Share product ideas and upvotes with our product team
It appears in Checklists, you can choose whether to launch URLs in the same browser tab or a new tab. I don't see that same option for links in Posts and Chats. We have a single page app and it's important for us to provide links to different areas of the app without loading each in a new window.For support, our agents often guide customers how to update their billing info (/admin/billing) or where to upload messages (/admin/messages). It would be extremely valuable to able to send links to these areas in the app w/o loading new windows.The exact same thing applies to proactive support where we want to guide new users to different features without opening the app in a bunch of windows.
Ability to edit titles of outbound messages after they're sent so they don’t just remain “Untitled” forever (organisational nightmare) if you forget to add a title. Or even just prompt us to add a title upon message creation / before sending so we don’t get into this mess.
I often forget to give an outbound message a title before it goes out - then I’m unable to correct my error and give it a title afterwards (meaning it shows in my list as “Untitled”).I want to be able to change the title at any time, and I can’t see any reason to prevent this.(My initial error would be mitigated by an unset title defaulting to the subject line, which I know is already a feature request - but even if that was introduced I’d want to be be able to change it.
We would love to have a 5-10 ding option instead of a single ding/sound when we get a message. We’d also like to suggest for that to repeat if we have not responded within 5 or 10 minutes. A quick short ding is being missed by our staff.
Currently, all spam filtering is automatic, with no options to adjust its sensitivity or manage spam (easily). I propose the following features to improve spam handling significantly:1. “Mark as Spam”Agents should be able to mark conversations, senders, or entire domains as spam.2. “Mark as Not Spam”Similarly, conversations, senders, or domains should be markable as not spam, with:• More granular controls, such as regex matching, rules, and white/blacklisting.• Integration into workflows for easy automation if “Mark as (not) Spam” options are available.3. Spam Box SortingAgents should be able to sort spam using existing headers.• Side note: This functionality should also be available in the standard Inbox “Table Layout.”4. Batch SelectionEnable agents to select multiple spam messages at once for actions like “Mark as (not) Spam” or deletion.5. Permanent DeletionConversations in the spam folder should be permanently deletable. Spam should not accumulate indefinitely.• Currently, we handle this via API and custom actions, but these conversations still appear in reports. Spam-marked messages should ideally be excluded from reports by default or have an option to exclude them easily. Thank you!
Our company has developed an AI to help patients to be directed to the appropriate healthcare professional within the optimal timeframe for their condition. We regularly update our AI and notify teams through the Intercom platform. Having an exact figure for the number of people who have seen our publication would be helpful, as we often receive reports requesting changes that were already addressed in the latest update.I really liked the idea of having the application on the home screen, as it allows teams to easily see the update.
We use the metric Hold Time for agent bonuses, and we were informed via Fin that it isn’t available in the UI nor is it available via API (I’ve tried searching for a method to calculate it using this Conversation Endpoint and it wasn’t very clear what conversation_parts we can use to calculate).Would love to be able to have this data readily available; we don’t want our agents putting our customers on hold for long periods of times, and we don’t have an easy way to measure this.
Hi,I really seach with various keywords but can’t find a similar request. For phone function there’s a huge need for an integration with custom SIP PBX. Most of the customers still call the phone systems of the companies. These companies all seek for a solution as a Voice agent but can’t figure out how to transfer the call to an advanced AI agent and take it back to their staff back. Elevenlabs have this SIP integration function but transfering back to human is not working as expected.
There isn’t the ability in the UI to bulk apply “Not Spam”. We recently had over 200 conversations listed as spam that should not have been and I have to go in 1-by-1 to click Not Spam. That’s not a great experience.
Currently, when using the messanger sdk I don’t know if a user has started a conversation. I would like to react to that event but currently it is not easily achievable.I have to poll the Intercom API for all the conversation created events and then filter for that particular userIt would be much much easier for me to just react to a callback from the intercom messanger sdk that a conversation has started and then perform business logic that I want to have. Please consider addign this callback to the messangeer sdk.Similar callback exists in the mobile SDK according to FIN AI and it is called IntercomDidStartNewConversationNotification. I’m not sure how to use it since I am now developing this on a mobile application.
I would like a reporting metric to be able to view the number of customers that have opened more than 1 conversation or ticket within the filtered period.Basically any user id where the count of conversation ids is greater than 1, within filtered parameters.Alongside a count of distinct conversations and distinct users within the filtered period, this will give our team a much better idea of how well issues are being resolved the first time, if the same issues are constantly coming back up, and if the same customers are consistently opening new conversations.
Given we need to reference knowledge base articles from external systems (ex: in banners, in release emails, etc. )Once I’m done editing my article and publishing I want an easy way to get the URL link to that article directly from the editor. I think that was present previously but got removed. pls add it back.
It would be helpful to be able to customize more options for the Phone channel, such as:Edit the default delay after phone menu options are shared before they are repeated (20s is too long and leads to more abandoned calls when they don’t pay attention the first time around) Edit the speed of the AI voice (the voice is a bit too fast and leads to customers missing the phone menu options that are stated) (nice to have) More options for the hold/wait musicThese customizable options will help our customers navigate the phone menus easier, and reduce abandoned calls or calls that get routed to the wrong team.
I think it would be a great improvement to have the ability for the AI to be able to respond to conversations before routing them to a team member, either providing more context as to why the routing is taking place. This would be great for some guidance changes that would allow for responses before closure etc.
At present when taking the ‘Overview’ report as an example, phone calls are automatically grouped under “Closed With No Reply” in reporting. Whilst this classification makes sense for missed calls, it doesn’t accurately reflect answered calls, which are genuine customer interactions handled by teammates.Why this matters: Customer engagement is underrepresented. An answered phone call is a meaningful teammate interaction, even if no written message follows. Grouping these with “no reply” means reporting doesn’t reflect reality. Managerial insights are skewed. Team responsiveness and workload data appear distorted, making it harder to track performance, allocate resources, or demonstrate value. Consistency across channels. With Fin Voice resolutions expected to be categorised correctly, teammate calls risk being unfairly bundled into “no reply”, creating inconsistencies in reporting. Proposed solution: Keep missed calls grouped as “Closed With No Reply.” Reclassify answered calls as “Teammate replied,” since these are direct agent engagements. Ensure consistency so that both agent calls and Fin Voice resolutions are reflected properly in analytics. Impact: Reporting would become more accurate and trustworthy. Managers would gain visibility into the true volume of customer conversations handled by agents. Support performance could be measured consistently across chat, calls, and AI voice interactions. This small distinction would have a significant impact on reporting clarity, helping teams to better understand and optimise the customer experience.
We tested the Slack integration and it workes fine.I´d suggest 2 things:To make it possible to create a ticket directly from these threads. When setting status to resolved in a ticket (since this wont be possible in a conversation I guess), create a trigger that reacts on the initial Slack conversation with an emoji, eg checkmark ✅ for solved. This would make it easier to see if a thread in Slack already is solved.
The new dataset export page promises, “ It’s faster, more flexible, and supports everything the old version did—and more.” This isn’t true when fields that we were able to add to reports are missing. In the legacy dataset export tool, I was able to add a column that showed up in the export as “Participated (email)” that would list the email addresses of everyone included in the conversation.I’ve attached a copy of a recent dataset export file. Column J has the field that is no longer available.This field is essential to my weekly reporting and analysis on my client conversations. Please add this field to the new dataset export tool ASAP.Thank you,Kieran
We increasingly use AI tools that generate documentation in markdown. The current Articles editor doesn’t support markdown or auto-convert it, which makes updating to Intercom’s own formatting tedious and error-prone.Why it matters:Faster authoring: paste markdown or docs without manual reformatting Fewer formatting mistakes: consistent headings, lists, code blocksWhat would help:Native markdown rendering in the editor, or “Paste as markdown” auto-conversion, or Import/Upload a .md file with mapping to Intercom blocksMarkdown is a basic documentation standard—supporting it would streamline knowledge base workflows for many teams.
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.