The best part of FIN is that we were already using Intercom, which has been a real success for us and at a rapid pace. It was very easy to set up, as we had the proper documentation and repository options for training the AI, which was straightforward. Also, the Fin team at Intercom has really developed many features. The one aspect I like most about FIN is the guidance section, where you set guidelines for Fin, and it’s very simple. I believe Fin has been designed exceptionally well for a chat tool like Intercom, and it integrated seamlessly with our existing environment. This allowed us to assess everything, implement, and go live within a month. I think even that might be a bit ambitious, but I just wanted to be cautious and say it was around 30 days.

One thing about Fin that I can say is the most disliked aspect for us is the assumption of resolved cases. What I mean by that is, let's say a customer asks a question and does not come back to confirm that the answer is correct, which often results in a number of responses. Fin then takes it as a confirmed case or a resolved chat, which I believe is a little problematic because we can't truly assess all the chats in this category as resolved. It takes us some time to review these chats. Although they provide an option to filter these, it would be much better if we could send a couple of follow-ups and then close the case. I think if that process is implemented, it would help address this issue.

FIN is helping us to reduce head count in the support team by at least 50% and also it also allows us to be available for support 24/7.

Fin is easy to use and manage. We love how Fin does about 30% of our work. He has more knowledge than all but one of our team members and maybe not even him. We love the amount of information we can find in intercom on how it's performing. We love Fin suggests new content and shows us where our content is lacking. We are currently working with fin-first for our incoming e-mails and are hoping to implement fin-first workflows for all our incoming conversations soon.

Sometimes he leads customers in a loop of answering or requesting more information. Also in very rare occassions Fin infers information that is not true and gives it to our customers.

We always have a staff shortage on our helpdesk. We have vacancies, but it's difficult to fill them. Now Fin does a lot of the work, the workload is manageable for the team and it also gives us time to work on sideproject to drive the business forward.

I think for an AI bot it learns very quickly and has seemed to be very accurate for the most part in our company - it's saved us lots and lots of hours, so thank you!

Sometimes some of the rules you add for Fin to take certain actions when speaking to a customer can be a bit delayed, or not quite right, but this is to be expected!

It's saving us lots of time! It takes about 50% of our customer conversations now, so it frees up time for us to spend on the higher impact things

There are several things I really like about Fin:

Guidance functionality: It works consistently well, even with the many guidance rules I’ve added.

Translation feature: It’s impressive—accurate and maintains the company tone across languages.

AI Categorisation: This is a recent feature I’ve started using, and it’s been working great so far.

It boost our customer support too in the repetive query and we are getting amazing 50% delfection with so many restrictions we added at the moment.

Just a couple of things I don’t necessarily dislike, but see as areas for improvement:

AI Duplication Detection: It would be helpful to have a feature that automatically identifies duplicate articles. This would assist us in keeping the knowledge base clean and well-organized.

Fin's Handoff Experience: Fin often asks customers if they’d like to connect with a human agent, but doesn’t always follow through. I've noticed that there's a slight delay in assigning the chat to an agent, and if the customer types something during that window, Fin starts responding again—leading to frustration. If this behavior could be improved, it would really enhance the experience.

Fin has been a major benefit for us—it's delivering excellent deflection and providing 24/7 support to our clients.

We’re continuously adding new knowledge content, which is helping make our Fin agent even more advanced and effective at resolving Tier 1 queries.

We've been incredibly impressed with Fin by Intercom, and it has truly transformed our customer service operations. As the highest-performing AI agent in customer service, Fin lives up to its promise, delivering not just higher-quality answers but also adeptly handling more complex customer queries than any other AI solution we've encountered.

The impact on our chat volume has been remarkable. Fin is exceptionally helpful in resolving customer inquiries, significantly reducing the load on our support team. Its ease of use is a major highlight; the interface is intuitive for both our team and our customers, making interactions seamless. This efficiency directly translates to happier customers who benefit from speedy and accurate solutions, a critical factor in today's fast-paced environment. We find ourselves relying on Fin with high frequency of use due to its consistent performance and reliability.

What sets Fin apart is its robust training capabilities and ease of implementation. The setup process was surprisingly smooth, allowing us to quickly train the bot using our existing support articles. This has been a game-changer, as we've witnessed a substantial improvement in Fin's knowledge base and its capacity to provide comprehensive and relevant information. The number of features packed into Fin is impressive, covering everything from proactive support to complex query resolution, and its ease of integration with our existing platforms was excellent. Furthermore, Intercom's customer support for Fin has been responsive and helpful, ensuring we maximize its potential.

Fin by Intercom is more than just an AI agent; it's a strategic partner that empowers us to deliver exceptional customer experiences with unparalleled efficiency. Highly recommended for any business looking to elevate their customer support with intelligent AI.

While Fin is generally excellent, one area we've noted is that on occasion, the AI may provide customers with misinformation. This can lead to customers having an incorrect understanding of our product or its features when they eventually connect with a human agent, requiring additional clarification. We are actively working on refining our training data to further minimize such instances.

It benefits us by:

Reducing chat volume: Fin is highly effective at resolving customer inquiries independently.

Improving customer satisfaction: Customers are happier because they receive quick solutions.

Enhancing knowledge base: Fin can be trained on our support articles, significantly improving its ability to provide comprehensive and relevant information.

Increasing efficiency: It acts as a strategic partner, helping to deliver exceptional customer experiences with unparalleled efficiency.

Fin consistently outperforms competitors in customer service tests, delivering accurate, high-quality responses that meet customer needs quickly and effectively. Its ability to resolve a significant share of customer queries—typically around 50–65%—is powered by its proprietary Fin AI Engine, which is specifically designed to understand context, interpret intent, and provide precise answers based on a company’s own knowledge base. This advanced AI engine not only speeds up response times but also reduces the workload for human support agents, allowing them to focus on more complex or sensitive issues. As a result, organizations benefit from improved efficiency, reduced support costs, and higher customer satisfaction levels.

Fin’s pricing model can be a significant hurdle, particularly for smaller teams or organizations with limited support budgets. At approximately $0.99 per resolution, each interaction that the AI fully handles—whether it’s a detailed exchange or just a quick, single response—incurs a charge. This means that even simple or incomplete conversations, such as when a customer leaves mid-chat, can contribute to costs. Over time, especially for businesses with high customer inquiry volumes, these charges can add up rapidly.

The unpredictable nature of these expenses can make budgeting difficult, as monthly costs may fluctuate depending on traffic spikes, seasonal demand, or unexpected events that trigger more customer inquiries. In some cases, teams have reported dramatic increases in their monthly support spend after implementing Fin, highlighting how quickly the expenses can scale. For organizations trying to keep costs predictable and manageable, this variable pricing structure may require close monitoring or additional strategies to limit AI-handled interactions.

Fin by Intercom addresses the challenge of delivering fast, accurate, and scalable customer support without overloading human agents. By using its AI engine to understand customer intent and pull answers directly from a company’s knowledge base, Fin can resolve a large portion of inquiries instantly. This reduces response times from minutes or hours to just seconds, ensuring customers get the information they need without waiting in queues. It also handles repetitive, low-complexity questions that would otherwise consume valuable agent time, freeing human support staff to focus on more complex, sensitive, or high-priority cases.

The benefits extend beyond efficiency. Because Fin is available 24/7, customers receive consistent support regardless of time zone, improving satisfaction and loyalty. It also integrates across multiple channels—such as live chat, email, and messaging apps—ensuring customers get help on their preferred platform. For organizations, this means not only reduced operational costs but also better scalability, as Fin can handle high volumes of inquiries during peak times without sacrificing quality or requiring additional staffing.

It's great that you can give Fin guidance and make him use the right tone in answers. We also love the features that come with Fin such as evaluating topics and AI generated suggestions for quick amendments in our help center. Fin can already answer a huge chunk of our support tickets by himself and we're excited to optimise that even further.

Currently the new features are only available for English content. We could really use this for other languages such as German.

Fin provides clear, structured breakdowns of customer queries, which helps us identify trends—especially useful for planning newsletter topics. It also highlights potential feature gaps based on recurring customer issues.

With the growing volume of customer requests, Fin helps us maintain consistent response times by handling simpler, repetitive support tickets. This frees up our team to focus on more complex and high-impact customer interactions.

Fin gathers all our support articles and information to create thoughtful and humanlike answers for our customers. The answers are typically spot-on (which is all determined by the information you feed it, so as a user, you can improve the answers) and supportive for our customers. Fin handles a large number of inquiries that come in, taking a load off our support team.

Sometimes, customers do not want to engage with AI at all, and it may take longer than they like to be routed to a human.

Answering a large amount of questions that come in with answers that are listed within our support articles. Taking a load off from our small team.

Fin is one of those tools that just works. It’s easy to set up, doesn’t require a dedicated a full team to manage, and gives us the power to keep our support team small and agile without sacrificing quality. The pricing is refreshingly straightforward, with no confusing tiers or upsells to worry about, which I really appreciate.

The best part? It’s genuinely helpful out of the box. We didn’t have to spend weeks training it or customizing complex workflows. It’s like having an extra team member who never sleeps and always gives solid answers.

If I had to nitpick, I’d say the analytics and customization options could go a bit deeper. While it covers the basics well, there are moments when I wish I could get more granular data or fine-tune how Fin handles certain types of queries. But honestly, that’s a minor ask given how much value it delivers already.

Fin helps us provide 24/7 support coverage without needing to scale our human team. It handles a large volume of common customer questions accurately and instantly, which means our support team can focus on the more complex or high-touch issues. That’s allowed us to stay lean and efficient without compromising on response times or customer satisfaction.

It’s also solved the problem of complexity that often comes with AI tools. Fin doesn’t require a huge implementation effort or a dedicated team to manage. It’s easy to launch, easy to maintain, and it works well with our existing Intercom setup. That simplicity saves us both time and money, and gives us peace of mind knowing we’re not spending weeks configuring something only to get mediocre results.

Fin is terrific in covering for our team outside business hours as well as helping us to learn more as we go.

Average limitations for AI models. No real dislikes to note.

It helps with limitations in working hours and covers for us.