Product feedback | The place for Zendesk users to come together and share
Skip to main content

Filter by feedback status

Filter by product

7571 Requests

Shaun11Newcomer

Real-time AI Suggestions for Voice Calls Should Use Auto-Assist ProceduresFeedback submitted

Please give a quick overview of your product feature request or feedback and note who in your organization is affected by this issue [ex. agents, admins, customers, etc.] (2-3 sentences)We would like Zendesk’s real-time AI suggestions for voice calls to support Auto assist procedures as a source, not only help center content. This affects agents who rely on live suggestions during customer calls, admins who maintain approved procedures, and customers who need accurate, complete guidance during live support interactions. Zendesk’s current documentation shows that real-time voice suggestions search help center content, while Auto assist procedures are a separate feature and cannot currently be configured on voice channels. [support.zendesk.com], [support.zendesk.com], [support.zendesk.com]What problem do you see this solving? (1-2 sentences)Real-time voice suggestions can paraphrase or simplify information from help center content instead of following our exact operational procedures. Allowing voice suggestions to use Auto assist procedures would help agents receive step-by-step, admin-approved guidance during calls, improving consistency and reducing the chance that important procedural details are missed. [support.zendesk.com], [support.zendesk.com]When was the last time you were affected by this lack of functionality, or specific tool? What happened? How often does this problem occur and how does this impact your business? (3-4 sentences)We are affected by this problem every time an Agent uses Real-time AI Suggetions in a call. This problem is especially important for voice calls because agents need concise, accurate guidance in real time and do not have the same opportunity to review long articles while speaking with a customer as they do during Messaging or Email tickets. Zendesk’s voice suggestion documentation currently describes help-center-based suggestions, while the Auto assist documentation says procedures are the primary knowledge source for Auto assist and that Auto assist cannot be configured on voice. [support.zendesk.com], [support.zendesk.com], [support.zendesk.com]Are you currently using a workaround to solve this problem? (If yes, please explain) (1-2 sentences)Yes. We are trying to maintain help center content and procedures separately, but that creates a gap: real-time voice suggestions rely on help center content, while our more precise step-by-step guidance lives in Auto assist procedures. Zendesk’s documentation says procedures can include macros, actions, agent instructions, help center content, other procedures, and ticket fields, but the real-time voice suggestion documentation does not indicate that those procedures can be used as a source for live voice suggestions. [support.zendesk.com], [support.zendesk.com]What would be your ideal solution to this problem? How would it work or function? (1-2 sentences)Ideally, Zendesk would allow admins to enable Auto assist procedures as a source for real-time AI suggestions during voice calls, either globally or by group, line, brand, intent, or call reason. During a live call, Copilot should use the real-time transcript to identify the relevant procedure, then surface the procedure’s approved steps, macros, actions, and agent instructions in the Knowledge or Copilot panel without requiring the agent to leave the call workflow. [support.zendesk.com], [support.zendesk.com]

Ian11Contributor

Ticket summary - automatic, not manualAccepted

Follow-up fromhttps://support.zendesk.com/hc/en-us/articles/8037649972634-Summarizing-ticket-comments-using-generative-AI OverviewWe would like ticket summaries to be automatically generated for all tickets, without requiring agents to manually click “View Ticket Summary.”Currently, the summary field only populates once an agent triggers this action, which leaves the field blank until that point. As a result, summaries are unavailable in Views, Explore reports, and the API until manually created. This affects agents, admins, and analytics teams who rely on complete, consistent summaries for reporting, automation, and intent analysis. ProblemOnly a small fraction of tickets (approximately 3% of the thousands we receive daily) currently have summaries. Most tickets (e.g. “Where is my order?” enquiries) are resolved quickly and never trigger manual summarization.Without these summaries, it’s difficult to perform meaningful, large-scale analysis of customer interactions beyond the assigned intent predictions. The missing summaries create significant blind spots in analytics, automation, and performance benchmarking. Problem SolvedAutomatic summarization would ensure every ticket includes a generated summary, allowing it to appear consistently in Views, Explore, and API exports.This would deliver full data coverage for analytics, intent validation, and automation, without requiring any manual agent intervention. Impact & FrequencyWe handle thousands of tickets per day, but only ~3% contain summaries due to the manual process.This limitation impacts us daily, hindering:Automation workflows dependent on the summary fieldLarge-scale analytics and intent modellingAccurate reporting and performance insightsThe lack of automatic summaries results in incomplete datasets and reduced reliability in Zendesk Explore and API-driven reports. WorkaroundWe developed an internal ZAF app (“AI Summary Auto-Trigger”) that attempts to simulate the summary click. However, ZAF does not allow interaction with UI elements outside its iframe, making this workaround technically infeasible. Ideal SolutionTicket summaries should auto-generate automatically upon ticket creation or update, with results immediately available via API, Views, and Explore datasets, eliminating any manual dependency.This would:Enable complete coverage for intent and topic modellingImprove anonymized analytics and performance insightsAllow validation of Zendesk-assigned intent taxonomy accuracy Without automated summarization, these capabilities remain limited to roughly 3% of our data, and manual triggering is not viable at enterprise scale.

Feature Request: Ability to prevent indexing of Zendesk Help Center system pages (e.g. /requests/new)Feedback submitted

Problem I’m trying to solveMy Zendesk Help Center includes system-generated pages (such as /hc/*/requests/*, including /requests/new) that are served by Zendesk and not by my own website. These pages are currently crawlable and indexable by search engines, and I have no way to control this behavior.Why the current functionality isn’t sufficient Zendesk’s robots.txt file is shared across all customers and cannot be customized at the account level. While noindex meta tags can be added via the Help Center editor, they are applied globally and cannot be configured on a page-by-page basis. Requesting URL removal via Google Search Console is only a temporary workaround and does not prevent future crawling or re-indexing. Desired outcomeI’d like a way to prevent specific Zendesk Help Center system paths from being indexed, such as: A Zendesk-managed robots.txt disallow for paths like /hc/*/requests/*, or An account-level option to apply noindex to specific system routes, or Any supported mechanism that allows customers to opt out of indexing for low-value or sensitive system pages. Business impact / contextThese system pages are not intended to be landing pages from search engines and provide little to no SEO value. When they appear in search results, they can: Confuse users who expect informational content but land on a ticket submission form Surface operational or sensitive pages that were never meant to be indexed Degrade overall SEO quality by allowing low-value pages into the index Being able to reliably prevent indexing of these pages would help ensure that only intentional, user-facing content appears in search results and would align Zendesk Help Centers with standard SEO best practices.

Allow customization of the anonymous request email-verification success page textFeedback submitted

Feature request summaryAllow customization of the text and button label on the anonymous request email-verification success page (/requests/verification/success).DescriptionWith the anonymous help center request verification workflow, end users land on a verification success page after confirming their email. This page (e.g. https://help.teamnext.de/hc/de/requests/verification/success) is rendered outside the Guide theme – it loads no theme assets (no script.js, no theme CSS, no document_head.hbs). As a result, the on-page text and button label cannot be customized by any supported means. ​​​Theme code, JavaScript redirects, and CSS overrides all fail here because the page never loads the theme. Zendesk support confirmed there is currently no supported way to edit this text.Business impactThe fixed wording doesn't match our brand voice. We use "Helpdesk" instead of "help centre" and refer to our "support team" rather than "an agent". Crucially, the page uses the formal German address form ("Sie"), while our help center consistently uses the informal form ("Du") — and we can't change it. The text also can't be localized per language (DE/EN). As the final step of the request flow, this mismatch is visible to every anonymous requester.Requested solutionA supported way to edit the heading, body text, and button label of this page — ideally per language, and with control over the formal/informal address form (German "Sie" vs. "Du") — so the text matches our brand's tone of voice. For example via an editable field in Admin Center, inclusion in the Guide theme, or translatable text strings / dynamic content.Who is affected?Affects all anonymous help center requesters since the March 2026 verification rollout.Screenshots/hc/en-150/requests/verification/success/hc/de/requests/verification/success 

Raida
RaidaNewcomer

Auto-Assist Admin insights should use a more actionable metric than full resolution timeFeedback submitted

Please give a quick overview of your product feature request or feedback and note who in your organization is affected by this issue.I’m requesting more actionable Admin insights for Auto-Assist performance. The current insight uses full resolution time, which can include broader ticket lifecycle time and may not accurately reflect agent handling or response speed. This affects admins and CS leadership who review performance insights, and agents when those insights make it appear that their workflow is slower than it actually is.What problem do you see this solving?This would help teams understand whether Auto-Assist is actually improving or slowing agent workflows. It would also prevent misleading performance alerts caused by customer wait time, Pending time, reopened ticket behavior, or other lifecycle factors outside the team’s direct control.When was the last time you were affected by this lack of functionality, or specific tool? What happened? How often does this problem occur and how does this impact your business?We were affected recently when Zendesk Admin showed an Auto-Assist insight indicating that resolution time for the “Enable Auto-Assist (Non-API tickets)” trigger rose by 859%, from 39 minutes to 6 hours, for the June 7 to June 14 insight period. After reviewing Explore reports for the same time frame, we found that reopened Auto-Assist tickets had a median requester wait time of 24 minutes, and reopened auto-reply tickets had a median requester wait time of 12 minutes. This made the Admin insight difficult to act on because the 6-hour figure appeared to reflect full ticket lifecycle time rather than typical agent response or handling time. This can impact our business by creating confusion around whether Auto-Assist is helping or hurting performance, and it requires additional manual reporting to validate the insight.Are you currently using a workaround to solve this problem?Yes. We are using Explore reports with requester wait time to better understand how long customers are waiting once tickets are back with CS. This helps, but it requires manual validation and does not change the Admin insight itself.What would be your ideal solution to this problem? How would it work or function?The ideal solution would be for Auto-Assist Admin insights to offer a metric focused on agent-controlled time instead of only full resolution time. For example, the insight could measure from when Auto-Assist is enabled, or when the ticket enters a support-owned status, to when the agent replies, solves, or otherwise moves the ticket forward.  

Elena11Newcomer

Calls from Suspended Users Do Not Generate Tickets, Resulting in Loss of Audit Trail and Interaction RecordsAccepted

Hi team,Currently, a suspended user can successfully place and complete a phone call through Zendesk Voice, but no corresponding ticket is created. Zendesk Support has confirmed that this behavior is expected: user suspension only prevents ticket creation, not phone calls, and calls from suspended users do not generate tickets.As a result, agents can handle lengthy customer interactions without any ticket record being created, leaving the interaction outside the standard support workflow.Business ImpactThis creates significant compliance, governance, and auditability concerns:Customer interactions can occur without any corresponding ticket record. There is no centralized audit trail containing the conversation details, agent actions, or interaction history. Quality Assurance, compliance reviews, and operational audits cannot reliably verify that the interaction occurred. Call recordings are not retained for suspended users, further reducing traceability. Organizations operating in regulated environments may be unable to demonstrate complete customer contact records.In our case, an inbound call from a suspended user lasted approximately 88 minutes and was successfully handled by an agent. The call was visible in Voice metadata but no ticket was generated, no interaction record was available through standard workflows, and the recording was unavailable.Expected BehaviorOrganizations should have the ability to retain an auditable record of calls from suspended users. At minimum, Zendesk should:Create a suspended ticket when a suspended user places or completes a call. Preserve call metadata and interaction details within Zendesk. Retain call recordings according to account retention policies. Clearly surface that the requester is suspended while maintaining a record of the interaction.Current BehaviorSuspended users can call Zendesk phone lines. Calls are connected to agents and can be completed successfully. No ticket is created. Call recordings are not retained. Interaction visibility is limited to backend Voice metadata.Why This MattersSuspension is typically used to restrict or control user activity, often for risk, fraud, abuse, compliance, or security reasons. Allowing successful voice interactions without generating a traceable record creates a gap in governance and customer interaction auditing.Even if the current behavior is technically expected, organizations should have the option to maintain a ticket-based audit trail for all customer interactions, including those involving suspended users.Requested EnhancementProvide a configurable option that allows administrators to:Automatically create suspended tickets for phone calls from suspended users. Retain call recordings and associated metadata. Ensure all customer interactions remain auditable and discoverable within standard Zendesk workflows. Generate reporting and analytics for these interactions in the same way as other customer contacts.

Entity detection - improvementsFeedback submitted

Please give a quick overview of your product feature request or feedback and note who in your organization is affected by this issue [ex. agents, admins, customers, etc.] (2-3 sentences)More control over Entity detection - include plurals, not include words as “misspellings” if they are a different word with a different meaning OR allow adding antonyms, so similar words could be proactivelly expluded. This affects the admins and data team. What problem do you see this solving? (1-2 sentences)Reporting is a really key feature for us and has to be precise, the more control we have over how entities get detected, the more acurate the data and reporting. When was the last time you were affected by this lack of functionality, or specific tool? What happened? How often does this problem occur and how does this impact your business? (3-4 sentences)We are currenly setting up our instance and noticed this in testing, we have to get this right as there can be a serious organisational impact if we don’t. Are you currently using a workaround to solve this problem? (If yes, please explain) (1-2 sentences)As far as we know there isn’t a workaround for this. What would be your ideal solution to this problem? How would it work or function? (1-2 sentences)Entity is able to detect synomyms and plurals of synonims, correctly notes misspellings and we could add antonyms too - words to be activelly excluded because they are similar to other words and captured as misspellings or because they can have multiple meanings. 

Multi-way branch / switch in Action Flows (more than if/else)Feedback submitted

ProblemAction Flow branching today supports only a binary split: If condition is met / Else.When a value can be one of many options (for example channel, brand, group, status, or any categorical field), there is no native way to route the flow into N paths in a single step.Current workaroundWe nest multiple Branch steps under each Else path (if A → … else if B → … else if C → …). This works, but it quickly becomes hard to read, maintain, and review—especially as the number of cases grows. It also increases the risk of configuration mistakes and makes simple process changes expensive.   Requested improvementAdd a multi-way branch step (switch / case), for example:Evaluate one variable once Define multiple cases (value equals A, B, C, …) Provide a default / else pathIdeal extras (nice to have):Match operators beyond exact equals (contains, is one of, is empty) Reusable case labels for clarity in the canvas Clear validation when cases overlap or a default is missingWhy it mattersMany real support processes are categorical (“for each X, do Y”), not binary. A native multi-way branch would reduce nesting, improve readability of production flows, and make Action Builder closer to how teams already model these processes.Expected benefitCleaner flows, fewer nested branches, faster edits, and lower maintenance cost for automations with multiple mutually exclusive paths.