Discussions

Hey Folks!We have been running Zendesk for our university student community for over a year now, and while our knowledge base continues to grow, we are now focused on improving its effectiveness rather than simply increasing the number of articles.Some areas we are currently exploring include:• Identifying outdated or underperforming content• Measuring true article effectiveness beyond page views• Improving searchability and content discoverability• Reducing recurring tickets through knowledge enhancements• Creating stronger departmental ownership of articlesFor organizations that have had their knowledge base in place for a few years, what were the biggest improvements you made after the initial implementation phase?Would love to learn from your experiences.

Hello Everyone!Is there any way in Zendesk (Explore or native Time Tracking) to report time spent at the individual update/comment level per agent, instead of only ticket-level totals?Specifically, I want to see each update’s duration (not repeated totals) and aggregate it per agent over a given date range.Is this possible natively, or only via API/workarounds?

I think AI is pushing content strategy from “Do we have an article for this?” to “Can this content produce a clear, trusted answer?”That has been the biggest shift for me.A lot of help center content was originally written for humans to browse, skim, and interpret. But when AI becomes the delivery layer, the content has to work differently. It needs to be easier to retrieve, easier to summarize, and less dependent on a person connecting the dots.The main changes I’m thinking about are: Making answers more direct and explicit Breaking long articles into cleaner sections Using headings that match real customer questions Removing duplicate or outdated guidance Adding the “what to do next” when an answer has exceptions Paying closer attention to the questions agents keep answering manually I also think governance becomes a much bigger deal. If two articles say slightly different things, a human might notice the nuance. AI may just pick one and confidently deliver the wrong versio

I am Admin for our ZD instance, and I work on the business side, not IT. Back in a former life, I was a software engineer, but not anymore. So I use Power Automate when I need to do code-like things. I have a number of Power Automate flows that invoke the ZD API (via the HTTP action) and teach myself the API via Postman. Auth for both Power Automate and Postman uses API keys.API key auth is being eliminated for Support.I have no idea how to transition to OAuth for my needs. There is not documentation that provides me any guidance. I have no experience with OAuth, and this looks like something I’ll need IT assistance to set up - but I have no idea what to even ask for. Can someone provide me some guidance?Note that Exploring Zendesk APIs with Postman on the ZD developer website instructs to use API key auth for Postman.

What level of control do administrators have over AI-generated responses? Specifically, can organizations customize the structure, formatting, tone, response templates, or output behavior to align with internal standards and business requirements?

Does Zendesk Knowledge Copilot support the ingestion of customer-specific internal data sources beyond Zendesk Help Center content? If so, what options are available for integrating and maintaining proprietary knowledge, and how does the AI leverage that data when generating responses?

We've been building a custom dashboard on top of the WFM API and were looking forward to pulling time-off data into it.However, when we call GET /wfm/public/api/v2/timeOff using Basic auth (email + API token), we're getting a 403 JWT not found error. Our existing WFM report endpoints (adherence, productivity) work perfectly fine with the same Basic auth setup, so it seems like the time-off endpoints may not yet support Basic auth the same way the report endpoints do.Is this a known limitation, or is Basic auth support for the time-off endpoints planned? Would love to know if there's a workaround or a timeline for when this might be available. Thanks!

pls help what ErrorMessage I will get when JWT token is expired? "401" or "HTTP 401" or something else?"The most common errors encountered are HTTP 401 unauthorized errors or expiration errors caused by using a JWT with an expired timestamp. In this case, a new JWT should be generated and a call made to LoginAsync."

I am using a Python script that exports (to CSV) a set of tickets using the Search API (/v2/search.json). In my script, I'm also making a call to get all ticket comments and combine into a single field (/v2/tickets/{ticket_id}/comments.json.  For the majority of the tickets/comments, everything exports as it should and all comments combine, but it looks like that if a ticket has a quite large amount of comments, when exported to CSV, it overflows to multiple lines below the ticket and makes the ticket data unusable. I've included a screenshot (with some content hidden) to give you an idea of what is happening. I have tried getting the comments by using 'body' & 'html_body', but it doesn't make a difference.  Any thoughts on this? Is there a limitation in how many characters can be included in Excel and/or a CSV column? 

Can you please show me how to export to excel a list of all of my tickets with their corresponding comments?

Hi everyone — I’m trying to analyze customer questions across a subset of tickets (for example, tickets related to a specific product). Ideally I’d like to export the ticket content and conversations into a CSV so I can review patterns in customer questions.Right now I can view tickets individually, but I can’t find a way to export the ticket body / conversation content for a filtered group of tickets.Is there a recommended way to:• export ticket conversations (not just metrics)• filter by keyword, tag, or product first• then download the data for analysis?We receive about 1,000 tickets per month, so reviewing tickets one by one isn’t scalable.Any guidance on the best workflow would be appreciated.

We have noticed, that all the managed accounts are missing AI Agent’s Allowance usage -information in their Admin Center. As we are a Zendesk-reseller in EMEA, this feature would be very crusial for us and our customers, in order for them to track how many Automated Resolutions they currently have left. Based on that information, they would be able to decide when to purchase more Automated Resolutions, before they run out of them. Without the information, it’s also impossible for them to know whether they are currently running on the pay-as-you-go pricing (2€), or the regular pricing (1,50€) - this issue would also be solved by showing the Allowance usage -information in the Admin Center.Below are the screenshots how a regular account looks like, compared to how a managed account looks like.We hope this would be developed soon!  

Hi everyone,I’m looking for the best practice to route outbound WhatsApp replies to specific Use Cases within the Zendesk Advanced AI Agent.My Scenario:I have 3 different Meta WhatsApp Templates used for proactive outbound messaging. All three templates start with a "Yes/No" Quick Reply button. In the Advanced AI Agent (Bot Builder), I have created 3 separate Use Cases (one for each template) to handle the different follow-up flows.The Problem: When a customer clicks "Yes" or "No", the AI Agent fails to identify which Use Case it should trigger. Since the reply is a simple "Yes", the NLU (Natural Language Understanding) cannot distinguish which template the user is responding to, leading to incorrect routing or a generic fallback.What I’ve tried:Sending metadata during the outbound trigger. Identifying the Template ID in the API call. Despite this, the Advanced AI Agent doesn't seem to "read" this context to select the correct Use Case automatically.My Question: What is the most effect

[Issue Details] Previously, the connection test worked perfectly when clicking the "Test your website" button under Step 3 in the "Install Google Tag Manager" screen.However, without making any changes to the configurations in either GTM or Zendesk, it suddenly stopped connecting. Now, it consistently displays an orange warning message stating, "No Google tag was detected on your website."[Steps Checked / Verified] I have already verified the following points:Source Code Output: I checked the page HTML using "View page source" and confirmed that the official GTM container snippets (starting with "GTM-") are definitely and correctly outputted within both the <head> and <body> tags. Actual Tracking Status: When users access the site via regular browsers, GA4 (Google Analytics 4) successfully and normally tracks data, including page views visible in the Realtime report. Zendesk Security Settings: In the Zendesk Guide admin settings, the "Display unsafe content" option is check

Hi.I want to set up a trigger:If a specific domain send an email to our support mail address, then it needs to be routed to a specific Zendesk group. However i cant find a condition based on domain. I have added the domain to the organization.Is this possible?This is is Zendesk Support.

After 15 Years, Zendesk Makes A Major Update Open. Pending. On-Hold. Solved. Four words.  Four categories that defined a generation of requests made from end users from startups to enterprises.  When Zendesk launched in 2007, it led to a revolution in how the industry viewed customer service. For all of the positive change, there has been one question that has surfaced in configuration after configuration… “How do we change the status of tickets to something that meets our needs better?”  Until last week, the answer was: You don’t (or, more appropriately, let’s create a custom workflow that extends the functionality of the tool beyond its originally designed functions and purpose).  But when you have more than 170,000 global customers, sometimes you just need a bit more flexibility… and Zendesk has finally agreed. The Back Story As many people are already aware, Zendesk launched in 2007 as a game-changing CX ticketing platform.  Although it wasn’t the first CRM

Hello, I noticed that I can't seem to revoke/delete an oauth token via the API. I am using an authorization token with full access. I try:curl https://$ZENDESK_SUBDOMAIN.zendesk.com/api/v2/oauth/tokens.json -v -u $ZENDESK_EMAIL/token:$ZENDESK_API_TOKEN And I get a page of results like{"tokens":[{"url":"https://xyz.zendesk.com/api/v2/oauth/tokens/1234567890.json","id":1234567890,...If I then do:curl https://$ZENDESK_SUBDOMAIN.zendesk.com/api/v2/oauth/tokens/1234567890 -v -u $ZENDESK_EMAIL/token:$ZENDESK_API_TOKENI get{"error":"RecordNotFound","description":"Not found"}I tried using the OAuth Client page in the admin, found the token I was interested in by paging through lots of pages manually (you need a better UI on the Oauth Client > View Tokens page), but I found it, and tried to delete from there.  It did not delete there, either, I also got a 404 I tried another token and was able to get it. I didn't want to actually delete it, so I just retreived it, but I did no

Is it possible to search for Macros in the admin panel and sort these by usage(per hour, day, week, month). Right now, sorting by usage only works for all macros. Alternatively, can the reporting feature be used to create a list of specific macros and their usage? 

Hi, we are using Zendesk Advanced AI and our chat bubble is an ongoing conversation (it doesn’t create a new one every time the customer opens it but it’s a continuous conversation). i noticed that dialogues or procedures don’t get applied when the chat conversation isn’t completely new. need advice on how to include existing or reopened conversations on procedures/dialogues. 

I know that these two are no longer distinct technically but reading the documentation it isn’t super clear to me if there’s actually a significant change as Zendesk would have you believe. I can see that the upgraded AI can access API’s, and that might be something that would be a game changer as I’ve just started messing around with managing our help center via API, but if anyone has done that already, I’d love to hear examples of how you’re using that. 

Curious how other Zendesk teams handle early escalation detection.In my experience managing support ops, the pattern is almost alwaysthe same — the signals are visible early (repeated contacts, legallanguage, competitor mentions) but they're buried under ticket volumeand only get flagged after the customer is already angry.I've been building tooling around this problem and recently submitteda Zendesk Marketplace app (EscalationAI — pending approval) that scorestickets in real time. Happy to share access for anyone who wants to testit on their actual tickets.But genuinely curious — what's your current process? Manual tagging?Supervisor review? Something automated?

Let’s talk about Copilot Ticket Summaries!I didn’t really want to talk about them. In fact, we already did talk about them, tried them, and decided not to buy them. Then you decided to make it weird by forcing them on everyone as a freemium upsale, in the hopes we’ll love them so much we want to pay a severely discounted price for them for a short time before the AI compute bill comes due for us all.  How do we opt out of having this unwanted and already-rejected feature enabled for our account? In advance, if you please.I don’t want to have to wait for them to be switched on to switch them off. And if I may say so: This is a pattern of feature rollout with Zendesk that is starting to get annoying. 

As part of our ongoing goal to improve our help center articles for AI AND human consumption, we’ve been attempting to do some format standardization. Zendesk has some features that can help towards this, but there was a significant gap that I found, which led me to using Claude Skills to assist.What Zendesk CAN do: you can create help center article “templates” to guide the creation of new articles. you now have more flexibility to select a specific font color. It used to be a limited selection, and our brand color was not included! But I see that recently, that has changed and we could put in our exact brand color #784FFF if we wanted.What Zendesk CAN’T do:You can’t apply a template to existing articles. You can’t add borders to a screenshot. My previous workaround was to create a 1x1 table to act as a border but it was not pretty! You can’t ask Zendesk to remember your general formatting rules and ask it to apply them across many articles. How I have used Claude:I worked with Claude

ProblemWhen having extensive flows I can’t easily connect multiple templates together in one use case/flow.What I would like.User ask question (case)Template A - Authorize user and create temporary access token for that user. Template B - API call - show list of entities and select entity (`/api/v1/entities/`) Template C - API call - show list of entities and select entity (`/api/v1/entities/id/items`) Template D - API call - present solution for entity item  What happens now;Case; I have problem with item.Set `destination` parameter (D) Link to template A: On success → destination === D→ link to template B Template B: On success → destination === D → link to template C Template C; On success → destination === D → link to template C Template D; Answer question for user. This creates dependencies between templates instead of a flexible flow you can plug between flows the D is now tied to B and can’t be use in a different case where I need to swap D with another case.What I would pref

We’re continuing Zendesk Voices, our series highlighting the expertise of Zendesk customers and partners.For our second post, we want to hear how your content strategy is evolving as AI becomes the primary way people get answers. Share what is changing, what you are learning, and how you are adapting your content to support customers in this new channel.Share your story right here in our Knowledge forum, then visit the Spark reward hub for full details and to make sure you are eligible for rewardsOnce we review and approve your post, you will receive points.