Ask questions, share tips, and discuss best practices about chat and messaging setup, widgets, and features
Recently active
A simple flow builder with different input and output types that are supported by meta is impossible to build currently via Zendesk. I wish this was there.. This is a basic feature available on all omnichannel platforms. If a customer replies to a notification the agent does not see, the context of the message sent.
Hi,On our site, we load the Messenger using the core messaging API and then ask the customer a couple of questions and ask them to fill in their name, email and phone number. At some point in their journey, we already have that data available and would like to pre-fill that for them - similar to the identify call in the Classic webwidget API v1. Is there a way of pre-filling this information for the customer?I have looked at using conversation fields (see this documentation), however that only seems to allow me to pre-fill custom fields - I might be wrong here. If I am wrong, how can I get the id's of the standard fields?Another option would be to log the user in using a JWT token (see this documentation). However, in the flow I want to implement this, the user is not logged in and this requires quite a bit more development work, so definitely not idealAm I missing something obvious? Could you help me out here?Thanks!
I wonder how bot detect customer language through the different channels I have read that with Web widget it should be as the help center it embed ,if it on a website it should detect user's browser language but how it works if I connect the bot on Messenger or whatsApp or other channels .
Hi Im trying to add a phone number on chatbot, so customer can click on the button, it would start a call to us. is that possible?
If I've backgrounded and foregrounded the app, messages stop working in the sdk's MessagingActivity. Either the support member's messages aren't displayed, the user's messages aren't displayed, or both.
Hello, Curious if others have experienced the Massage Grouping updates of the Agent Interface no being reflected through the Beta trial?
I am currently working on setting up the Zendesk conversation bot to function effectively as a standalone virtual assistant. Our goal is to eliminate customer interaction via live chats entirely and have the bot handle all initial queries. Here’s the desired behavior we aim to achieve: The bot introduces itself to the customer. The customer provides their query or prompt. The bot recommends an answer, either based on custom predefined answers or by suggesting relevant Help Centre articles. If the customer requires further assistance, they are prompted to leave a message. The chat session then terminates. The customer is prompted to leave a Customer Satisfaction (CSAT) rating. Issues/Limitations:We have encountered some limitations with the core functionality of the Zendesk conversation bot in achieving this setup. Specifically, for analytics and reporting purposes, it would be ideal for a support ticket to be created when a customer leaves a message. However, it appear
When a new ticket is created through a live chat, is there a way to choose which form is created with. For example our customer creates a live chat related to a Payment, however when a ticket is created the “form” chosen (on the left side panel) is another one but not Payment Has this happened to anyone else? if so, have you been able to solve it ?
I would like to configure a webhook notification for each message being recieved through the chat widget in the account.I was able to set a webhook to ticket being created/updated by this is doesn't seem to be triggered when a new message arrives.Is there a way to support webhook for messaging? if not, is there any workaround that I can use?
Hi!We are looking into merging our multiple brands into one brand. This means that all of our products will be in one help center. This also affects the conversation bot as it will now look for relevant articles and I also assume answers (based on intent) throughout the entire which covers multiple products. Obviously, this is not ideal as we don't want the users to be presented with information of a different product. The solution I see for this is similar to what is available in the Zbot here in Zendesk. There is the possibility to select a Product area and then the bot, it seems, look at the content within that specific selected area (surely the actual technicalities of this might be different, but at least that is the general idea I see happening). It seems that this is achieved via a custom solution or integration. The question is - is there a way how to achieve this with Zendesk features for the conversation bot? Is there a way how to offer a Category dropdown in
I am trying to understand how proactive messaging should work, I am relating to this article 2. Under Frequency,choose how often a customer will receive a proactive message: Once per customer. Message will only show the first time the customer meets the conditions. After clearing the browser’s cache, the customer will be treated as a new user at their next visit and will see the message again. Once per visit. After the message is opened or dismissed, it won't show again until the customer’s next visit. Every time the conditions are met. Recommended only for critical or time-sensitive messages. In my understanding, when I set up the frequency as “Every time the conditions are met”, the proactive message will come up every time the conditions are met, so for example if the trigger is opening a certain web page, every time when this page is opened proactive will fire. The problem I am having is that it will only fire one time, and then it will not fire at all when I refres
Afternoon all. Can anyone confirm that the Classic Widget search bot does in fact search across Federated Searches? It would seem so, but it's not explicitly stated anywhere from what I can see.
Is there a way to give messaging end users a ticket form to fill out with some pre-filled fields based off of the data collected in the messaging conversation? Something like this: https://support.zendesk.com/hc/en-us/articles/4408839114522-Creating-pre-filled-ticket-forms but the values should be unique to every messaging interaction the end user faces. I haven't found a built-in feature like this in messaging, but perhaps there is still a way to do it?
I'm working on a help center page where I implemented 3 icons with links. In one of these links I need that when the customer clicks on it, the chatbot present in the Help Center opens to start customer service. I've already tried using the codes provided by Zendesk itself on the developer documentation portal (https://developer.zendesk.com/documentation/classic-web-widget-sdks/web-widget/quickstart-tutorials/creating-a-launcher- for-the-web-widget/) and I tried to use some other codes I found on the internet (https://stackoverflow.com/questions/31247753/activate-zendesk-widget-using-javascript-trigger), but none of the codes used It worked. Below I leave the code where I want the user to click to open the chat with the last attempt I made. home-page <div class="column-container-principal container-link"> <a id="start-chatbot" href="javascript:$zopim.livechat.window.show();"> <img title="chat" src="https://theme.zdass
Hi there, hope all is well. On the Zendesk chat interface we can see the average wait time of chats in queue for each department we have set up. I was wondering if there was a way to see real time information of what the average wait time for a department was for the current hour period. I have set up an explore report but this doesn't update in real time.
Using the messaging web widget, I am using JWT to authenticate my users. When I generate the JWT payload, I include the user's name, email, set “email_verified” to “true” and have “scope” = “user”. When a user is created in my system I create a user in Zendesk via the API. However, when the user opens the messaging widget, none of their information is prefilled and unless they type their exact name and email, they are not recognized as the user I've created for them in Zendesk.
I'm not sure why Zendesk deleted this tip I wrote previously, but I keep backups of everything, and figured I'd put it back up, since this is still something people are asking about chat, and seems helpful. When working with chat data, it became clear that when you have automatic chat assignment on, there is a need to better understand why a chat was marked as Missed. I did some brainstorming with team mates, and we broke down missed chats into a few buckets:1. Nobody was available - Someone chatted in, waited a reasonable amount of time, but no agents were assigned to the chat as they were busy, and then the chatter left, without being served.2. The chatter left quickly - Someone chatted in, then left very quickly, before an agent could join. 3. Tried Agents - Someone chatted in, waited, agents were assigned to the chat, but none of the agents picked up in the 30 second window. As you read these, you might start to see why these buckets are needed. If you had a high volume
The customer contacts us on WhatsApp with one query - we deal with it and close ticket. Same customers contact via WhatsApp again with different query - but all go under one original ticket. Then reporting is not accurate? How best to manage these as different queries so we can tag and classify each query.
Hi All, How can I report on the # of messaging “chats” that turn into email? For example, the support request starts in messaging but turns into email because it is outside business hours, no agents were available, or the customer timed out due to inactivity. We ultimately want to understand who had a “true” messaging experience vs who had an email experience. Thank you!
I'm trying to apply “noprint” css class so widget doesn't get printed out… anyone have any idea how we can achieve this? TIA!
Hi,Is Sunshine Conversation needed when using Conversation Extensions in messaging?
Hi everyone, I would need your help to understand why it is now possible, through the agent workplace in Zendesk, to respond to a WhatsApp tickets even after 24 hours. Let me explain a little better: until last week, when a WhatsApp ticket exceeded 24 hours, Zendesk did not allow the option to write a public comment with WhatsApp within the ticket. Since a couple of days now, we've found out that it has become possible to insert and send message that anyway fails to be delivered because Meta prevents it due to its 24 hours limit rules. We are afraid of possibile issues, not to mention that it is now more difficult to identify the tickets that agents can actively work on. Hope I have been clear enough. Thank you in advance for your support. Stefano
I have a question to our Zendesk Community! So, my team is still considering the transition from Chat to Messaging.The reason for the delay is that Messaging authentication is only possible with JSON web token, which requires us to create a costly new server just to receive authentication requests.This is unlike classic Chat which is already authenticated by the user's Zendesk Help Center login (via SAML SSO in ours). While we can still use the Messaging widget without authentication, one limitation of this is that the widget cannot provide article search results if the Help Center itself requires sign-in (even if the articles themselves are set to “Everyone” viewing permission). My question is, if you are using Messaging within a Help Center that requires sign-in, how did you manage its limitations?Did you purchase a server just for the JWT?Did you forego article recommendations altogether?Did you force your Help Center to become publicly available?Or, did you find a wo
I want to know when will be released a new version of the iOS SDK with Privacy Manifest as required for Apple for 3rd party sdk. Do you have an approx release date?
Hello, Is there any option to reply to Facebook comments without selecting the option to create every single comment as a ticket? Regards,
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.