Support

I am not seeing a way to set agent status based on schedule.  When I first started configuring Omnichannel routing, all agents were set as Available even though many work different shifts.  It would be great if we could apply a schedule to an agent's profile so that we can ensure tickets do not get assigned to them off schedule (unless the agent manually updates their status).

I have seen some conversations on this that are several years old, but I wanted to know if there was a solution for Contributors or Light Agents that would allow them to post Public comments on tickets where they are the Requestor or CC’d in the ticket.   

I’ve been thinking about what happens when a customer replies to a ticket after it has already been solved.For support teams handling a large number of tickets, reopened conversations could easily get mixed in with new requests.How do other Zendesk teams handle this?Do you use triggers, automations, views, tags, or ticket priorities to make sure reopened tickets are identified and handled quickly?I’d be interested to hear what workflows have worked well for other support teams.

Hey, we’ve been having issues syncing some comments/internal notes from Zendesk to Jira. It appears that comments that have a complex HTML design/format are not syncing directly, and they appear as a blank space on Jira (e.g. email responses to a ticket that have a signature, messages in bold, etc).Also, it seems that attachments aren’t syncing to Jira. We had a .mov file not sync to Jira.I’ve checked on Jira and we do have Wiki Style Renderer activated for comments.Any help would be appreciated!   

Hello Zendesk Support,We use a third-party bot with Sunshine Conversations and AI-agent-only messaging tickets. We want the standard Brand field and a custom AI status dropdown populated when the ticket is first created.The Brand is active. When creating a conversation using POST /sc/v2/apps/{app_id}/conversations, we include:{ "type": "personal", "participants": [ { "userId": "000000000000000000000002", "subscribeSDKClient": false } ], "metadata": { "dataCapture.systemField.brand_id": "900001", "dataCapture.ticketField.900002": "handling" }}Creation succeeds with HTTP 201. However, the AI-agent ticket generated by the first user message retains the default Brand and an empty AI-status field.The same metadata key format works during human handoff:POST /sc/v2/apps/{app_id}/conversations/{conversation_id}/passControl{ "switchboardIntegration": "zd-agentWorkspace", "metadata": { "dataCapture.systemField.brand_id": "900001", "dataCapture.ticketField.90000

Hi everyone,We built an RMA hardware inspection report in Zendesk Explore using a Standard Calculated Attribute that renders a thumbnail preview from a custom ticket URL field:IMG([Ticket custom field: Attachment Preview URL], 120, 60)Inside the Explore report builder, the table rows automatically expand vertically to fit the 60px image height. However, when we add this report to a Dashboard or export it via Scheduled PDF delivery, the table row height locks to ~32px and clips the bottom half of every thumbnail image:Full screenshot of the clipped rows in Dashboard export: https://media2url.com/m/579ad6b9a97d40 (uploaded via screenshot to url)Things we have already checked under Chart configuration > Columns:Confirmed that Text wrap is checked on the calculated attribute column. Tried leaving the height parameter blank in IMG([Ticket custom field: Attachment Preview URL], 120), which still gets clipped at the exact same row boundary.Is there a setting in the new Explore Dashboard bu

I noticed that recently when we are responding to customers with their order information, zendesk is autoreplacing the numbers with a link to a ticket that does not exist. For example, when I tell a customer “there is a problem with order #1234567,” the 1234567 becomes link to ticket number 1234567. This goes to the customer so when they click it, it takes them to an error page that says “oops the page you are looking for does not exist”This is a terrible feature and I do not see what setting to turn off for this. 

Welcome Let me introduce you to kBackup! This little app will allow you to back up your help centre content to your local hard drive at the click of a button.  Using this app makes backing up your help centre content a breeze.  Features The app currently has several features and may be expanded to include additional features if demand arises. The app will back up your help centre content to a specified folder as HTML pages in the format of [ArticleId].html. The app will first verify that the user exists in the given Zendesk domain. This is included for security reasons, so only users who exist in the domain can back up the content. The app will also back up any images that are embedded in your article and save these in the same backup folder. The app will work for any Zendesk help centre, even if a custom domain is used. Using the app to back up your help centre content is completely safe and will not affect any existing data. It will simply take a copy of it, so there is n

Sometimes customers miss an email, forget to reply, or simply don’t engage with the first message. In these cases, sending a follow-up can help keep the conversation moving without creating a new campaign from scratch.If you’ve already sent a mass email campaign, you can create a follow-up based on how recipients interacted with the original message.The workflow is pretty straightforward:1. Open the Details of the completed campaign and select Create Follow-up Campaign2. Choose who you want to follow up with: Customers who opened the email Customers who didn’t open it Customers who replied 3. Give the campaign a name and continue with the usual campaign setup.4. Аdjust the message or schedule if needed, then launch the campaign.This can be useful for reminders, re-engagement, onboarding, or announcements where the next message depends on what customers did with the first one.

The password expiry notification should not include a URL as it sets off phishing alarm bells to recipients. It should just instruct to log in directly at the agent’s established zendesk login page to change their password. 

Dear Zendesk Team,We are considering an integration with Microsoft Teams for another unit that provides internal support to our employees rather than external customers.Our main goal is to have the following workflow:Employees should use Microsoft Teams as usual, without needing to interact directly with Zendesk. The support team should use Zendesk to receive, manage, and respond to employee requests as tickets. Messages sent by employees in Teams should be received in Zendesk, and the support team's responses in Zendesk should be delivered back to the employee in Teams. We would also like to know whether it would be possible for the support team to initiate a Teams chat with an employee directly from Zendesk, even if the employee has never contacted the support team through Teams before. Is it possible to integrate our employee directory/data with Zendesk so that employee information is available when creating or managing requests?We would also like to understand how this could be kep

I have an issue with Api call for side_conversations and scope definition in header. I’m trying to  take json for tickets using:https://{domain}/api/v2/tickets/{ticket_id}/side_conversations Token request: scope requested = 'side_conversations:read', HTTP 200 Token response body: { "access_token": "<masked, fingerprint=36c17344>", "token_type": "bearer", "expires_in": 1799 }but I’m getting: 403 - {"error":{"code":"InvalidScopes","title":"Cannot access this resource","message":"Missing one of the following required scopes: side_conversations:read,read","allowed_scopes":["side_conversations:read","read"]}}I tried also scopes: “read”, “tickets:read” and without scope in header.All examples failed. There is no error for example for comments endpoint where url is similar. I can run with no error : /api/v2/tickets/side_conversations/events?start_time={start_time}, but jsons don’t contains all fields.How to define scope to run my api call with bearer token without error?

AI is evolving quickly, and keeping up with new capabilities, use cases, and terminology can feel like a lot, especially when every team has different workflows, priorities, and customer needs.That’s why Zendesk is creating different ways to learn, explore, and find inspiration. Whether you’re looking for product guidance, real-world examples, industry perspectives, or learning resources, here are a few places to start. Explore Custom Agents in practiceJoin the upcoming What’s New at Zendesk session to learn more about Custom Agents and how they can be adapted to your business’s workflows, knowledge, systems, and approval processes. The session will include practical guidance and live Q&A to help you understand where Custom Agents may fit into your service operations.📍 Register hereExplore what’s nextThe AI Summit 2026 will bring together perspectives from AI and service experts, product insights, technical demos, and conversations about how teams are applying AI to real business

Hello,We're coming from a helpdesk system that allowed us to simply have our agent emailed when a user responds back to a closed ticket, rather than reopening the ticket (and emailing the agent about the reopen). We ended up cutting out tons of "Thanks" email reopens this way, which ultimately make our reopen stats useless and adds more time for an agent to close the ticket, which causes more email to the end user.Is there a way to do this in ZenDesk? Just stop, cold turkey, all response-to-closed-tickets from turning them into reopens, and instead just have the agent get emailed so they can reopen manually if need be?Many thanks. 🙏

Hi Zendesk Community,We are currently setting up a support workflow in Zendesk Support to handle a high volume of incoming user inquiries, specifically separating technical bug reports from content and feature requests.Right now, we are relying on basic triggers, but during peak times, tickets tend to overlap in the default view before agents can manually categorize them. - Are there recommended best practices for using conditions and tags in Triggers to automatically route tickets based on keywords in the user's message?- How do other teams manage macro shortcuts efficiently for agents handling repetitive request types?Would love to hear how you structure your triage workflows for faster response times!

Olá,Com a mudança na forma de cobrança das mensagens do WhatsApp Business a partir de 1º de outubro de 2026, acredito que a Zendesk deveria disponibilizar um painel simples e de fácil acesso para acompanhamento do volume de mensagens trocadas.Hoje é difícil ter uma visão clara desse consumo, e isso passa a ser ainda mais importante com o novo modelo de cobrança. Ter informações consolidadas sobre a quantidade de mensagens enviadas e recebidas ajudaria as empresas a entenderem melhor o impacto financeiro dessa mudança.Esse acompanhamento também permitiria avaliar quais canais, filas ou operações geram mais consumo, ajudando na tomada de decisão sobre a viabilidade de manter determinadas estratégias de atendimento ou até mesmo rever processos para otimizar custos.Acredito que essa visibilidade seria muito útil para nós que somos clientes da Zendesk, trazendo mais transparência e previsibilidade em relação aos gastos com o WhatsApp Business.Espero novidades em breve!Merlen KeitiAnalista d

On every email we respond to, the below text is added every time:[Company]I know it cannot be removed per ZD Support, but has anyone else had to change this and if so, what did you put in place of the company and/or brand? 

Over time, Zendesk can accumulate users who no longer have any active conversations. If these records are no longer needed, cleaning them up can help keep your customer database more manageable and support your data retention practices.If you need to find and delete these users in bulk, use this quick process within the GDPR Compliance integration:Step 1: In Combined Lists, select the predefined condition Users without requested tickets. Name the list and add a description if needed, then load it. Step 2: Review the users included in the list and click Process GDPR Request. Step 3: Select Delete Data, review the process, and start the cleanup.The predefined condition saves you from having to build the filtering logic manually, while the list gives you a chance to review the users before anything is deleted.This can be useful for regular Zendesk data cleanup when you have a large number of users who no longer have active support conversations. 

Hi everyone,We use the native X (Twitter) channel and send agent replies to public mention tickets through theTicketing API. Zendesk tweets the reply from the correct monitored handle, but the reply is ALWAYSthreaded under the ticket's first (original) tweet, even when we are answering a later tweet in thesame conversation.WHAT WE SEE (reproducible)1. @customerA tweets T1: "@brand main post"            -> Zendesk creates a ticket2. @customerA replies to their own tweet with T2         -> added to the same ticket3. @customerB replies in the same thread with T3         -> also added to the same ticket4. We add a public comment via the API, intending to answer T2 (or T3)Result on X:- The reply is posted under T1, not under T2 or T3.- The reply text starts with "@customerA" only. @customerB is not mentioned, so they are never  notified, even though we were answering their tweet.The request we send:  PUT /api/v2/tickets/{ticket_id}.json  {"ticket": {"comment": {"body": "Thanks, che

We have SLAs set up in our instance for first reply and next reply, for normal and high priority tickets.  This generally works well, but some tickets (most if not all of which appear to involve side conversations) have no SLA running on them.  I can't figure out what the reason might be, but it's becoming a big issue because our views are set up to be ordered by SLA breach and this is hiding any tickets without SLAs.

What I am trying to do is to get the number of days since the last response. IF Current Day - Last Updated = less than a week (7days) → give me the number in daysIf Current Day - Last Updated = less than 4 weeks (a month) → give me the number in weeksIf Current Day - Last updated = more than 4 weeks → give me the number in months This is what I want to do. TO essentially tell the user that their ticket has been outstanding X days, weeks or month. How can I do this in a macro using the placeholder references / scripting language

Hi all - I wanted to see how other companies are enforcing human agents to document resolution steps in their tickets. Often our IT Technical support team communication and resolve issues in person with a user. Often they close the ticket out without documenting what the fix. I am going to try a checkbox that must be checked for the ticket to close as a reminder to note in resolution steps. We want to enforce the resolution steps as well to training agent copilot for better auto assist suggestions. Has anyone else had this issue and found a system solution for it? 

We respond to many Messaging tickets daily, but our agents must use up their valuable time assigning the ticket to themselves and updating the status, before being able to respond to the Customer. Is there a way to streamline this? It seems like an awful lot of clicks unnecessarily. 

Bit of a sanity check before we commit to a full Zendesk setup.We're a small property services outfit in Glasgow; four of us on the tools, one in the office. Most of our customer contact is straightforward: someone rings up worried about damp or timber issues, we book a damp survey report, then follow up with quotes and scheduling. Right now that's all handled through a shared inbox and a spreadsheet, and honestly it works, but things slip through the cracks when we're busy.We've been trialling Zendesk to see if it can tighten up the process; mainly ticket tracking so every enquiry gets a proper follow-up, and canned responses for the questions we answer ten times a week ("how long does treatment take", "do I need to move furniture", that kind of thing).The concern is whether we're overcomplicating something that a simpler tool could handle. We don't need chatbots or AI routing; just a clean way to make sure nothing gets missed between the initial call and the job being booked.Has anyo