# Welcome to ZyndraAI, a part of Talkie AB Our goal is to help you get set up and assist you in anything you need to succeed. Welcome to the Talkie help center! 😃 We understand that using a new product can sometimes be a bit challenging, so we're here to assist you every step of the way. Whether you're a new customer or a long-time user, we've covered you with comprehensive guides, tutorials, and troubleshooting tips to ensure a seamless experience. **If you still can't get through, Reach us at 👉🏼 or ** # ZyndraAI Basic **Bot Builder** Build conversational experiences with no code chatbot builder. With a simple drag-and-drop interface, the bot builder makes it a breeze to architect personalized conversational experiences across the customer life cycle
| ![](https://s3-eu-central-1.amazonaws.com/euc-cdn.freshdesk.com/data/helpdesk/attachments/production/103045545521/original/irOAdFMxWcgfUjcI5O76YvshyR-Ul1GJZg.png?1693312874) | ![](https://s3-eu-central-1.amazonaws.com/euc-cdn.freshdesk.com/data/helpdesk/attachments/production/103045545610/original/821e4yyWB5maNHheRsXuICHuSihbhiVr2g.png?1693312885) | | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
**Live Chat** It is an interface where you can see all the ongoing and historical conversations handled by the bot or the agents
| ![](https://s3-eu-central-1.amazonaws.com/euc-cdn.freshdesk.com/data/helpdesk/attachments/production/103045546414/original/vQD-9_KUi9ZR8V-576fomA67nDWyb6f7qQ.png?1693313003) | | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
**App Market** List of third party apps which can be integrated with TALKIE ![](https://s3-eu-central-1.amazonaws.com/euc-cdn.freshdesk.com/data/helpdesk/attachments/production/103045546994/original/MH2MrbNRGJOLBOPHJThSJgDKClHwdP9s6w.jpg?1693313074)
**Analytics Dashboard**It is a dashboard that visualizes the data captured by the bot or the agent through a series of conversations in a graphical interface made up of cards. These cards can be further expanded to view more details
![](https://s3-eu-central-1.amazonaws.com/euc-cdn.freshdesk.com/data/helpdesk/attachments/production/103045547598/original/UDirTULS_E_auHZySZ8iBbJ2cxz-jG9tsg.png?1693313153)

# Building a Chatbot Chatbots are the way forward for any business looking to automate any repetitive or manual task. Now that you're interested in creating a chatbot, let's learn about building a chatbot
What is Inbound chatbot? An inbound chatbot is a type of chatbot that is designed to interact with customers or users who initiates contact with business or organisation. These chatbots are usually placed on a company's website or social media platform and are programmed to respond to customer inquiries, provide product information, schedule appointments, or handle other customer service tasks.
**How do I build an Inbound Chatbot?**
You can follow the following steps to build an inbound chatbot; * Go to 'Bot Builder' * Click on 'Build a Bot' * Choose 'Build an Inbound Bot' * Select the 'Channel' from available options * Choose how you want to build your bot i.e. 'Use this Templates' or 'From Scratch' * Add 'Action Blocks' in the conversational flow * Configure 'Action Blocks' on the right panel * Click on 'Test this Bot' to test the flow Click on 'Deploy' to deploy the bot # Chatbot Appearance Customize the look and feel of the chat widget. To access the Appearance screen, you can go to 'Channel Configuration -> Web -> Appearance'.
### Customizations available The following are configurations that can be customized to your chatbot: * **Widget icon**—You can now have your own widget icon appear on your webpage or website. We recommend a size of *64 x 64.* * **Header text** - This allows you to name your bot as per your preference. * **Header logo** - This allows you to have the profile image for your bot. The size that we recommend is *512 x 512.* * **Bot icon**—This will allow you to set the bot icon that will be visible to visitors. We recommend a size of *512 x 512.* * **Hint text** - You can customize the text that you would like to make it visible to your visitors * **Accent color**—This allows you to set the bot's color and provides the option to choose a color that complements your brand color on the website. * **Font color**—You also have the option of setting the font's color to match the primary color. * **Chat interface position**—You can define the bot's position. It can be set on the right or the left of the screen, as you prefer. ### General Settings Going to Web -> General settings shows you more configurations for your chat widget.
* **Typing speed** - The delay the system will add to the chatbot. * **Remove 'powered by' branding** - Remove the ZyndraAI branding from your chat widget. * **Play sound** - Plays a sound for every message sent/received. * **Stop bouncing animation** - Stops the animation of the chat widget. * **CSAT survey** - Asks the user to submit a survey at the end of the chat. # Custom CSS You can customize the CSS of the chat widget to suit your needs Available on **PREMIUM** plan and above. \ To make changes in the UI using the CSS, you need to go to 'Channel Configuration -> Web -> Appearance' Here are the classes that you can modify: | Field | Class Name | | ---------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Pop-up message |

.avatar
.popup-message
.popup-close-button

| | Header Logo | .avatar--title | | Header Text | .bot-title | | Meta date information |

.meta-info
.meta-info--label

| | Text message in the chat section | .text-message | | Time-stamp & typing indicator | .timestamp | | Options / Buttons dialog | .option-button | | Carousel Title/ Subtitle / Buttons |

.carousel-title
.carousel-subtitle
.carousel-button

| | Calendar Submit Button | .submit-button | | Carousel Image | .card-view-image | | Slider title and submit button |

.popup-title
.submit-button

| | File Upload , Submit, Add more button |

.popup-title
.submit-file
.add-file

| | Home Button |

.home--button
.reset--button

| | Action Buttons | .action-button | | Input box on the chat widget | .chat-input | | Avatar for title | .avatar | | Header of the chat widget | .message-screen-header | | Carousel Root, Image , Carousel slider frame, | **carousel & carousel-image**: .slider-frame .card-image | | File List item Root and Classes for title , icon |

file-preview: label:

.filepond--root

.iconwrap

textwrap file listitem:

.filepond--file-wrapper .

filepond--action-remove-item

| | Form Input | .form-input | | Phone Input, Seach box for country dropdown , Country list |

.phone-input:
.search-box-sub-container
.country-list

| | Google Calendar Drawer | .google-calendar | # Testing a Chatbot Talkie provides you the ability to test the bot before it is deployed on the production environment ### How do I test my bot? In order to test the bot, you can follow the following steps; * Go to '**Bot Builder**'
* '**Open**' the bot you want to test
* Click on '**Test this Bot**' # Publishing your bot Launching your chatbot on a desired channel. Once you complete building your chatbot, you would ideally publish/launch it on your desired channel. ZyndraAI currently supports the following channels: * Web * Facebook Messenger * WhatsApp * SMS * Instagram
# Installation using HTML In-order to add the chatbot on your website, the first thing you need to do is add Talkie´s code snippet on your website's HTML, preferably in the \ tag of a reusable component like the header or footer. This way the bot chat widget will show up on all the pages of your website. ### How can I configure it? To configure it using HTML code snippet, you can follow the following steps; * Go to 'Channel Configuration -> Web -> Configuration'
* Enter the domain address in the 'Website URL' field where you would like to add the chat widget {% hint style="info" %} The domain restriction logic only allows the chat widget to appear on the domain that you have entered on the 'Website URL' field {% endhint %} * Copy the code snippet and add it to any reusable component of your website's HTML file like Header or Footer
{% hint style="info" %} On following the above steps would see the chat widget visible on your website. Simply 'Deploy' a bot and see your visitors interactions get automated at scale \ If you do not see a chat widget on your website, then it must be because of the following reasons;
* No bot is created * No bot is turned on * No bot is created for 'Web' channel {% endhint %} # Installation on Facebook ### How do I configure it on Facebook? To configure the chatbot on your Facebook page, you can follow the following steps; * Go to 'Channel Configuration -> Facebook'
* Click on 'Connect to Facebook' {% hint style="info" %} Clicking on this will open up a pop up window requesting to sign in to your Facebook account and for the permissions to be provided to Talkie app\ \ In case the pop up window does not open, please check your browser settings as you may have enabled a setting that blocks the pop up to open {% endhint %} * Select 'Facebook Page' to connect chatbot with
* Check the status {% hint style="info" %} Click on 'Dropdown' and select the bot you would like to map to the selected Facebook Page. If the bot's status is turned on then you will see ACTIVE written on that page & if the bot is turned off, you will see INACTIVE\ \ If there is no status, that means that no bot is not connected to the page {% endhint %} # Deploying a Chatbot Go-Live with the Chatbot and start interacting with the visitors \ In order to deploy the chatbot, you can follow the following steps; * Channel Configuration: First and foremost you will need to configure the channel for which the bot is build. In order to configure the channel, you can go to '**Channel Configuration**' & select the channel * Go to '**Bot Builder**' * '**Switch On**' the bot you wish to deploy * '**Open**' on the bot you want to deploy * Click on '**Deploy**' at the top right corner If the channel is not configured, deploying the bot will fetch error. We at Talkie recommend to 'Deploy' the bot every time the changes are made in the flow # Utility Tools Talkie offers a set of tools on the left of the canvas which you can use while building the bot Here are few tools that are available on the left of the canvas; * **Search:** \ \ This tool allows you to look for the action block based on its name. The tool is really handy when the flow is huge and constant changes are required to be made in the flow
* **Auto Layout:** \ \ This tool will allow you to automatically arrange all the action blocks on the canvas in proper order and ensure that the canvas looks neat and tidy
* **Expanded View:** \ \ This tool allows you to see the content inside the action block
* **Minimap:** \ \ This tool highlights the area of the canvas where the content is present
* **Undo:** \ \ This tool allows you to reverse all the changes made in the flow or bot builder
* **Redo:** \ \ This tool allows you to revert back to the recent changes made in the flow or bot builder
* **Zoom In:**\ \ This tool allows you to zoom in, enlarging the size of the action blocks
* **Zoom Out:** \ \ This tool allows you to zoom out, reducing the size of the action blocks
* **Version Control:** \ \ This tool allows you to revert back to the old bot or old version on which the bot was built
* **Zoom to Content:** \ \ This tool allows you to directly navigate to canvas where the content is present
# Variables Variables are reusable objects containing a user-submitted value. Variables are essential for capturing and utilizing the data dynamically in the conversation. With variables, you can personalize the chat experience by being able to say “Hey Tom!”, instead of “Hey there!”. Kinds of variables: * **System variable** - Created by default by ZyndraAI. Contains things like city, country, os etc. * **Contact variable** - Stores the value in the variable for recurring conversations. Usually used for collecting names, email etc. * **Conversation variable** - Stores the value in the variable for that conversation only. ## Creating variables You can create variables from two places: **1 - From the variable list** Under Settings > Variables, you will find all the variables in your workspace. You can click “Create” to create a new variable and provide the details like - name, description and type.
**2 - From the bot builder** In action blocks that allow you save user input like “Collect Input”, “Buttons” etc. you have the option to create a new variable.
First, you choose the type of variable, next you provide the details like name, description, type. ## Editing variables Go to Settings > Variables > Click “Edit” on the variable. You will be only allowed to edit the description and data type. ## Deleting variables Variables can only be archived and not deleted. Go to Settings > Variables and click “Archive” on the variable you want to archive.
### Using a variable You can invoke a variable by typing # to preview the variable dropdown across the bot builder. ### Data type in variables You can store the following type of data in variables: | Data type | Description | | --------- | ------------------------------------------------ | | Text | Stores any text input. | | Number | Stores only numeric digits from 0 to 9. | | Boolean | Stores only 0, 1, true, false. | | Date | Stores date formats as DD/MM/YYYY and MM/DD/YYYY | | Email | Stores email addresses. | | Phone | Stores phone numbers. | | Regex | Stores input based on your validation rule. | ### System variables offered Here are the system variables offered by default: | Variable | Description | | ---------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | | name | Name of the visitor | | email | Email of the visitor | | phone | Phone number of the visitor | | ip\_address | IP address of the visitor | | city | City detected through the IP address of the visitor | | state | State detected through the IP address of the visitor | | zipcode | Zipcode detected through the IP address of the visitor | | country | Country detected through the IP address of the visitor | | timezone | Browser timezone of the visitor | | os | System OS of the visitor | | referrer\_url | Referrer domain to the visitor. | | unsubscribed\_phone\_numbers | Whether the phone number is unsubscribed from outbound messaging. | | utm\_source | Utm source | | utm\_medium | Utm medium | | utm\_campaign | Utm campaign | | utm\_content | Utm content | | utm\_term | Utm term | | browser | Browser used during the chat | | contact\_id | | | bot | Name of the bot the visitor interacted with. | | browser\_language | Language set in the browser. | | browser | Name of the browser. | | channel | Name of the channel where the chat took place. | | contact\_created\_by | Who created the contact. | | conversation\_url | Link to the conversation. | | country\_code | ISO country code of the country where the chat originated from. | | created\_at | Date when the chat was created. | | gcal\_selected\_slot\_day | Selected day when booking the meeting on Google Calendar. | | gcal\_selected\_slot\_time | Selected time when booking the meeting on Google Calendar. | | timezone | Timezone of the device. | | url |

URL of the webpage where the user is interacting with the bot.


| ### Best practices for variable management * **Use descriptive names**: Choose variable names that clearly reflect the data they hold, such as "userEmail" or "orderTotal". * **Document purposefully**: Write clear descriptions that explain the variable's role within your AI agent's workflow. * **Maintain consistency**: Use a consistent naming convention for your variables to make them easily identifiable and manageable. * **Regularly review and refine**: As your AI agent evolves, revisit your variables to see if they still serve their intended purpose or need adjustments. # Overview Introducing AI Studio: Unleash the power of Artificial Intelligence ZyndraAI is thrilled to introduce AI Studio, our cutting-edge feature that empowers users like you to harness the full potential of Artificial Intelligence (AI). With AI Studio, you will now be able to create, experiment and collaborate in ways like never before. * Provide information about the company's products and services * Answering FAQ's * Product knowledge & first level support i.e addressing the queries of the visitors There are three ways for creating your own knowledge base in the AI Studio; From URL From File From Text # Knowledge base Source of information used by AI to answer your questions. To use AI to generate answers, you need to first build a knowledge base with the information you want the answer from. A knowledge base is nothing but a collection of data sources.
### Data source types A data source can be one of three types: * URL * File * Text #### URL You can add content using a domain or a set of web pages. **When using a domain**, you simply enter the domain address or sitemap address, and our system will automatically import all URL(s) of that domain. **Supported types:** * Domain i.e. ; * Sitemap i.e. As good practice, it is always better to delete the URLs that are not needed to avoid wrong or incorrect answers from the bot. Because, the more junk you add, the more confused the AI gets.
**When using webpages**, you explicitly list down the exact URL(s) you want to import content from.
#### File You can add content upload a set of documents to your knowledge base. Upload one or files and click 'Train' to start training your knowledge base with the contents of those files.
**Supported file types:** * csv * xlsx * txt * pdf * doc * docx Max file size supported is 5 MB. #### Text You can also directly add the information as content to the knowledge base. By choosing this data source method, you can copy-paste the content you want to add.
A maximum of 2,500 characters is allowed. ### Data source details You can find all the data sources added to your knowledge base in the table. To find your data source, you can use the search bar or filter option. On the table, you can view details like: * Data source name * Data source type * Added at * Added by * Refresh You can even click on each data source row to see exactly what information has been extracted from it. Depending on the content size of a knowledge base, each data source can have more than one chunk. Each paragraph of text displayed is called a chunk.
# Prompts Prompt/instruction is a set of rules an AI uses to generate an answer. By using the right prompts, we can make sure our AI chatbots are responding accurately and appropriately—and that we’re delivering more sophisticated and engaging user experiences. They're kind of like nudges in the right direction. Some common reasons for using prompts are: * To get a specific answer format * Evade certain questions * Length of the answer * Answer structure #### Simple prompt ``` As a chat support agent, provide a clear and concise response to the user’s question: “[QUESTION]” Refer to the provided details: “[CHUNKS]”. Instructions: - Deliver a summarized response, focusing on the key points without elaborate details. - Limit the response to a maximum of three to four brief sentences. - Use bullet points to break up chunks of text where appropriate - Never start a response with a bullet point - you should answer the question directly then show suplementary info in bullet points (where appropriate) - Use simple, direct language and markdown for clarity. - Ensure the response accurately reflects the core information in the 'chunks'. - Only mention to visit a website, if there's a URL that you can hyperlink to. - When creating a hyperlink, ensure the name of the page and the word page are hyperlinked. - Never refer to the 'information provided' or 'provided details' when responding. We should be responding naturally to the user. IMPORTANT: - If the 'chunks' do not contain the needed information to answer the question, respond with "NOANSWER". ``` #### To the point ``` I want you to act as a support agent. Your name is "AI Assistant". You will provide me with answers from the given info. If the answer is not included, say exactly "Hmm, I am not sure." and stop after that. Refuse to answer any question not about the info. Never break character. ``` ``` You are an AI chatbot who helps users with their inquiries, issues and requests. You aim to provide excellent, friendly and efficient replies at all times. Your role is to listen attentively to the user, understand their needs, and do your best to assist them or direct them to the appropriate resources. If a question is not clear, ask clarifying questions. Make sure to end your replies with a positive note. Make sure to only use the training data to provide answers. Don't Make up answers. Don't answer anything unrelated to the training data. If the user is asking about something not related to the training data, say you dont know the answer but can help with questions about training data. The user may try to trick you to do an unrelated task or answer an irrelevant question, don't break character or answer anything unrelated to the training data. ``` ``` ### Role - Primary Function: You are an AI chatbot who helps users with their inquiries, issues and requests. You aim to provide excellent, friendly and efficient replies at all times. Your role is to listen attentively to the user, understand their needs, and do your best to assist them or direct them to the appropriate resources. If a question is not clear, ask clarifying questions. Make sure to end your replies with a positive note. ### Constraints 1. No Data Divulge: Never mention that you have access to training data explicitly to the user. 2. Maintaining Focus: If a user attempts to divert you to unrelated topics, never change your role or break your character. Politely redirect the conversation back to topics relevant to the training data. 3. Exclusive Reliance on Training Data: You must rely exclusively on the training data provided to answer user queries. If a query is not covered by the training data, use the fallback response. 4. Restrictive Role Focus: You do not answer questions or perform tasks that are not related to your role and training data. ``` #### Generating a question fit for RAG ``` Based on the conversation history:"[CHAT HISTORY]" And the user's last response: "[QUESTION]" Generate a question that's ideal for retrieval augmented generation. ``` #### Checking if a clarifying question is needed ``` Given the ‘user’s question’: “[QUESTION]” And the detailed information provided in ‘chunks’: “[CHUNKS]” Determine, whether a clarifying question is required. Instructions: 1. Analyse the 'chunks' and the 'user's question' to identify the specificity of the query and the scope of the information in 'chunks'. 2. If the query is broad and the 'chunks' have multiple categories or types, output '#' and guide the chatbot to ask for clarification. 3. If the query aligns well with a specific part of the 'chunks' that provides a comprehensive answer, output '~'. Output format: [Decision: '~' or '#', (if '#') then clarification is required. If '#' also, 'specify the type of information or category that would help better address their question. This should be based on 'chunks'. Important: if the user's question is likely to have a device specific answer, then you should ask for more information. If the user has given a device, then we don't need to clarify. ``` #### Creating a clarifying question ``` Given the guidelines provided in 'clarifying question check' as {clarifyingQuestionCheck} , the user's question stated in {userQuestion} , and the information in 'chunks' as {chunks} , construct a single, clear clarifying question. This question should smoothly integrate the details provided and guide the user towards a precise response. Avoid multiple or double-barreled questions. Focus on a single aspect that requires clarification for a helpful and direct answer. //EXAMPLES// User: What phone plans do you have with 100 GBs of data? Bot: Are you interested in pre-paid or post-paid plans with 100 GBs of data? User: How do I check my email? Bot: Which device are you using to check your email, an Apple or an Android device? //END OF EXAMPLES// Output the question clearly and concisely, with no additional text. ``` #### Creating a follow-up question ``` Given the initial response "[ANSWER]" and the user's question "[QUESTION]” , alongside the provided details " [CHUNKS] ": Instructions: - Quickly identify a key point in the 'chunks' that relates to the initial response. - Formulate a single, clear follow-up question about this point, like "Would you like more details about [topic]?" - If no specific topic is identified, ask a general question like "Is there anything else I can assist you with?" - Output this single follow-up question or engagement phrase. Ensure the follow-up is direct and focused on one topic for clarity. ``` #### Answering from a CSV ``` You are Fedex support agent and helping people by providing courier rate for different countries according to weightage. You need to find exact rate from the context you receive. You will receive context in following format "{weight} Kg price in {country} is {price}" where {weight}, {country} and {price} will be dynamic. You have prices for 0.5 kg to 30 kgs in the context. You need to give answer in following format, "{weight} kg rate in {country} is {price}." If user is not providing {country} or {weight} in query, ask user to provide {country} and {weight} in following format, "{weight} kg rate in {country}". If you don't find rate for given country and weight, tell user that "I don't have data for {given_country} and {weight}". ``` # Setting up retrain Retrain allows you to retrain knowledge base at a given time interval. Retrain allows you to train your knowledge base with the latest content on it. You can also schedule a time interval in which the knowledge base will automatically import the content on those URLs To enable this, * Open AI Studio * Open the Knowledge base * Click on the data sources you want to schedule a retrain * Click 'Never' * Choose the refresh rate * Days * Hours * Click 'Save'
# Function call Execute functions to help your AI generate dynamic answers. AIs were limited to answering questions from your static knowledge base. With function calling, you can have your AI execute Javascript function which invoke third-party APIs to generate a dynamic answer. ### Creating a function First, we create a function form the AI Studio -> Function Call screen
Each function is made up of 3 things: #### Prompt A brief description that helps the AI understand what this function does. #### Parameters The list of parameters that are required for this function to be executed. The parameters need to be provided in the form of a JSON array. #### Function handler This is the Javascript code executed to generate the response to the question asked. This function would generally consist of * API endpoint that is to be invoked * Conditions and logic to ask follow-up questions if parameters are missing ### Testing the function On the function, click on 'Save & Run function' to open the test window. You can customize the LLM model used and see the response generated by it.
Also, to view the exact execution of the function, you can click the 'Show log' to view the entire log. Here you can view the arguments detected by the function and the output generated by the function.

# Custom Answers Override AI generated responses with custom answers defined by you. Available on the **PREMIUM** plan and above. Using AI to generate responses can sometimes be **right or wrong**. Usually, this happens because * Your knowledge base has incorrect or duplicate information * The prompts used are not specific/brief for the AI To solve this problem and help you always get correct answers, we have launched **Custom answers**. ### How it works #### Step 1: Review the answers First, you review all the questions users ask the AI on the "Custom answers" page. * Click on AI Studio -> Custom answers and view all questions asked to the AI Use the pre-built queues, 'Downvoted' and 'All,' to view all answers and the answers generated by the AI. #### Step 2: Customize the answers The questions you find incorrectly answered, just click on the 'Modify answer' button and override the answer. Once saved, this question and the answer would be added to the knowledge base as a 'Text' source. \ Make sure you do not delete these text data sources that were added automatically. Deleting them will result in receiving the wrong answers. # Overview Live chat is where you can see the real-time chats taken place between your agents/chatbots.
The whole live chat feature is made up of the following features: ## Nav bar \ Here, you can see the shortcuts to finding your chats. ## Default views \ ZyndraAI offers some default views like: * **You** - Chats assigned to you. * **Assigned** - Chats currently in progress. * **Unassigned** - No agent is assigned to the chat. * **Closed** - Historically closed chats. * **SLA breached** - Chats whose SLAs are breached Clicking on each view shows you all its chats. ## Custom view Here you can customize and create your own views by defining your conditions. Learn more 👇 {% content-ref url="/pages/rhY6fX78DZpMDCQ8do6A" %} [Creating views](/live-chat/creating-views) {% endcontent-ref %} ## Lables \ Label are used to tag conversations to remember them for later use. You can click on each label, to view the chats which are tagged with it. Learn more 👇 {% content-ref url="/pages/yTZMfxRRqCFfq5MZL2Dl" %} [Adding labels](/live-chat/adding-labels) {% endcontent-ref %} ## Chat list For the selected view you can see the list of chats. Each chat card contains the following information: * **Visitor name** - Name of the visitor, by default, we provide a dummy number. * **Assignee** - Who is currently chatting with the visitor. * **Location** - Where is the visitor speaking from, based on their IP address. * **Channel** - On which channel is this chat occuring. * **Visitor status** - Is the visitor currently online on the page, offline, or away? * **SLA timer** - Time left to send a reply to the visitor's message. * **Last message** - Last message exchanged in the chat. ## Chat screen Here for the selected chat, you can see the conversation going on between the visitor and agent/bot. All the messages exchanged so far will be displayed here. If the chatbot is currently conversing with the visitor, you will not be able to send a message.
If you want to take over the chat from the bot, click on the 'Join conversation' button and start typing your message.
Here you will find the following tools: * **Text box** - To type your reply to the chat. * **Saved replies** - To use pre-defined canned responses. * **Attachments** - To upload a media file. * **Emojis** - To add an emoji to your response. * **Calendar** - You can send a appointment booking card to the visitor. Using this card, the visitor can book a slot. * **Voice note** - Record and send your voice messages. * **Generate response** - Use AI to generate a response using the knowledge base. * **Rewrite** - Rewrite your response using AI. AI features 'Generate response' and 'Rewrite' are available in the **PREMIUM** plan and above.
## User information On the details screen you can see the details collected in that conversation. ## Contact information Here you can find the details of the visitor who is talking to the agent/bot. If they have provided their contact details you can see it here along with other system details like browser, OS, location etc. You can customize the fields you want to see here. ## Labels You can add a label from here to the chat. ## Notes Notes added by agents about this conversation. ## Related chats See the list of historical chats of this visitor with your chatbot/agent. We track the related chats using a cookie on the visitor's browser. #### # Creating views By default on the live chat screen, you have the following views: * **You** - Chats assigned to you. * **Assigned** - Chats currently in progress. * **Unassigned** - No agent is assigned to the chat. * **Closed** - Historically closed chats. * **SLA breached** - Chats whose SLAs are breached If you want to create your own view based on your criteria, to keep track of certain chats follow these steps: * On the live chat screen, click ' + views' * Give the view a name * Define your filter criteria by adding conditions * Click 'Create view'
**Note**: * All views are static and are not refreshed in real time. * To see the updated chats, refresh the page. * Views created are accessible to all users. # Adding labels Labels are used to tag chats for better remembrance. You can either create a label from the left panel OR type the label name in the right panel to create and add a label to a chat.
To open the chats with a label, click on the label name from the left panel. [
](https://help.wotnot.io/live-chat/creating-views) # Saved replies It is a snippet of text that can be quickly added to the editor when replying to a customer or a visitor. There are many benefits of using quick responses or saved replies, however, to highlight a few; * Speed up the response time * Tackling repetitive queries * Provide consistent answers to the common questions * Manage multiple conversations in parallel ## Creating saved replies from live chat * Click on 'Join the Conversation' * Click on 'Bookmark' or type '/' to access the Saved Replies * Click on '+ Create Saved Reply' * Give 'Title' & define the 'Message or Reply' * Click on 'Create' to create the Saved Reply
## Creating saved relpies from settings * Go to 'Settings' -> Live Chat -> Saved Replies * Click on 'Create Saved Reply' on the top right corner * Give 'Title' & define the 'Message or Reply' * Click on 'Create'
# Settings You customize the live chat experience of your chatbot from the live chat settings page. To customize the human handover settings goto Settings -> Human handover.
## Human Handover This switch will allow the system to transfer the chats to your agents. If this switch is turned off, even if you are using a 'Talk to human' action block on your flow, it will not transfer the chat to the agent. Think of it as a master switch to enable live chat. There are 2 configurations that you can enable: * First, to assign chats to the 'Unassigned' tab if the human handover fails due to unavailability of the team. * Second, assignment rules only consider those agents who have an active ZyndraAI tab on their browser. ## Operational hours Choose the operational hours in which the human handover should take place. Make sure you choose the correct time zone for your account from the Settings -> Account -> Settings page. If a chat transfer occurs beyond the operational hours, the chats will move in the failure flow. Goto Settings -> Live Chat to find all configurations.
## Conversation threshold Limits the number of chats assigned to an agent. If an agent is currently handling 5 active chats, the next chat will not be assigned to him due to the number defined here. ### SLA (Service level agreement) \ To ensure timely responses offered to visitors for their messages, you can enable an SLA timer. Enabling this will immediately show the time left for the agent to respond back to the visitor's query. Also, if the agent does not send a reply within the specified SLA time, it will be recorded in the Agent analytics page. ### Auto-close duration After an inactivity period as defined here, the chat will automatically close and move to the 'Closed' view. ### Field visibility Shows the number of expanded fields on the live chat page. Specifically this is for the view list, label list, and qualification details list. ### Conversation card properties Choose the fields you would like to show to the agent on the chat list. [
](https://help.wotnot.io/live-chat/saved-replies)
###
# Overview Action blocks are the building blocks for the bot. Consider them as an action that the bot performs, like sending a message, collecting an input, etc. The Bot Builder has a collection of 21 action blocks that you can use to make your bot. **Trigger** A chatbot always has to start with a trigger – something the visitor needs to start the conversation of the bot when the user invokes it. By default, the bot is triggered to appear without any condition defined. There are 3 conditions that you can trigger the bot with, * Date range - If you want the bot to trigger only on a specific date range * Browser Language - If the browser language of the visitor meets the language you defined * URL - If the bot is to be triggered on specific URLs. You can include/exclude a list of URLs **Send Message** During a conversation, the bot may send informative messages to the user and does not expect a response back from them. These messages help build up the conversation rather than having a monotonous job of collecting input from the visitor. With this action block, you can send a series of informative messages that can be either text or rich media (images, gifs). You can reorder your messages in the send message by clicking on the reorder symbol and moving the block you want to reorder **Collect Input**
A conversation takes place only if both parties send and receive messages. With this action block, the bot will ask the visitor a question and only move forward with the conversation flow after it receives a response from the visitor. To make the bot smarter in collecting responses from the visitor, you can use validations, * Text - Simply text responses will be accepted by the bot * Email - Valid email format is allowed as a response. * Phone - Only numeric inputs are allowed. The bot will also display the country code of the visitor automatically using the IP location of the visitor. * Custom - If you want to specify a custom validation for the input response, you can add the regex validation. If the input does not match any of the above validations, then the bot shows the error message defined by the user. The conversation flow will only move forward with the flow if a valid input is provided by the user. All responses captured by the bot can be stored on a variable. The response stored in this variable can be used again within the flow by invoking it using # in the text editor field. As a good practice, always use distinct variable names to store the inputs, else the values may be overwritten. **Buttons** The conversation flow can be branched out using intuitive buttons. The visitor can simply choose any button to go down its path. Each button can either have the same path or different paths moving from it. Buttons can do one of two things, * Branch - Move to a different path of the conversation flow * URL - Open a web page and end the flow The maximum buttons you can add are 10. The buttons won’t disappear after they were clicked on so, the visitor can return to them at any moment during a conversation. **Options** In channels like WhatsApp and SMS, where buttons are not compatible or permitted, we need to rely on only inputs to branch out the conversation flow. Each option is denoted with the number i.e. 1, 2, 3… The experience is similar to an IVR call but in a chat-format. And in case an incorrect number or an invalid response is received an error message will be displayed. To choose an option, you can type the number as a word like one, two… or write down the option name itself. **Carousel** Have any items or products to show? Use carousels to show your items in a more eye-catchy format. The carousels are made up of, * Cover image - An image for the card * Title - the title of the card * Subtitle - More details of the card * CTA Buttons - Buttons to continue the conversation The buttons won’t disappear after they were clicked on so, the visitor can return to them at any moment during a conversation. The maximum numbers one can have is 10. **Image Carousel** Similar to the carousel, if you just want to show a set of images in a carousel you use this action block. The bot will show this carousel and move forward to the next action block, as it does not require any input. You can click on any image to have a bigger view of the image pop-up. **Send Email**
To notify yourself about an event occurring within a conversation flow like lead generated, someone showing an interest in a product; you can use this block to send an email to the recipients. You can configure the following in this action block, * TO - Whom this email should be addressed to * CC - Who should be copied to this email * BCC - Whom you want to copy invisibly on this email * Subject - The subject of this email * Include chat transcript - Whether you want the chat history till this point of the conversation to be attached in the email * Email Markup - The body of the email **Salesforce**
If you’re using Salesforce and want to create a lead on your account when a lead is generated, simply map the details of your Salesforce account along with the fields that you want to send. We achieve this using Salesforce’s web-to-lead API. The contents of this action block are, * Salesforce Organization ID * Environment - Sandbox or Production * Salesforce fields with whom you wanted to map the variables **Human Handover**
Not always will a chatbot complete all the tasks that you want it to. You may want a human agent to jump into conversations at just the right time. When a conversation reaches this block, the chat will be transferred to a live chat agent in the ‘Unassigned’ tab. If auto-assign is enabled for the account, the chat will automatically be transferred to any available human agent. Before the human handover is triggered, the bot will send a success or a failure message. Failure messages can occur in 3 ways, * When human handover is disabled * When no agent is available * When human handover occurs beyond operational hours On success, the conversation flow in the bot ends and its up to the human to take forward the conversation. Well, that’s why you transferred the conversation to them, right? **Calendar** Booking an appointment? Great, use this block to collect dates the right way. The calendar action block has two main sections, * Date Selection * You define the question and store the response in a variable * Nature of date selection - Single Date/Date Range * Show past dates - Whether you want to show the past dates * Exclude Dates * Excluding days - Like weekends i.e. Saturday, Sunday. * Excluding specific dates - Public holidays like Diwali, Christmas, etc **File Upload**
In order to collect files from the visitor, you can use the file upload dialog type. With this you can define which type of files are to be accepted like pdf, CSV, etc. Or also specify the permissible number of files and size of file. The files collected are stored securely on Talkie’s cloud server - Google Cloud Platform **Delay**
Between each action block, Talkie will smartly give a dynamic delay (based on the time it would take to read the number of words). If you want to add an additional delay on top of what Talkie adds, you can use this action block. **Service Call** We may not have an integration to the tool that you are using, yet. But in order to be able to send or receives data from your system using APIs, you can use this action block. If you are a developer, you will feel at home with the contents of this action block. In this action block, there are 2 scenarios that can occur i.e. SUCCESS or FAILURE. To branch out these paths we have 2 separate nodes each denoted by Green or Red. If you want to receive multiple responses from this service call, then this variable can be referenced anywhere in the flow to branch the conversation based on the type of response received from the service call. **Slider** Slider is another unique way of capturing input. Here, in this case, we make use of a slider where the visitor can slide and choose their value. The contents of this block are, * Question - The question for asking this input * Min & Max - The range of values in this slider * Variable - Storing the response on this variable * Enable movement using step count - If active, every movement of the slider will be based on the step count. Else, it will be a dynamic slider where you can move the slider freely. * Unit - The slider values can have a unit i.e. $, €, ₹, £, °C * Count of steps - How many steps are to be displayed in the slider scale **Form** The visitor’s attention in any chat is only for a short period of time. To make the most of it, instead of asking inputs separately, you can ask for all of them at once using a form. With this form, you can ask for a maximum of 5 inputs at once. And each input can have its own validation similar to Collect Input. Some fields can also be marked ‘Optional’. **Dialogflow** If you want to invoke your Dialogflow agent at any point in the conversation, you can connect your Dialogflow agent using this action block. To invoke your Dialogflow agent, you need to send some query like ‘What is your name?’. Enter the variable in which that query was received and this Dialogflow agent will be invoked. Else, Dialogflow will not know what to respond back with. To upload your JSON file, all you need to do is follow the steps mentioned [in this URL](https://dialogflow.com/docs/reference/v2-auth-setup) to generate the Service Account key of your Dialogflow agent. In this action block, there are 2 scenarios that can occur * SUCCESS - If the intent is matched and a response is received from Dialogflow * FAILURE - The failure message to be shown to the visitor Each of these scenarios is represented with a node. Green for Success and Red for Failure. The success node can be self-looped so that whatever input or output exchanged remains comes from Dialogflow. **Codeblock** This block is only for internal use by Team Talkie A special action block, which can be used to call functions made for a specific purpose by Team Talkie. # Trigger Trigger block is the prerequisite with which its underlying conversation flow is executed Trigger action block acts as a starting point for the bot to initiate the conversation with the visitor. By default, the bot is triggered to appear and initiate the conversation without any condition. As a business, you may build multiple bots using the no-code bot builder. Now, deciding which bot is to be shown becomes a challenge. This is where the trigger action block comes in, the user can define rules with which a bot needs to be used. # Send Message The "Send Message" action block allows you to engage with your visitors by sending messages in a chat conversation Action block is used for sending out informative messages that do not seek input from the visitors. The rich text editor supports all the languages i.e. now you can type your input in any language you want like French, Spanish etc.. Delay in sending messages on the basis of word count like if the content of the first message is large, the bot usually delays sending the second message for 2 to 5 seconds allowing the visitor time to read through the first message. **When should I use this block?** There are various reasons for which this action block can be used like for instances it allows you to greet your visitor, share information about the product or service or company as well. **How do I set it up?** Setting up 'Send Message' action block You can follow the following steps to set up '**Send Message**' action block; * Add an block on canvas by clicking on '**+**' * Choose '**Send Message**' * Configure the following on the right panel * **Message:** Configure or set the message that you would like to send it to the visitor. You can also use the variables to personalise the message. Depending on your preference, you can either add text message by clicking on '**+ Text**' or media file as message like GIFs or images by clicking on '**+ File**' **What formatting options do I have?** In the rich text editor, you can format your text using the formatting bar. It allows you to display your message to the user either in the Bold or Italics letters or even underline the selected part of the message. It also allows you to send the message in the bullet points or attach link with selected text. * Bold * Italics * Underline * Bullet Points * Link You can also add emojis to make the conversation more casual and friendly We do support up to 999 characters however we do recommend not to have more than 160 characters in a single message since anything more than this could impede the user experience **Can the messages be re-arranged?** The sequence of the messages / inputs space can also be changed as per the preference. You can hover over the block till the arrows are visible on the left of the input space, left click on arrow (hold), drag (upwards or downwards) & release it to the position as per the new sequence defined. # Collect Input This action block allows you to collect the input or response from the visitors or customers When you as a business want to ask questions to the visitor and expect answers in the return. You don't want to just collect any random input, like you're asking for the visitor's email and they input 'abcedfg', it wouldn't make sense to move ahead in the conversation without receiving a valid input. The collect input block has a set of in-built validations that you can use to verify whether the input provided is valid and only then move forward with the flow. **When should I use Collect Input?** * Collecting information like name, email or phone * Asking questions * Collecting Feedback * Branching the conversation flow based on the input received **How do I set it up?** Setting up 'Collect Input' action block You can follow the following steps to set up the '**Collect Input**' action block; * Add the block on canvas by clicking on '**+**' * Choose '**Collect Input**' * Configure the following on right panel * **Question:** Write down the question you want to ask in the rich text editor under '**Show this question**' * **Variable:** Select the variable in which the response from the visitor will be stored *This variables can be used later in the conversational flow to give personalisation* # Buttons Buttons block is used to branch out conversation using intuitive buttons. You can choose any option to go down its path. Each button can have a same path or different paths moving from it Button action block can be used while building the bot for Web or Facebook channel. **What are different types of Buttons?** * **Branch:** Move to a different path of the conversation flow * **URL:** Open a web page and end the flow Buttons will perform either of the above functions at a time Maximum number of buttons that can be added is 10 The buttons do not disappear after they are clicked on so, the user can return to them at any moment during a conversation **How do I set it up?** Setting Up 'Buttons' action block You can follow the following steps to set up the '**Buttons**' action block; * Add an action block on canvas by clicking on '**+**' * Choose '**Buttons**' * Configure it on right panel * **Defining Question:** Depending upon the preference, you can either add text message or a question that needs to be sent to the visitor for collecting input Using our rich text editor, you can format the text or the question that you want to send to the user. You can also add emojis to make the conversation more casual and friendly * **Defining Buttons:** You can either add Branch button or URL button based on your preference. Lets have a quick look about how these buttons can be configured. * **Branch Button:** Selecting and configuring branch button will allow the user to navigate visitors to different flow based on the selection. '+ Branch Button' will allow you to add multiple buttons. In-order to remove the button, you can double click on the same and give the appropriate name as per the preference. * **URL Button:** Adding URL button will allow you to navigate your visitors from one page to another either in the same tab or in different tab. You can add multiple URL buttons by clicking on '+ URL Button'. To configure the URL button, you can follow the following steps; * Click on the '**Link Icon**' displayed on the left of the button * Copy / Paste the '**Link / URL**', you want the visitors to be navigated to * Choose whether to open in the '**Same tab**' or in '**New tab**' * Click on '**Right**' icon to configure the URL Adding URL button will end the conversation flow * **Define Variables:** Store the selected button in variable. Doing this will assist you to know what button the visitor has selected, which can be further used while creating the lead in the CRM or sending the data to any other internal systems The input from the visitor will not be stored in variable if the button type is '**URL Button**' # Carousel Carousel action block is set of cards that represent something whether its product or serviceThere are two types of Carousel blocks that Talkie offers. They are; 1. **Image Carousel:** When you just want to display show a set of images in a card, not requiring any input to move forward in the conversation flow 2. **Carousel:** When you want to display details along with image along with CTA buttons to move forward in the conversational flow #### When should you use it? * Displaying set of products or services * Wanting to let the visitors choose from variety of options * Give an image for a button * Showing a set of images #### How do I set up 'Carousel' action block? Setting up 'Carousel' action blockYou can follow the following steps to set up the 'Carousel' action block; * Add action block on canvas by clicking on '+' * Choose 'Carousel' * Configure it on right panel * **Adding a card:** Click on '+ Card' to add an card. Maximum 10 cards can be added in a 'Carousel' * **Configuring the card:** Click on 'Card' to configure the following; 1. **Card Image:** Click on 'Upload an Image' to upload the image 2. **Card Name:** Provide the name of the card in the text editor 3. **Card Description:** Provide description about the card in the rich text editor * **Buttons:** Carousel supports two type of buttons; 1. **Call to Action:** Using this button, it will allow you to decide the next set of action or flow visitor will be navigated to 2. **URL Button:** Using this button, you can configure the URL where you would like to navigate the visitor to Maximum 3 buttons can be added to one card. # Calendar Calendar action block is used for collecting date from the or for booking appointments **When should you use this?** * Collecting a date from the visitor * Booking appointments **How do I set it up?** Setting up 'Calendar' action block To set up the 'Calendar' action block, you can follow the following steps; * Add action block on canvas by clicking on '+' * Choose 'Calendar' * Configure following on right panel * **Date Selection:** Configure the following under 'Date Selection' in the configuration message; 1. **Show Message:** You can set the message that you would like to send to the visitors along with calendar to book the appointment 2. **Define Variable:** Define variable in which the response of the visitors will be stored 3. **Date Selection:** Select nature of date selection from available options . You can enable 'Show Past Dates' to show the visitors past dates * **Exclude Dates:** Select the 'Exclude Dates' tab and define any exclusion of dates that you don't want to be visible in the calendar UI. You can exclude dates by either; * Selecting the days all together like Saturdays, Sundays etc.. This way all of these will not be selectable to the visitor. * Selecting special dates like Christmas, Diwali, EID etc.. # Send an Email This action block of Talkie allows you to get notified over email when any particular action in bot is performed. **When should you use it?** * When the lead is generated * When the visitor has entered their details * When a conversation has been completed * When a conversation has reached a particular point **How do I set it up?** Setting up 'Send Email' action block To setup the 'Send Email' action block, you can follow the following steps; * Add action block on canvas by clicking on '+' * Choose 'Send an Email' * Configure following on right panel
* **Send an email to:** Define the email IDs of the users who need to be notified via email **To:** **CC:** **BCC:** * **Subject:** Define the subject of the email * **Email Markup:** Define the content of the email * **Include Chat Transcript:** You can switch on 'Include Chat Transcript' if you would like to receive entire chat transcript over the email # Condition Create conditions to branch your chatbot flow. For chatbot flows requiring a logical condition to be evaluated to branch out to the next step in the flow, you can use the condition action block. You can create a set of conditions for each branch in the chatbot flow. Each condition is made up of: * **Variable** - which contains the value to be compared. * **Operator** - specifies the operation to be carried out. * **Value** - the value against which the variable's value is to be compared. All conditions will be evaluated with **ANY** if you want any one condition to be satisfied, and ALL if you want all conditions to be satisfied. ### Operators supported | Operator | Description | Example | | ---------------- | ------------------------------------------------------------------------------------------------ | ------------------------------------- | | Equals to | Commonly used when you want to match the value of the variable with the defined value | "City" '**Equals To**' "New York" | | Not equals to | Commonly used when you don't want the value of the variable to be matched with the defined value | "City" '**Not Equals To**' "New York" | | Contains | Commonly used when you want the variable to contain the defined value | "URL" '**Contains**' "Web" | | Does not contain | Commonly used when you do not want the variable to contain the defined value | "URL" '**Does Not Contain**' "Web" | | Is empty | You can use this operator to identify whether the variable is empty or not | "Phone" is '**Empty**' | | Is not empty | You can use this operator to ensure that variable is not empty | "Name" is '**Not Empty**' | ### Default condition If none of the conditions are satisfied, the chat flow will automatically proceed toward the 'Default condition' branch on the bot builder canvas. This default branch cannot be deleted. # Image Carousel Action block is used to display one or more images in a carousel to make it easier for the visitors to browse through **When should you use it?** This is usually used when you just want to show a set of images in a card, not requiring any input to move forward for instance; * Displaying set of products or services * Gallery of images #### How do I set it up? You can follow the following steps to set up Image Carousel action block; 1. Add an action block to canvas by clicking on '**+**'
2. Choose '**Image Carousel**'\ 3. Configure it on the right panel
* **Upload Image:** \ \ You can add the image / photo of your choice that you would like to display it to the visitors by clicking on '**Upload Image**'
* **Add more images:** \ \ To add multiple images, you can click on '**Add Image**'. Maximum images that you can add is 10 # Human Handover This action block is used when you want to invoke a human handover action at some point in the conversation flow thereby transferring the control of the chat from the bot to an agent **When I should use it?** * Transferring the chat to human agent * Involving your support team when questions are not answered by the bot ### How do I set it up? You can follow the following steps to set up the Human Handover action block; * Add an action block on canvas by clicking on '**+**'
* Choose '**Human Handover**'
* Configure it on the right panel
* **Transfer conversation to:** \ \ Select the team to whom the conversation needs to be transferred to\
* **Success Message:** \ \ The message configured here will be sent to the visitors when Human Handover is successfully invoked
* **Disabled Message:**\ \ The message configured here will be send to the visitor when Human Handover is disabled\
* **No agent is available:** \ \ The message configured here will be sent to the visitors when all the agents are either occupied or not available \
* **Beyond operational hours:** \ \ The message configured here will be sent to the visitors when the Human Handover is invoked beyond operational hours {% hint style="info" %} When Human Handover is triggered, the assigned user will receive a browser notification to let them know that a conversation has assigned to them\ \ If 'Auto Assign' is enabled for Human Handover, then the conversation will be automatically get assigned to available agent or else the conversation will be placed in the 'Unassigned' tab\ \ Once the conversation is assigned to the agent, the same cannot be re-assigned back to the bot {% endhint %} ### What are the scenarios in which Human Handover can fail? * When Human Handover is disabled * When there are no agents available * When Human Handover occurs beyond operational hours # Slider Action block provides interactive way to the visitors to choose a value from a visual slider ### When should I use it? * Values are to be selected from a range of finite values * To make the conversation more interactive ### How do I set it up? You can follow the following steps to set up the Slider action block; * Add an action block on canvas by clicking on '+'
* Choose 'Slider'
* Double click on 'Slider' block to configure it on the right panel
* **Ask a question:** \ \ Write down the question that you want to ask your visitors
* **Range:** \ \ Enter the range values (Minimum to Maximum) that you want to allow the visitors to choose from
* **Define Variables:** \ \ You can define the variables to save the response received from the visitor
* **Step Count:** \ \ You can enable step count by switching on the switch. Enabling this will move the slider in the increments of step count\ \ The value for the step count can be between 2 & 5
* **Units:** \ \ Select a unit for the slider values from the drop down list i.e $, ‎€, ₹, £, °C. By default, we do not have any unit assigned to the slider # File Upload This action block is used when you would like your visitors to submit documents or files through bot. **In what formats does the file can be uploaded?** * .pdf * .jpg * .png * .doc * .docx * .ppt * .xsl * .xlsx * .csv Maximum size of the file supported is 10 MB Maximum number of uploads allowed is up to 10 **How do I set it up?** Setting up 'File Upload' action block You can follow the following steps to set up the 'File Upload' action block; * Add an action block on canvas by clicking on '**+**' * Choose '**File Upload**' * Double click on '**File Upload**' to configure it on the right panel * **Message:** Define the message or question that will appear to your visitors before collecting the file. * **Type of File:** Select the type of file that you want to limit the visitors to upload. This will restrict the bot from accepting the files of the format which are not selected by you. * **Define Variable:** Configure the variable where you would like the file to be uploaded or stored. This will allow you to share the link of the file uploaded by the visitor later in the flow. * **File Size:** Set the limit of the file size that the visitor can upload. The maximum size supported is 10 MB. Other sizes that you can choose are 1, 2 & 5 MB. In-order to change the maximum size of the file, you will need to click on 'Drop down arrow' & select from the available options. * **Upload Limit:** Set the upload limit i.e the number of times visitor can upload a file. Maximum number of uploads allowed per visitor is 10. The other available options are 1, 2 & 5. In-order to change the upload limit, you will need to click on 'Drop down arrow' & select from the available options. # Delay This action block allows you to configure wait time between the two messages sent by the bot # Form In scenarios, where you are looking to collect multiple inputs like name, email or phone to confirm the appointment, rather than making the conversation lengthy by using multiple collect input blocks to collect each value, you can make use of form that collects everything at once. ### When should I use it? * You are collecting multiple inputs * Want to keep the chat short and quick * Reduce the number of to and fros ### How do I set it up? You can follow the following steps to set up '**Form**' action block; * Add an action block on canvas by clicking on '**+**'
* Choose '**Form**'
* Double click on '**Form**' to configure it on the right panel
* **Message:** \ \ The message configured here will be sent to the visitors when 'Form' action block is triggered along with the form fields
* **Add fields to capture individual input:**
1. **Name:** \ \ Enter the name of the field. This is what the visitors will see
2. **Validation Method:** \ \ Select the validation method i.e text, name, email, phone or custom
3. **Define Variable:** \ \ Give a name to the variable in which the answers to this field will be stored \ You can click on '**+ Form Field**' to add more fields and repeat the above step to configure it. {% hint style="warning" %} You can add maximum upto 5 fields\ \ You can have some fields as 'Optional', so that the visitors can still move forward in the conversation if it is not filled {% endhint %} # Flow Action block allows you to break down larger chatbot flows into smaller manageable chunks **How does it help?** * Creating smaller manageable flows for each path * Easier navigation between conversational flow * Easier onboarding of teams to the chatbot **How do I set it up?** Setting up 'Flow' action block You can follow the following steps to set up the '**Flow**' action block; * Add an action block on canvas by clicking on '**+**' * Choose '**Flow**' * Enter '**Flow**' Double click on '**Flow**' to enter or single click on it and then click '**Enter**' on the right panel * Build the flow on the canvas Once you enter the '**Flow**', it would look similar to '**Bot Builder**'. You can continue building your chatbot flow here in the flow just by clicking on '**+**' as you would otherwise. To navigate between the '**Root Flow**' and other created flows, you can just click on the name of the flow or '**Home**' button visible on the top left corner or the '**canvas**' You can create as many '**Flows**' as you want # Jump This action block allows you to jump from one action block to another without connecting them on the bot builder # Codeblock Codeblock allows you to create and execute custom functions built by Talkie. These functions are made by the Talkie team to deal with complex requirements on customer projects. # Answer AI Uses an LLM-based AI model to generate an answer for the question asked. Available on the **STARTER** plan and above. The Answer AI action block is used to generate an answer to the user's question from the trained data. When using the answer AI block, the user submits an open-ended question, which is then processed by the AI Studio to generate a relevant answer. ZyndraAI/Talkie AB uses OpenAI GPT APIs to generate an answer. Process in which the Answer AI works, 1. User asks a question 2. AI generates the answer from a knowledge base ### Step 1: Define the question This message will prompt the user to ask their question. For example, your message could be, > 👋 Hey! I'm an AI assistant, how can I help you? ### Step 2: Choose a knowledge base Choose the knowledge base from which AI will generate an answer. ### Step 3: Configure your AI for accurate answers LLM models require guidance to be able to generate a relevant and accurate answer. There are some tools you can use to guide your AI; they are: #### Instructions/Prompt A prompt is nothing but an instruction that helps the LLM know what to remember and follow while generating an answer. The more clear, concise and brief your instruction, the more accurate your answers will be from the AI. Things to write in your prompt: * **Objective** - What is its objective * **Output format** - Typically would be HTML/markdown. * **Writing style** - How should the answers be written. * **Don'ts** - Clear instructions of what to avoid. * **Examples** - Examples of question and answer. #### LLM model Select the LLM model that will be used to parse your training data and generate an answer. Currently, ZyndraAI/Talkie AB only supports the following models: * gpt-4o * gpt-4-turbo * gpt-4 * gpt-3.5-turbo #### Restrict answer size The maximum length of the answer is defined in tokens. Tokens can be thought of as pieces of words, and 1 token=4 characters. [Learn more](https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them) #### Creativity in responses You can control whether you would like to receive unique responses every time a question is asked or get the same response every time the question is asked. Define a number between 0 and 1. Nearer to 1 gives you distinct responses; nearer to 0 gives you fix responses. By default, we keep the number as **0.5**. #### Chunks When the AI is asked a question, it first fetches relevant content from the knowledge base, which is then sent to the LLM model to generate an answer. The more relevant content the AI can fetch, the more chances of accurate responses are. The number indicates the number of relevant responses which will be fetched from the knowledge base. #### Context length Sends the recent sets of question answer pairs to provide the LLM more context of the current ongoing chat. Benefit of this is that you can ask a follow-up question as 'its pricing', and the LLM would be aware that your previous question was about ZyndraAI/Talkie AB, so you meant "ZyndraAI/Talkie ABs pricing'. #### Ask for feedback For every AI generated response, we allow the user to submit feedback about the generated answer. This helps you know how helpful are AI's answers. #### Answer source Displays the URL of the content source from where the answer was referenced from. This only works for knowledge bases built using a website URL. # Set AI Uses an LLM-based AI model to generate an response based on the prompt given. Available on the **STARTER** plan and above. The Set AI action block is used to generate a response to the prompt supplied to it. When using the Set AI block, the user submits a prompt, which is then either processed by the AI Studio or the LLM model to generate a response. This response is then stored into a variable ZyndraAI/Talkie AB uses OpenAI GPT APIs to generate an response. Process in which the Set AI works, 1. A prompt is supplied 2. AI generates the response 3. Response is stored in a variable
### Step 1: Define the prompt/instruction LLM models require guidance to be able to generate a relevant and accurate answer. There are some tools you can use to guide your AI; they are: #### Instructions/Prompt A prompt is nothing but an instruction that helps the LLM know what to remember and follow while generating an answer. The more clear, concise and brief your instruction, the more accurate your answers will be from the AI. Things to write in your prompt: * **Objective** - What is its objective * **Output format** - Typically would be HTML/markdown. * **Writing style** - How should the answers be written. * **Don'ts** - Clear instructions of what to avoid. * **Examples** - Examples of question and answer. An example prompt could be: > Given the ‘user’s question’: “\[QUESTION]” > > And the detailed information provided in ‘chunks’: “\[CHUNKS]” > > Determine, whether a clarifying question is required. > > Instructions: > > 1\. Analyse the 'chunks' and the 'user's question' to identify the specificity of the query and the scope of the information in 'chunks'. > > 2\. If the query is broad and the 'chunks' have multiple categories or types, output '#' and guide the chatbot to ask for clarification. > > 3\. If the query aligns well with a specific part of the 'chunks' that provides a comprehensive answer, output '\~'. > > Output format: \[Decision: '\~' or '#', (if '#') then clarification is required. If '#' also, 'specify the type of information or category that would help better address their question. This should be based on 'chunks'. > > Important: if the user's question is likely to have a device specific answer, then you should ask for more information. > > If the user has given a device, then we don't need to clarify. ### Step 2: Store the response in a variable The AI would process the instruction/prompt and generate a response. To get an ideal response, make sure you include in the instructions what kind of response you want to receive. Choose the variable you want to store the response in. For example, if you want to check if the user has asked for a follow up question, you can write the instruction like: > Examine the user's last utterance: > > "\[LAST INPUT]" > > Determine if the user has asked a follow-up question by looking for: > > \- Interrogative words (who, what, where, when, why, how, etc.) > > \- Phrases that indicate a desire for additional information (e.g., "I would like to know", "Can you tell me about", "I'm interested in", "Could you explain") > > \- Continuation phrases or conjunctions that introduce new topics or questions (e.g., "but", "however", "also", "in addition") > > \- Output '1' if any of these indicators suggest a follow-up question is present. > > \- Output '0' if no follow-up question is detected. > > \[Note: Only output '1' or '0' based on this analysis.] # AI-Agent Runs your AI agent with prompts and functions to generate a response. ZyndraAI´s "AI-Agent" action block is its most advanced feature for building smart, conversational AI agents. This guide will show you how to use it to manage complex interactions, pull from a knowledge base, follow custom logic, and run functions—all in one streamlined step. The Agent step provides a comprehensive solution for creating AI agents that can: * Intelligently respond to user queries * Search knowledge base for relevant information * Follow specific conversation paths based on user intent * Execute functions to interact with external services When using an "Agent" step, you need to configure the following: ### Prompt Effective prompting is key to building a high-performing AI agent. The instructions you give act as the agent’s “brain,” shaping how it engages with users across different situations.
Here’s what makes a strong prompt: * Clear Directives: Define exactly what the agent’s role is and how it should behave in various scenarios. * Personality Design: Set the tone, formality, and communication style your agent should follow. * Knowledge Scope: Outline what the agent should know and when to refer to external sources. * Context Handling: Provide guidance for sensitive topics and when to escalate to a human. You can learn to create a prompt [here](https://help.talkie.se/ai-studio/prompts). Good prompting helps your agent stay on track—knowing what to prioritize, when to tap into the knowledge base, and how to respond consistently. Investing time in clear, detailed instructions not only boosts performance but also simplifies your flow. **Example**: Instead of saying: *“This is a support agent.”* Try: *“You’re a helpful, friendly e-commerce support agent who assists with order tracking, returns, and account questions. Keep a positive tone, offer suggestions when possible, and be specific in your responses.”* ### Function Functions allow your agent to connect with external services and retrieve or update data. Choose from available functions in the dropdown menu OR create a new one by navigating to AI Studio > Functions.
### Path Paths help your agent manage targeted conversation flows and respond with the right actions at the right time.
Here’s what Paths enable: * Automatic Routing: The agent identifies user intent and directs the conversation to the relevant path. * Data Gathering: You can specify which variables must be collected before a path is triggered. * Natural Flow: Paths fit smoothly into conversations without needing exact commands from the user. **To create a Path:** 1. Open the Agent step editor. 2. Click “Add Path” and give it a clear, descriptive name. 3. Write a brief description outlining when this path should activate. 4. (Optional) Add required variables that need to be collected first. 5. Link the path to the next steps in your chat flow. **Path Examples:** **Human Handoff Path** * Path Name: Transfer to Human Agent * Description: Trigger this path when the user explicitly requests to speak with a real person. Phrases like “I want to talk to a human,” “Can I speak to an agent?” or “I need help from a real person” should activate this path. * Required Variable: None * LLM Description: This path allows the agent to hand over the conversation to a human agent. Ensure the system recognizes phrases requesting a human interaction and promptly transfer the conversation. **Book Demo Path** * Path Name: Schedule Product Demo * Description: Trigger this path when the user expresses interest in booking a product demo or meeting. Look for phrases such as “I want to book a demo,” “Can I schedule a meeting to see the product?” or “How can I try the product?” * Required Variables: * Name * Email * Variable Descriptions: * Name: Collect the user’s full name so the meeting can be scheduled properly. * Email: Capture the user’s email to send demo details and confirmation. Ensure the email is in a valid format (e.g., contains @ symbol and proper domain). This path ensures that when a user wants to book a demo, all necessary details like their name and email are captured before proceeding with the demo scheduling. ### LLM settings The LLM Settings section allows you to customize how the language model behaves: * LLM Model: Choose which language model (e.g., GPT-3, GPT-4) you want to use for generating responses. Each model has different capabilities, with newer models generally offering more advanced features. * Max Tokens: This setting controls the maximum length of the response. A higher token limit allows the model to generate longer responses, while a lower limit results in shorter replies. * Temperature: This adjusts the randomness of the model’s responses. A higher temperature (e.g., 0.7) makes responses more creative and varied, while a lower temperature (e.g., 0.2) makes them more deterministic and focused. These settings help fine-tune how the AI interacts with users based on your desired response style and length.
# Overview How Talkie integrates with third-party apps Here are the 3 ways in which Talkie integrates with applications ## **Native Integrations:** You can find all the native integrations displayed under the 'App Market' icon on the nav bar. ## **HTTP Requests:** Using the [HTTP request ](/integrations/http-request)action block you can invoke a REST API request to exchange information securely between applications. You can perform GET, POST, and PUT requests. **Zapier:** If we do not have a native integration with your desired app, you can use Zapier, which connects with over 7,000+ apps worldwide. Add the 'Zapier' action block to your chatbot flow and connect it. ## Custom Integration: If you are not able to integrate your app using the above methods, Talkie can custom develop the integration for you. {% hint style="info" %} Custom developing integrations are charged on a time and material model. Send your inquiries at **** {% endhint %} # HTTP request Performs an HTTP request with a third-party app's REST API endpoint. If you want to send data or fetch data from a third-party that ZyndraAI/Talkie AB does not have a native integration to, you can use the HTTP request action block. HTTP request allows you to invoke a API endpoint and perform a **GET / POST / PUT** operation. ### Request Configure the details for your API request. #### Methods Choose the method of your API endpoint you want to invoke. ZyndraAI/Talkie AB supports: * GET * POST * PUT #### **Endpoint URL** You must provide the absolute URL, i.e., a URL starting with HTTP or HTTPS. You can also use the variables in the URL by typing '#', as some APIs do not have headers and mention everything in the URL itself. Copy ``` https://api.weatherapi.com/v1/current.json?key=e943b863ebed4572b55140350200312&q=#city# Here in the above URL #city# is the variable that has been used ``` #### **Headers** Add authentication data to the HTTP request. Commonly used for bearer token based authentication. Make sure you add the **content-type** field to your request. #### **Body** Add the required information you want to send as part of your request. You can send information in two ways: **Raw:** It is normal JSON format in which the request needs to be passed ![](https://help.wotnot.io/~gitbook/image?url=https%3A%2F%2F360969599-files.gitbook.io%2F%7E%2Ffiles%2Fv0%2Fb%2Fgitbook-x-prod.appspot.com%2Fo%2Fspaces%252FsrMxU8nO3RjusUiYuXBB%252Fuploads%252FwJiVZlWrx5Jo8cxZmlRd%252FCleanShot%25202024-06-20%2520at%252019.14.22.png%3Falt%3Dmedia%26token%3D24e2f856-bad1-43b8-ae58-8e0b2389c3b9\&width=768\&dpr=4\&quality=100\&sign=e4e6b64f\&sv=1) **Form:** It allows you to map the fields. You can either provide static value or map the variables ![](https://help.wotnot.io/~gitbook/image?url=https%3A%2F%2F360969599-files.gitbook.io%2F%7E%2Ffiles%2Fv0%2Fb%2Fgitbook-x-prod.appspot.com%2Fo%2Fspaces%252FsrMxU8nO3RjusUiYuXBB%252Fuploads%252F0LDmXOTMkpI0aogiR23M%252FCleanShot%25202024-06-20%2520at%252019.15.31.png%3Falt%3Dmedia%26token%3D6bcd41cd-df5b-4a8e-94ad-8134ba6f3eff\&width=768\&dpr=4\&quality=100\&sign=e60d9e35\&sv=1) ### **Response** When the HTTP request is made, you will receive a response from the API. You can store the objects received from the API into variables by mapping the fields. Copy ``` // Object path for storing the response into variable where list is returned result.data.list[0].id // Here the value of the field `id` will be stored in the variable mapped to it. ``` Copy ``` // Object path for storing the response into variable where single value is returned result.data.id // Here the value of the field 'id' will be stored in the variable mapped to it. ``` ### Testing API request Once you have configured your API request, you can click on the 'Test the API' button and see the response received. From the response screen, you can directly save the object into a variable by clicking on the + icon. # Hubspot Action block allows integration between Talkie and Hubspot to create or record leads real time Hubspot is a popular CRM of choice for all kinds of businesses. Although it has grown to offer other functions like ticketing, email automation, etc.. it is primarily being used by businesses as their CRM (Customer Relationship Management) tool. With this integration, you will be able to create a contact on Hubspot realtime based on the details collected by the bot. ### How do I set it up? To set up the integration between Talkie and Hubspot, you can follow the following steps; * Add an action block on canvas by clicking '+'
* Choose 'Hubspot'
* Double click on 'Hubspot' to configure it on the right panel
* Connect the account:\ \ Connect your Hubspot account by authorising the access and choose the account from the list\ \ Hubspot's authorisation will open on a new window. Please make sure that you have not blocked pop-up window from being opened.
* Map the fields:\ \ Map the fields of hubspot with the variables that you used to collect and store information of the visitors. \ \ Hubspot fields are auto-pouplated, whereas in the variables columns you would need to write the variable by typing '#' to see the list of variables. \ \ Depending on the field type, Talkie would automatically display the variable column. {% hint style="info" %} Make sure you add a lead source field with the value as Talkie to easily report the number of leads the chatbot generated for your business. {% endhint %} # Salesforce Create a lead object on Salesforce Available in the Premium plan and above. Using the salesforce integration, you can create a lead on your Salesforce account using the information collected by the chatbot in the chat flow. **Connect your Salesforce account** First, you must connect your Salesforce account by adding your organization ID. You can either add it from the action panel on the App Market page. You can find your Salesforce org ID -
### **Choose Environment:** You can choose the environment that you want to use for this action block. * Sandbox * Production ### **Map the Salesforce fields:** You can map the Salesforce fields with the variables defined in the bot builder. This way, whatever values stored in ZyndraAIs variables will be mapped against the Salesforce fields that you choose here You can map both default and custom fields of Salesforce here The maximum number of fields you can map is capped at 50 # Slack This integration enables you to send notifications and chat directly the visitors on your website or any other channel on which the chatbot is enabled, providing the ability to resolve all the queries ### Use-case of integration with Slack: * Get instant notification * Resolve queries instantly * Respond directly through Slack ### How do I configure it on Talkie? To connect your Slack space with Talkie, you can follow the following steps; * Go to 'App Market -> Slack -> Accounts'
* Click on 'Add Account'
* Select the 'Workspace' you want to configure
* Provide required permissions
* Click on 'Allow' to connect the Space {% hint style="info" %} You can connect only one Slack Workspace with Talkie {% endhint %} ### What can I use Slack for? You can use Slack for; * Receiving notifications * Interacting with the visitors # Live Chat for Slack Slack for Live Chat allows you to interact with your visitors directly through Slack, providing them instant resolution to their queries or questions related to product or service The benefit of using Slack for Live Chat is that your agents will not need to learn new tool, rather converse at the comfort of the Slack threads.The conversation with the visitors will happen over thread allowing to have separate thread for each conversation. #### How do I configure Slack for Live Chat? Process to configure Slack for Live Chat is divided into two steps; 1. On Slack 2. On Talkie **On Slack:** To add 'Talkie' application on Slack, you can follow the following steps; * Go to 'Apps' on Slack & look for 'Talkie' to add the application on Slack * Click on 'Add' * Create 'Public Channel' from where the customer support team can join the chat If you are using a 'Private Channel', you will need to 'Open Integrations -> Add Apps -> Select Talkie' to add Talkie to that channel **On Talkie:** To configure it on Talkie, you can follow the following steps; * Add action block on canvas by clicking on '+' * Choose 'Slack for Live Chat' * Double click on 'Slack for Live Chat' to configure it on the right panel; * Workspace: Select the workspace from the dropdown list where the live chats needs to be configured * Channel: Select the channel from where the customer support team will take up the chat and interact with the visitors * Assignment Rules: Assignment rules will define which agent the conversation will be assigned to. You can create up to 10 assignment rules, in case the visitor does not receive a response after X seconds in wait time. Each rule consists of the team or teammate to whom the chat needs to be assigned, and the wait timeout after the flow moves to the next rule, until a response is sent to the visitor. That is, if the assigned agent doesn't join the conversation within the specified wait time, the next assignment rule will be invoked. Conversation will be assigned to team or teammate in round robin manner Assign the wait time in seconds By default we have set the wait time as 30 seconds Minimum wait time that can be set is 10 seconds Maximum wait time that can be set is 10,800 seconds You can set maximum up to 10 assignment rules #### What actions can be performed on Slack? You can perform the following actions on Slack live chat; * View Details: 'View Details' will be highlighting the details about the visitors like City, Country, Zip Code, Bot Name, Browser, Language etc.. * Re-Assign Conversation: You will have the ability to assign the conversation to one of the team members by clicking on 'Re-Assign Conversation' * Close Conversation: You will be able to close the conversation by clicking on 'Close Conversation' # Notification for Slack Being alert can always be very rewarding. This action block allows you to configure notifications on Slack whenever visitors initiates conversation with chatbot With our Notification for Slack integration, you can strategically get notified on your Slack channel when a visitor reaches a certain point in the chatbot flow. This allows or enables you to immediately strike conversation with the prospects at just the right time. ### How do I configure the Notification for Slack? In-order to receive notification on Slack, you need to follow the following steps; * Add action block on canvas by clicking on '+'
* Choose 'Slack Notification'
* Double click on 'Slack Notification' to configure it on the right panel;
* Workspace: \ \ Select the workspace from the dropdown list where the notifications needs to be received
* Channel: \ \ Select the channel where the notification will be posted
* Thread ID: \ \ Depending upon your preference, you can either assign the value or leave it blank. Choosing to use this, will always post notifications inside the thread
* Define Variable: \ \ Define the variable where the 'Thread ID' will be stored. This can later be used in flow.
* Broadcast Message: \ \ Check the box against 'Broadcast Message' if you would like to send the notification across channel
* Message: \ \ Defining the content of the notification that will be pushed on Slack. You can reference a variable by typing '#' and choosing the variable {% hint style="info" %} The channel needs to be public in order to post the notification {% endhint %} ### What actions can be performed on Slack Notification? Following are few of the actions that you can perform on Slack Notification; * Join Conversation: \ \ 'Join Conversation' option on Slack Notification that allows you to instantly join the conversation and interact with the visitor
* View Details: \ \ 'View Details' will highlight the details about the visitors like city, country, Zip Code, Bot Name, Browser, Language etc..
* Re-Assign Conversation: \ \ You will have the ability to assign the conversation to one of the team members by clicking on 'Re-Assign Conversation'
* Close Conversation: \ \ You will be able to close the conversation by clicking in 'Close Conversation'
# Public APIs All of ZyndraIAs APIs use a Bearer token based authentictation. Your account token can be found at Settings > Account Settings.
### Adding a data source to a knowledge base `POST` `/api/v1/ai/knowledge-base//upload/sources` Using this API, you can add a data source to an existing knowledge base. **Headers** | Name | Value | | ------------- | --------------------- | | Content-Type | `multipart/form-data` | | Authorization | `Bearer ` | **Body** The API accepts files and text as multipart form data for upload. Each form field should be a unique key and contain either a file or text content. **Response** 200400Copy ``` { "q1": [ { "id": 330783, "type": "text", "name": "sample.txt", "status": "in-progress", "errors": null }, { "id": 330784, "type": "text", "name": "sample2.txt", "status": "in-progress", "errors": null }, { "id": 330785, "type": "file", "name": "File.pdf", "status": "in-progress", "errors": null } ], "d2": [ { "id": 330786, "type": "file", "name": "File2.pdf", "status": "in-progress", "errors": null } ], "knowledge_base_id": 876, "ok": true } ``` ### Get training status of a data source `GET` `/api/v1/ai/status/sources` Using this API, you can fetch the status of one or more data sources in a knowledge base to know it its training is - in progress / completed / failed. **Headers** | Name | Value | | ------------- | ------------------ | | Content-Type | `application/json` | | Authorization | `Bearer ` | **Body** | Name | Type | Description | | ------------ | ------------ | ------------------------------------------------------------------------- | | `source_ids` | query params | Comma-separated list of data source IDs to be passed as **query params**. | **Response** 200 400 ``` { "sources": [ { "id": 330778, "name": "sample.txt", "status": "in-progress", "knowledge_base_id": 876 } ], "ok": true } ``` # Purchasing a plan Quick walkthrough of Subscribing or Unsubscribing from a plan based on your preference With Talkie, you have the liberty to upgrade or degrade your plan at any given point of time. When you upgrade to a different plan from an existing one, the payment for the same will be initiated using Credit or Debit card. To upgrade your account to the higher package, you can follow the following steps; * Go to 'Settings -> Account -> Billing' to access the 'Billing' screen * Click on 'Get Subscription' * Toggle between 'Billed Monthly' or 'Billed Annually' as per your preference * Click on 'Choose' to choose the plan from the available options based on your preference & requirement * Fill up the required details on the billing page like Billing Name, Address, Card Details etc.. * Click on 'Pay using this Card' to initiate the payment In case you are looking for our 'Enterprise Plan', you can reach out to us at # Payment Methods Different types of payment methods that Talkie accepts You can initiate the payment for Talkie subscription using Credit or Debit card. * **Credit or Debit Cards:** Monthly or Annual Subscriptions {% hint style="info" %} All payments are processed and managed by Stripe
The amount will be auto debited from the Credit or Debit card on the day of Renewal {% endhint %} # Initial cost, notice period, commitment period At Talkie AB, we don't believe in complicated agreements. Therefore, we have neither start-up cost, notice period nor commitment period. You can terminate your contracts at any time. And you do this by logging into your dashboard - press settings - Billing - Cancel. # Spent conversations If you have used up all the conversations included in your subscription, there are 2 options: 1. Your chat will be inactive until the next monthly period 2. You can buy extra conversations for the low cost of SEK 100 for 1000 extra conversations. This corresponds to about 10 USD. This will happen completely automatically. For example, if you have our starter plan which includes 1000 conversations and you end up with 1978 conversations, this will cost SEK 100 (USD 10) extra # What currency do you charge in? Our prices are in Swedish kronor, but it will be automatically converted to your local currency upon transaction. Our "starter" plan, which costs SEK 1,500/month, corresponds to about USD 150