The Answer Bot is an NLP-based chatbot exclusively designed to answer questions using the SalesIQ resources. This chatbot provides 360-degree support assistance to your visitors and users. When a visitor/user asks a question in the SalesIQ's live chat widget on your website, the answer bot will take up the chat, search for an appropriate answer from your Answer hub (articles, FAQs & webpages), and send it. If the visitor feels the answer is not solving their query, they will have the option to connect with you (operator), and the chatbot will transfer the chat, allowing you to engage personally. 

An interesting fact about Answer bots is that you don't have to design or build them. They are prebuilt, and you just have to set them up and enable them.

Why should businesses use Answer bot?

If your business deals with routine questions on a daily basis, instead of you personally handling each of these chats/queries repeatedly, you can use Answer Bot.

1. This bot specializes in addressing common, level 1 questions, saving you time for more complex questions. 
   
2. Handle a higher volume of queries efficiently by automating repetitive questions across sales and support.
3. Deliver consistent, business-aligned responses by grounding answers in your knowledge base instead of generic AI output
4. Reduce routine workload for operators using AI to handle common queries so operators can focus on complex conversations
5. Maintain control over AI responses by defining how the bot handles valid, invalid, or failed answers
   

If you want your bot to be more than just answering questions, we recommend using Zobot. With Zobot, you can build compelling chatbots to automate customer interactions. Zobot is entirely user-based, meaning it acts according to how you build and define it. It's highly customizable and can be designed based on whatever trends and business criteria you need.

You can bring an Answer bot inside Zobot using the Question & Answering card in the Codeless bot builder and the auto_pilot action for other bot-building platforms.

How does Answer bot work?

Zia

1. Zia uses the SalesIQ knowledge center/Zoho Desk's Knowledge base to generate answers.
   
2. Searches Articles, FAQs, Small Talks, and Webpages
3. Identifies the most relevant resources based on the visitor/user question. 
4. Generates a summarized response and displays as answers. 
   

OpenAI, Anthropic, Google AI, & DeepSeek

1. Zia retrieves up to 5 relevant resources from SalesIQ/Desk resource
2. These resource are sent to the configured AI service
3. The AI generates a response using both your data and its capabilities
   

Note: The OpenAI Assistant generates responses using its own knowledge. 

Custom service

1. You can also connect any AI service using a custom REST API. 
2. Custom services can optionally use SalesIQ or Zoho Desk knowledge sources. 
   
3. Upon enabling, relevant resources are retrieved by Zia and sent along with the visitor's/user’s query, allowing the AI to generate responses using both business context and its own knowledge.
4. Timeout: 5 seconds. Ensure the AI service responds within this duration for synchronous processing.
5. Asynchronous support: If the response takes longer than 5 seconds, asynchronous handling is supported to process and deliver the response.

Useful links

1. Learn how to bulk import resources
   
2. Learn how to sync desk articles in SalesIQ (only supported languages' articles will be synced)
   
3. Answer bot supported languages

Supported AI services

The Answer bot supports multiple AI engines, giving you flexibility to choose or integrate your preferred service

AI service	Supported models
Zia (Zoho’s in-house AI engine)	Zia uses the most optimal model (no selection required).
OpenAI	gpt-3.5-turbo, gpt-4, gpt-4o, gpt-4o-mini
OpenAI Assistant	Assistants configured in your OpenAI account
Anthropic	claude-sonnet-4-6, claude-opus-4-6, claude-haiku-4-5-20251001
Google AI	gemini-2.5-flash, gemini-2.5-pro, gemini-3.1-pro
DeepSeek	deepseek-chat
Custom Service	Any AI service via REST APIs

Note: To use external AI services, the respective AI must be integrated with SalesIQ. Click here to know more about the AI integration.

Answer bot configuration

To activate the Answer bot for your brand, you need to configure five different sections, which include profile, resources, response, behavior, and fallbacks.

Profile

To set up an Answer Bot, 

1. Navigate to Settings > Bots > Answer Bot.  Choose the brand which you have to enable the bot.
   
2. Give your Answer bot a name and description.
3. Choose the languages you want the bot to converse in Supported languages. Currently SalesIQ supports 34 different languages.  
   
4. From the chosen languages, select the primary/default language for the bot. 

1. When the visitor uses a language which is not configured/supported, the bot will converse in primary/default language. 
   

Language handling

1. Visitor-initiated chat - When the bot is triggered by a visitor message (reactive mode):

1. The bot detects the language of the visitor’s first message
2. If that language is configured, the bot continues the conversation in that language
3. If not, the bot falls back to the default language

Example

1. If Spanish is configured and the visitor sends “Hola!”, the bot responds in Spanish.
2. If Spanish is not configured, the bot responds in the default language.

2. Language switching during chat - Visitors can change the language at any time using the language dropdown at the beginning of the chat.

1. The selected language will be used for all subsequent responses
2. The bot will not automatically switch languages based on new messages
3. Once a new language is selected, the chat restarts from the beginning in the selected language 

Note: If the conversation is in English and the visitor sends a message in Greek, the bot will still respond in English unless the language is changed via the dropdown.

3. Bot initiated chats - When the bot is triggered automatically:

1. SalesIQ detects the website’s language
2. If that language is configured in the bot, the bot starts in that language
3. Otherwise, the bot uses the default language

Next, choose the Chat departments for the bot. The bot will be able to use the resources such as articles, FAQs, etc., associated with the department. 

Resources

Resources are the "Brain of the Answer bot". The resources will be used to train your Answer Bot. The bot will use these to answer the visitor's questions. Choose your resource, the Answer bot supports either Zoho SalesIQ's resource or Zoho Desk's knowledge base. 

1. Select the AI service to power this card
   

1. Identify relevant resource - Configure the answer source for the bot. 
   

Zoho SalesIQ Resource: 

1. The bot uses the SalesIQ's resource (FAQs, Articles & webpages)  to answer the question. 
   
2. Select specific resource or combinations based on your use case. 
   
3. Inside the Small talk dropdown, choose the Small talk package you want your bot to use. 
4. Training status: Click on "Resource training status" to view training status for each resource. If a resource is not trained, you can retry training.
   

Zoho Desk Resource:

1. The bot uses the Desk's knowledge base to answer the question. 
   
2. Zia should be enabled in the Zoho Desk (Setup > General > Zia > and enable Ticket Reply Assistant).
3. A minimum of 30 articles are required in the Desk Knowledge for the Answer bot to work. 
   
4. Zoho Desk integration is mandatory to select the Zoho Desk resource/knowledge base for the bot. 
5. The bot uses the articles from the desk department that is mapped to the bot's department under integration configuration.  
   
6. So, ensure to map the essential desk department with SalesIQ in the SalesIQ-Desk Integration page. 

Note: 

1. Answer bot will not function, when selecting "Choose on Demand" in the SalesIQ-Desk department mapping.
   
2. Visit the spreadsheet to view the list of languages supported by the Answer Bot for Desk resources.

Global data 

Enable global data to handle broader, non-business-specific queries.

1. Define keywords (tags) to guide the AI toward better contextual search.
2. Example - "Trending shoes"
3. Use case - You own a shoe shop, when a visitor asks a general query like “What are the trending shoes for GenZ?”, the AI uses these tags to understand the context and generate a relevant response, even if the query is not strictly business-specific.
   

Supports up to 20 tags, each with 20 characters.

This allows the bot to respond to general queries while still prioritizing business context.

Response 

In this step, you can configure the bot response text and follow-up actions for the Answer found and not found cases. 

Waiting state

1. While Zia generates a response, you can display interim messages. Example: “Searching…”, “Fetching best match…”, "Putting it together".
2. Supports up to 3 messages, displayed sequentially in a loop. 

1. Allow answer streaming - Streams responses in real time as they are generated, reducing perceived wait time.
   

Note: Streaming is not supported for the Small talks and Zoho Desk resources. 

1. List reference sources - Displays the resources used to generate the answer in list.
1. Title for source list - Title to display before the reference. Example: “Know more:”
2. Follow-up action - Configure the follow-up buttons for this case. 
   

1. To customize the display and response text of the follow-up actions, click on the edit icon next to the Follow-up actions header. A pop-up will appear where you can edit the messages for each follow-up action. 
   

1. No relevant knowledge is identified
   
2. Zia cannot generate a usable response
   

Configurations

1. Add a text/message to be displayed when the "Answer not found case is triggered.
2. Follow-up action -  Configure the follow-up buttons for this case. 
   

When the bot is engaged in a conversation in a specific language, only the resources associated with that language will be used. For example, you have an article about the "Return policy" in English but not in French. Let's say you have configured French in the bot's profile. When the visitor asks the bot regarding the return policy in French, the "Answer not found" case will be triggered. 

Ratings

Enable feedback collection for each response. This will collect the rating for a response. 

1. Displays thumbs up / thumbs down options
2. Allows users to rate each answer

Note: Ratings are supported only for the Web and Mobile SDKs. Social channels don't support this.

Trigger actions based on ratings

Define automated actions via plugs based on rating patterns.

1. Occurrence Count - Number of ratings required to trigger an action
2. Mode (one at a time): 
1. Consecutively - Back-to-back ratings
2. Entire chat - Across the full conversation

Unrated messages are not considered positive or negative.

Plug Configuration

1. Select a plug
2. Map required parameters
3. Trigger actions such as notifications, escalations, or workflows
   

Behaviour

1. During business hours: 
   
1. Do not attend any chats  - The bot will not pick any incoming conversations during business hours.
2. Attend all incoming chats  - The bot will pick up all incoming hats during business hours. 
3. Attend chats only when operators are not available  - If the operators are unavailable, the chats will be picked by the bot. 
   
2. During non-business hours:
1. Do not attend any chats  - The bot will not pick any incoming conversations during business hours.
2. Attend all incoming chats  - The bot will pick all incoming chats during business hours. 
3. Attend chats only when operators are not available  - If the operators are unavailable, the chats will be picked by the bot. 
   

1. By default, the audience criteria rule is set to  Visitor type is All . 
2. To customize the rule, click on Manage rules, and now you can choose your desired conditions and click Save. 
   

1. Now, you can set the trigger under the following conditions:
   
1. When the visitor lands on your site and spends more than the mentioned time. 
2. When the visitor clicks on the chat widget.
3. When the visitor performs a custom action.

Note: 

1. When the bot trigger/proactive interaction is set up, SalesIQ will analyse the website's language and checks if the respective language is configured on the bot. If it does, the bot gets triggered in the website language. 
2. If the website's language is not supported in the bot, the bot will trigger in the default language. 

1. Configure the greeting message you want the bot to send out to your visitor when the condition set is satisfied. 
   

1. Reminder  - Based on the inactivity time set in this section, the bot can send out reminders to the visitor that the conversation is still ongoing. You can customize the reminder message and add another message to send multiple reminders at different times. 
2. Conversation timeout  - After sufficient reminders, you can end the conversations after the desired time with the chat closing message.
   

• Drag the slider to desired value to set the bot's response delay.

1. Static handoff link  - If this option is enabled, an option to connect with an operator will be available throughout the bot conversation. You can customize the text and options shown in the static link. 
2. Conversational handoff  - This option will transfer the chat to a human operator when the visitor requests via chat. You can configure the response message once the bot initiates the transfer. 
   

Fallback 

1. Operator busy : The response message and selected follow-up actions will be shown when all the operators are busy when forwarding. 
2. Operators unavailable:  The response message and selected follow-up actions will be shown when all the operators are offline when forwarding. 
   

Test/Preview

Deploy

Useful Links

Learn how to set up SalesIQ's live chat widget on your site

Learn how to personalise the chat widget