight are its flexibility and ease to configure. Adaptation is pretty easy. It requires a little bit of programming skills, but comparatively, it's the shortest time window one can get comfortable with. People who have Python programming experience can get their comfort pretty easily.

Adaptability, scalability, and integrational flexibility are the top features at this point of time.

The capability of n8n to connect different applications and automate tasks while supporting over 200 pre-built integrations is pretty effective. So far, we have had all good positive thoughts about it. Our infrastructure teams are more vigilant about the security part and about data flow and data privacy at this point of time, but otherwise, the adaptation and flexibility of the tool is pretty good.

Regarding n8n supporting over 200 pre-built integrations, this capability contributes to the organization's communication in a pretty effective way.

I am familiar with the visual editor for designing workflows in n8n and it adds great value. It serves as a miniature digital twin framework to capture the existing as-is process flows. That design and visualization feature definitely adds a lot of value to the overall service portfolio.

The impact of real-time execution logs on optimizing the workflows is quite effective, but currently, the solutions we have prepared using this product are in a semi-production area only. We have not yet put it onto the production area.

What could possibly be improved in n8n to make this solution better for the next release is still a navigating process for us. I don't think I'm in a position to provide the improvement areas yet because we are still learning and discovering them.

Regarding the cost, scalability, and initial setup, I can think that for a bigger enterprise environment, the scalability, based on my observation, could be improved since we haven't fully explored that modular adaptive nature of n8n.

There are limitations in scalability.

We have been working with n8n for more than six months now. We started using and exploring it around December.

The deployment part is straightforward initially. If the environment grows and if n8n is scaled to the enterprise, customer-friendly plans would definitely help. Right now, it appears as a one-size-fits-all plan.

In terms of stability, n8n is pretty good; I don't see a challenge. I would rate it seven and above in stability.

There are limitations in scalability. For scalability, I would rate n8n above seven, as we are still in the process of setting up the platform and configuring it.

I have contacted tech support regarding n8n, but that is where we are struggling. We are not in SaaS platform mode of subscription, nor using a standard plan; instead, we are utilizing a customized plan of setting the server in our environment.

For the support rating, I would give it about a five; it's not too bad, but not too great either. It's acceptable.

We do have good experience with other AI automation tools, mostly the open-source ones. Another open-source tool we are working with is make.com.

Regarding the cost, scalability, and initial setup, for a bigger enterprise environment, the scalability, based on my observation, could be improved since we haven't fully explored that modular adaptive nature of n8n.

Our customers do see a return on investment with this solution; it absolutely offers value for money. One direct correlation is in terms of the amount of automation the tool provides, which directly impacts the number of people who can perform those tasks.

For a small license part, I find n8n is reasonable. But the minute you expand, I believe it becomes a bit on the high side.

I haven't had an opportunity to utilize the error management system. We haven't explored the security module features completely, so it wouldn't be fair to conclude in a generalistic way. Our teams are still exploring that module.

I would definitely recommend n8n to those who would want to use it, even though there are some challenges. It is one of the pretty effective products in the AI automation area, and it can grow leaps and bounds if some of these things are taken care of. It is pretty effective and one of the best tools I have seen.

On a scale of 1-10, I rate n8n an 8.