Share product ideas and upvotes with our product team
I want to be able to set the redetect on closure for Fin and for teammates separately. For example, I want to be able to set up an attribute to be redetected upon Fin closure but not teammate closure. The reasoning here is that we want to rely on Fin to set the attribute when a human is not involved, but when a human is involved, we want to trust the human.
Many companies are now providing llms.txt for their documentation to make it more efficient for agents to consume content. It’s an index of publicly available markdown files. See, for example, https://cloudinary.com/documentation/llms.txt. If Fin supported this as a content source (or even just markdown files as a source) it would really help us, as the markdown files contain all code examples, and expand content that’s in collapsed sections - content that’s not accessible directly from the HTML.
Ability to define and save a company style and tone guide. As employees compose messages in chat or email, provide real-time, AI-powered coaching to help align with that guide. Before reps press send, run a final check against our tone and style guides and prevent delivery if the message doesn’t meet standards, with clear guidance on what needs to be updated.
It would be good to be able to like/dislike and correct fin responses in real time. There is the “improve answer” option, but I would like to directly tell fin that the answer it gave was incorrect or misleading, and course correct for future responses. If it gave a good response, I would like to indicate that somehow. I think this would help a lot with training it outside of help centre materials
FIN AI should be able to monitor how many tickets come in in a given timeframe. If this is suddenly spiking, and especially if the same topic is mentioned in multiple tickets, then FIN AI should escalate directly to human support with high priority. Plus spark and alert.
When viewing chats in the All view, it would be helpful to have a small icon or a dedicated column showing the Assigned owner for each chat. Currently, I have to jump around within the All view to find the conversations assigned to me while also monitoring other ongoing chats.It would be great if the All section could auto‑sort and clearly label where each chat belongs, including which ones are still unassigned. This would make it easier to stay organized without constantly switching between different views. You would get the benefits of the All and Unassigned/Owned Views together, with easy access to your own chats and the capability to go and support a teammate in a quick chat or grab an unassigned one.
It would be helpful to know just how many conversations are required before copilot uses conversations as a data source. As we are launching Intercom and this is one of the most important features, it is concerning to not know when this will become usable. It would also be helpful to allow Copilot to start searching existing conversations even without reaching the volume required -- even if it were rudimentary search as opposed to an LLM.
We would like to request the ability to customize the e-mail signature for the chatbot alone, to clearly indicate that the response is coming from a chatbot when replying to emails. Additionally, we would like the option to edit the job title for Fin AI in the same way it is possible for customer advisors. This would help ensure clarity and transparency in communication.The current email footer “Composed by AI” is not sufficient to set the right expectations for customers.
We currently handle all written customer communication in Intercom, but our phone support is handled through a separate telephony system that is not integrated with Intercom. Many of our customer advisors have started using Copilot, and both the advisors working in Intercom and those taking calls would like the ability to use Copilot without having to open a written conversation first.There are many situations where an advisor needs to ask Copilot a question that is not related to an active customer conversation. It would therefore be extremely helpful to have:A dedicated place in Intercom to interact with Copilot, Or a stand‑alone Copilot window/app that stays open independently of which system the advisor is working in.Our advisors often work across up to 16 different internal systems when assisting customers, and we rely heavily on internal routines for each system. Before asking a colleague, many would prefer to ask Copilot to save time—if only it were more easily accessible.A separate, persistent Copilot window would greatly improve workflow efficiency and make Copilot more useful across all parts of our support operation.
I am using Fin Tasks to automatically generate Linear issues for certain types of customer issues. Independent of this, we use the Linear/Intercom integration, When an Intercom URL is linked to a Linear issue updates from Linear to be reflected back to the conversation in Intercom. This would ensure when this automated Linear issue is resolved that a note is created on the ticket for the team to review and act on. To do this, Linear needs the Intercom URL as a “links” parameter or through an attachment API call. I can create a data connector that does this, but Fin Tasks doesn’t have the conversation ID or full URL as an attribute I can reference -- which seems like an easy addition and could be useful for others. Even the ID would allow me to construct the URL manually.
We’re really happy with the new escalation guidance functionality and have started using it to control which users can escalate to human support in chat. Current setup Paying users can always escalate to human support via chat Non-paying users are limited to AI support in chat and are guided to email if they need human support We’ve achieved this by creating a separate audience for non-paying users and configuring escalation guidance that encourages them to continue chatting with Fin or reach out via email when they ask to speak to a human This works very well functionally 👍 Feature requestWhat would be extremely helpful is the ability to see when escalation guidance has been applied directly in the conversation events, similar to how other guidance or automation events are shown today.For example: “Escalation guidance applied: Non-paying users – AI only” Timestamped and visible in the conversation timeline Why this matters It would help us verify that escalation logic is working as intended Makes troubleshooting and QA much easier Improves internal visibility and confidence when reviewing conversations Especially useful when multiple escalation rules and audiences are in use Right now, there’s no clear way to confirm from the conversation view that the correct escalation guidance was triggered, which makes it harder to monitor and validate the setup over time.Thanks for building a great feature already — this would be a small addition with a big impact on usability and trust.
hi all, when copilot suggests answers, there is currently no mechanism for the agent to provide feedback (good or bad) about the suggestion itself. The only options are either to provide feedback on the source, or tag the conversation/ticket for review (through tag or back office ticket0.Ideally, Copilot would have a 👍/👎 option that would flag the specific answer that needs reviewing, rather than the whole conversation.Thanks
Proposed feature changeAllow the subject line of the conversation email notification for Fin Chat Simple Deploy conversations to reflect the topic of the conversation, ideally based on the first message the customer sends.ProblemWhen Fin is enabled using Chat Simple Deploy, and a welcome message is set e.g. “Hi! I’m Fin”, the first line of the conversation email is always Fin’s welcome message.As a result, the subject line of the conversation email notification is always Fin’s welcome message, rather than the customer’s question. This causes all conversation email notifications to have the same subject line, making it difficult for customers to quickly scan their inbox and identify the correct support conversation.ScopeThis change is only applicable for user initiated conversations when Fin is enabled using Chat Simple Deploy.
It would be useful to surface the Conversation ID directly in the preview panel when viewing a conversation in Fin.At the moment, if I need the Conversation ID, I have to navigate away from Content and open Conversations → Analyse to find it. This adds unnecessary friction, especially when debugging, sharing references with colleagues, or cross-checking data.Having the Conversation ID visible in the preview panel would:Save time and reduce context switchingMake troubleshooting and internal referencing easier Thanks for considering.
I would like to create an attribute in Intercom that I can use to identify customers who are messaging us via the app version of our SaaS product and not our web portal so that I can tailor the Fin AI agent’s instructions to their device type.
Hello Team,I submitted this as a question. Didn’t get any confirmation. So I’ll attempt to submit as a Idea.When using AI compose we are seeing em dashes inserted in the text responses generated as shown below. Example:"Thank you so much for sharing those examples—they're incredibly helpful!" Here’s another example “Absolutely! Could you please share the Fax example details with me? I can then open a new ticket for you. Alternatively, if you prefer, we can continue working within this current ticket—whichever works best for you.”Is it possible to remove these em dashes and replace them with periods? This would create a more natural tone. Em dashes are a dead giveaway that someone is using AI-generated responses. I think AI Compose is a wonderful feature, and with this quick fix, we can take the AI Intercom experience to the next level. Thank you for your consideration.
Everyone is just getting onto AI and starting to understand Fin and what it can do. But the reality is: most teams are sitting on a mountain of dirty, unstructured historical data. That blocks us from using our own history for trend analysis, routing, automation, and learning loops in any meaningful way. Intercom already shows the pattern works: Topics Explorer retroactively assigns topics/subtopics via a “backfilling” step (currently described as the last 90 days) and then does daily inference on recently closed items. That’s the right idea — now we need it beyond topics, and we need it for tickets and conversations as a first-class capability. What I’m asking for Give us the ability to run a durable, resumable backfill job on existing tickets and conversations that re-runs Fin AI classification/enrichment and writes results back to our chosen fields/attributes — without requiring us to open/close every ticket. This should feel like a real platform primitive: “Re-run Fin classification on dataset X” “Apply current labeling logic to historical items” “Populate ticket + conversation attributes consistently” Why product leadership should care This isn’t a “nice-to-have automation.” It’s foundational for making Fin real at scale: Data integrity & reporting If only net-new items are enriched, every metric you show to customers is biased toward “post-AI era” data and hides the baseline. That makes it harder to prove impact, especially as you expand Fin AI Agent reporting metrics/attributes and reporting labels. (And yes — this ties directly to the discussion I already raised in my Fin thread replying to the upcoming reporting-labels/metrics work.) Customer journey intelligence Teams want to map the full journey: pre-sales → onboarding → incidents → renewals. If history is unstructured, journey analytics stays cosmetic. Fin gets smarter when customers can iterate Intercom is actively enabling customers to improve Fin with “content from conversations.” Backfilling is the missing partner to that: as customers refine knowledge and classification, we should be able to re-run enrichment over history to validate improvements against real volume. Outbound + automation is about to create even more data You’re already pushing Fin deeper into outbound follow-ups. Great. But if we can’t retroactively structure historical data, outbound-driven automation will widen the “old data vs new data” gap and make reporting/segmentation messier over time. “This already exists… just not where it needs to” Intercom has job-style infrastructure for exporting data (including date-range export jobs, with concurrency limits). So this request is straightforward conceptually: a managed backfill job, except it writes back Fin-derived fields rather than just exporting raw records. Also: Intercom notes bulk exports for conversation content are API-driven and time-bounded (and UI bulk export isn’t available). That’s exactly why this should be a first-class back-end job with safe controls, not a hacky “open/close everything” workflow. Trust, privacy, and “don’t make me create a compliance incident” If you ship backfill, it should be designed like a serious platform feature with guardrails: Scoped access controls (who can run it, which inboxes, which time ranges). Audit logs (who ran what, when, what changed). No customer re-notifications (analysis + field updates only). Data minimization (store only needed derived labels; don’t replicate raw content). Resumable, idempotent processing (safe restarts; deterministic outputs). And this is where it gets bigger than “support ops”: backfill can become a trust & safety accelerator if it supports sensitive information classification and remediation workflows. Intercom already has security features like PAN redaction. A backfill pipeline could optionally scan historical conversations/tickets for sensitive categories and support governance controls (e.g., classify → restrict → redact → retain evidence). This matters because customers are dealing with real legal obligations, including: GDPR DSAR / portability / erasure workflows (exports + deletion expectations). DMCA / legal takedown style requests (content traceability + audit). CSAM and other illegal-content reporting requirements (classification, quarantine, strict access, audit trails). If you want to be the AI-first customer comms platform, you can’t ignore that “AI over historic data” intersects with compliance and safety — and doing it deliberately is a competitive advantage. Minimum viable requirements Background job + progress + completion summary (counts, failures). Scope filters: inbox/team, date range, open/closed, tags/attributes, ticket types. Choose outputs: conversation attributes, ticket attributes, selected fields. Idempotent + resumable + rate-limited. Clear “no customer impact” guarantee (no state changes, no outbound triggers). Optional: dry-run sampling + estimated impact. User interview I want to talk to the team building this. I’m in Dublin, Ireland, and happy to do a user interview (also happy to meet anyone in Dublin). If helpful, I can share the volume we’re dealing with, the exact attributes/labels we want to backfill, and the reporting gaps this would close.
We operate as an agency and handle support for multiple brands. Each brand has its own inbox and Fin setup with brand-specific knowledge and tone.The problem:Sometimes customers send emails directly to our agency email address, even though their request relates to one of our brands. Currently, there's no way to have Fin analyze and categorize the conversation first, route it to the correct brand inbox, and then let that brand's Fin respond.Current limitations:Fin Attributes are only assigned at the end of a conversation or as part of escalation rules – not before Fin responds Using message content filters isn't practical because inquiries are too varied and individual Manual routing defeats the purpose of automationDesired workflow:Email arrives in agency inbox AI analyzes the conversation and identifies which brand it relates to Conversation is automatically moved to the correct brand inbox That brand's Fin setup handles the responseRequest:Please add:The ability to use AI-based classification to route conversations to the correct inbox before Fin responds The option for Fin to respond to moved/rerouted conversations without requiring the customer to send a new messageAs of right now, even if we manually move a conversation, Fin cannot respond to it – the customer would have to reach out again. This makes multi-brand support impractical.
We want to connect Fin to our ClickUp workspace via MCP to give it access to our internal knowledge base (Docs/Wiki). This would significantly reduce manual work and improve our resolution rate.However, we've hit a blocker: MCP tools that require array parameters as input are not displayed at all in Fin's tool selection. They are completely unavailable.Specific example:The ClickUp MCP tool get_document_pages expects a page_ids parameter as an array of strings (e.g., ["page_1", "page_2"]). This tool doesn't appear in Fin's available tools list because of the array requirement.Workaround attempted:We tried building a custom API integration, but it doesn't work reliably and requires ongoing maintenance.Request:Please add support for array/list parameters in MCP tool inputs so these tools become available to Fin.
Already have an account? Login
No account yet? Create an account
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.