Product Wishlist | Community
Skip to main content

    Idea Pipeline

    Filter by idea status

    Browse by category

    518 Ideas

    Grace Ndabananiye
    Grace NdabananiyeNew Participant

    Export Individual Reviewer Notes from Monitor ScorecardsSubmitted

    Summary:Manual reviewer notes added to conversations within a Monitor should be exportable, either via the Dataset Export (CSV) or surfaced in the Reporting API, alongside scorecard scores.The problem:When manually reviewing conversations in a Monitor, reviewers can add qualitative notes to explain their scoring decisions. These notes are currently only visible within the Monitor UI, one conversation at a time, and can't be exported via CSV, the Reporting API, or S3 export.This creates a noticeable gap: the qualitative insight that gives context to scores, the why behind a pass or fail, is trapped behind individual conversation views. To extract patterns, a reviewer has to manually open and copy notes from each conversation, which isn't realistic at scale.What I'd expect instead:Reviewer notes should be included in the Dataset Export (scorecard_evaluation or scorecard_attribute_evaluation datasets) as an additional column, and ideally also surfaced via the Reporting API so teams can build dashboards or send data to external tools.Even a basic bulk export, such as a CSV of conversation ID, attribute, score, and note, would make a meaningful difference for teams using Monitors for structured QA workflows.Impact:Without exportable notes, qualitative QA analysis at scale just isn't practical. It limits Monitors to surface-level pass/fail tracking, rather than the richer insight tool they're designed to be.

    Jordan CantorNew Participant

    Feature request: A built-in QA sandbox for Fin AI Agent and OutcomesSubmitted

    Building a reliable Fin AI agent and Outcomes means testing more than the happy path. We need to know whether Fin follows the right procedure, asks for missing information, handles exceptions, and hands off when it should. We also need to know whether a change that fixes one workflow breaks another.Today, Preview is useful for debugging a single conversation, but running tests one by one takes too much time. Evals help with batch testing, but their scope and run limits make it hard to test several workflows together, give each workflow its own success criteria, and rerun the full set after every change.I’d like Intercom to add a built-in sandbox enviornment where teams can:Test safely. Run draft configurations against simulated conversations without contacting customers or triggering live actions. Simulate tool responses, failures, and handoffs. Batch test across workflows. Run refunds, onboarding, and escalations in the same batch, with separate success criteria for each scenario. For example, a refund test should pass only if Fin verifies eligibility before attempting the refund. Run QA automatically and without limits. Generate test cases from Procedures, save them as reusable suites, and rerun them on demand, on a schedule, or whenever the configuration changes. Sandbox runs should be free and unlimited, with no monthly quota on simulations or evaluations. See what passed, what failed, and why. Show success rates by workflow and criterion, compare results with the published configuration, and link each failure to the conversation and the step where Fin went off track.The goal is to make QA part of building rather than another feature that has a financial barrier. Before publishing a change, I want to run the full regression suite and see whether Fin follows our protocols and achieves the intended outcomes. We should be able to do that after every meaningful change, as often as we need without limits.

    Melissa MontesNew Participant

    Product Request: Role-Based Knowledge Visibility in CopilotSubmitted

    We would like to request the ability to configure role-based knowledge visibility within Intercom Copilot, allowing administrators to control which knowledge articles and guidance are accessible to agents based on their support tier, team, or role.Business ContextOur support organization operates within a tiered escalation workflow. Tier 1 agents handle initial troubleshooting and workflow steps before escalating to Tier 2 agents, who continue the process and follow additional guidance.As we scale our knowledge base and AI-assisted support workflows, we need to ensure that Copilot surfaces the appropriate guidance based on the agent’s role and responsibilities.We would like to understand whether Copilot can distinguish between agent tiers and restrict the retrieval or display of content accordingly.Requested FunctionalityWe would like the ability to: Configure knowledge articles or content sources to be accessible only to designated agent roles, teams, or support tiers. Restrict Copilot from surfacing Tier 2-specific guidance to Tier 1 agents. Dynamically tailor Copilot responses based on the agent’s role and escalation stage. Maintain a centralized knowledge base while supporting different levels of guidance and access. Expected Benefits Ensure agents receive guidance appropriate to their responsibilities and workflow stage. Reduce the risk of agents receiving instructions intended for higher-tier specialists. Improve the accuracy and relevance of Copilot responses. Support scalable knowledge management and AI adoption without duplicating content across multiple knowledge bases. Strengthen governance and control over AI-generated guidance. Example Use CaseA Tier 1 agent is working through an escalation workflow and asks Copilot for the next step.Copilot should surface only the guidance relevant to Tier 1 responsibilities and the initial troubleshooting process.Once the conversation is escalated to a Tier 2 agent, Copilot should be able to surface the additional guidance required to continue the workflow.