Here's how things worked before:

• Every AI-generated response cost 1 AI credit, no matter the size.

• A 200-token reply and a 40,000-token reply were charged exactly the same.

• This made costs unpredictable for us, and unfair for customers who keep their interactions short and efficient.

Here's how it works now:

• AI credits are charged based on actual token usage.

• Each LLM model has a defined credit rate per 1,000 tokens(Blended i.e. input + output tokens) consumed.

• Tokens include everything in the interaction — your prompt, knowledge base, conversation history, and the AI's response.

A simple example: if your model charges 3 credits per 1,000 tokens and a response uses 3,000 tokens, you're charged 3 credits. Use less, pay less. It's that straightforward.

This updated calculation will apply from Apr 20, 2026 10 am IST.

Learn more →

That changes today. Clicking unsubscribe in any campaign email now brings contacts to a WotNot-hosted Subscription Preferences screen, and their opt-out is recorded directly in WotNot.

You can also set up Subscription Groups when designing your campaigns, so contacts can choose exactly what they opt out of rather than going dark entirely.

You can now store sensitive credentials — like API keys and authentication tokens — directly in your project variables using the new Secret data type.

Secret variables are encrypted immediately upon saving, and once set, the value is permanently hidden: no one on the team can view it, only replace it. It's the safe, team-friendly way to manage credentials used in your JavaScript API calls.

Learn more →

With Data Retention, you can set a predefined period after which chats and conversations are automatically deleted — helping you stay compliant and keep your system clean.

What’s new

• Choose retention periods like 5 days, 1 month, 3 months, 6 months, 12 months, or 24 months

• Or keep data indefinitely with “Never”

• Automatic deletion — no manual cleanup needed

Why it matters

• Stay aligned with data privacy and compliance requirements

• Reduce unnecessary data storage

• Keep your workspace clean and efficient

Where to find it

Settings → Account → Data Retention

Previously, voice agents were limited to phone numbers provisioned via Twilio. With SIP trunk support, you now have greater flexibility in how you route and manage voice calls.

What’s new

• Connect voice agents using SIP trunks

• Configure your own provider, credentials, and routing details

• Support for custom protocols, URLs, and ports

Why it matters

• Use your existing telephony setup

• Reduce dependency on Twilio

• Gain better control over call routing and costs

• Enable enterprise-grade voice integrations

Where to find it

Settings → Voice Channel → Configuration

This gives you more flexibility to optimize for cost, speed, and voice quality based on your use case.

What’s new

• Multiple STT models supported (e.g., real-time and high-accuracy options)

  • gpt-4o-transcribe

  • scribe_v2_realtime

  • saaras:v3

• Multiple TTS models supported with different voice qualities and latencies

  • gpt-4o-mini-tts

  • eleven_flash_v2_5

  • eleven_multilingual_v2

  • eleven_turbo_v2_5

  • bulbul:v3

Why it matters

• Optimize for cost vs performance depending on use case

• Improve voice quality and responsiveness

• Support a wider range of languages and accents

• Greater control over your voice agent experience

Where to find it

Contact your account manager to enable it OR email cs@wotnot.io

You can now hide days with no data from charts in the Analytics reporting dashboard.

With the new “Hide empty data” toggle, charts will exclude dates where no activity was recorded — making reports cleaner and easier to analyze.

This setting is available in any drill-in report that includes a chart. Simply enable the toggle to remove empty days from the view.

Cleaner charts. Better insights. 📊

You’ll now see a configuration under advanced settings for “Enable free text” that lets you enable an input box alongside existing options like buttons or quick replies. When turned on, users can either select from the predefined options or type their own response directly in the chat widget.

This unlocks more natural, two-way conversations by:

• Allowing users to express intent in their own words

• Avoiding forced, linear journeys

• Making interactions feel more flexible and human

For any free text input, you can route the response to any AI agent to handle user queries dynamically.

This update helps you design conversations that guide users—without restricting them.

What is the average length of chats?

On the “conversation duration” metric, under its drill-in view we have introduced a insights tab which shows all the duration of chats handled in pre-defined ranges — <5, 5-10, 10-15, 15-30, >30 mins.

This helps you get better sense of how long agents take to complete a chat.

Time to response trend

We’ve enhanced the Time to Response metric to give you deeper visibility into how quickly your agents reply to visitors. This update introduces two new visualizations that make it easier to spot patterns and track improvements over time.

Response Time Heatmap

The heatmap shows median response time across days of the week and hours of the day. This helps you quickly identify when response times are on track and when they tend to slow down, so you can optimize staffing, automation, or workflows for those specific time windows.

Response Time Trend (Line Graph)

The line graph displays how your average time to response changes over a selected period. It gives you a holistic view of whether response times are improving, declining, or remaining consistent, making it easier to measure the impact of process or team changes over time.

Time to first response

The line graph displays how your average time to first response changes over a selected period. It gives you a holistic view of whether first response times are improving, declining, or remaining consistent, making it easier to measure the impact of process or team changes over time.

SLA breaches

The line graph shows the number of SLA breaches over time, making it easy to spot trends at a glance. You can quickly see whether breaches are increasing, decreasing, or remaining stable across the selected period.

You can now filter bots based on:

• Creation date — quickly view bots created within a specific time range.

• Bot status — see bots that are active, inactive or in draft/sent mode.

• Created by — filter bots based on the user who created them.

• Channel — which channel is the bot built for.

These filters help you stay organized, reduce clutter, and manage bots more efficiently—especially when working with multiple teams and environments.

Why it matters

Audit Logs bring transparency, security, and accountability to your operations. You can:

• Monitor who did what, and when

• Track access and activity for compliance and security reviews

• Debug issues faster by identifying recent configuration changes

Where to find it

Go to Support & Billing → User Management → Audit Log.

How it works

If you’re storing a JSON object in a variable and want to use a particular field from it, follow this simple approach:

• Create a TEXT type variable

• Store the JSON object in that variable

• Type # followed by the variable name and append the field you want to access

• Hit Enter to create the variable chip

Why it matters

This makes working with structured data much simpler inside your flows. You can dynamically pull values from JSON objects and use them across messages, conditions, and actions—saving time and reducing complexity.

Use cases

Ideal for handling API responses, passing data between steps, and building more dynamic, data-driven conversations.

Learn more

Why it matters

This gives partners greater flexibility and control over usage and costs. You can:

• Decide which LLM models customers can access

• Set custom credit multipliers per model

• Better manage margins and optimize AI usage

Who it’s for:

Built specifically for partners and resellers who want fine-grained control over model access and billing.

Why it matters:

This gives you deeper visibility into how users engage with your messages. You can now:

• Update CRM records based on message opens and replies

• Track delivery and engagement patterns

• Trigger smarter marketing and follow-up automations

Use cases:

Perfect for teams looking to build data-driven marketing automation, improve follow-ups, and maintain accurate engagement data across systems.

Learn more

With the new Mobile SDK, you can seamlessly embed AI agents into your iOS and Android applications, enabling in-app support, engagement, and automation — without ever leaving your app.

No redirects. No browser hops. Just smooth, native experiences your users will love.

Learn more

Key Benefits:

• Native experience: Fully integrated, app-like chat interactions.

• Higher engagement: Keep users within your own app ecosystem.

• Developer-friendly: Easy SDK setup with complete control over branding and UI customization.

You can also personalise this message using variables such as {{agent_name}} and contact fields, giving a warm and customised first touch.

WotNot now supports major SSO providers including Auth0, Okta, and Azure Active Directory (Azure AD)

Learn more

This makes it incredibly quick to reuse logic, duplicate flows, and speed up bot building across environments.

This helps improve the conversation experience.

This makes it easier to quickly read, understand, and respond without needing to play the audio every time.

A small change that makes conversations faster, clearer, and more accessible.

This enhancement ensures consistent training times, higher availability, and efficient use of compute resources, so you can deploy updates faster and scale effortlessly across multiple environments.

In addition to seeing past versions, you can open any version in read-only mode and even test it directly — making it easier to compare flow changes, A/B test experiences, and confidently roll back if needed.

A simple way to experiment, learn, and optimize your bots faster.

Key benefits:

• MM Lite API delivers comparable or a greater number of marketing messages than Cloud API

• New features exclusive to MM Lite API and not available on Cloud API

• Easy-to-transition from Cloud API

Steps to implement:

1. Reconnect your WhatsApp Business API

2. Your Meta registered email address will receive an email from Meta

3. Click on the link to acknowledge

4. Start running your campaigns

This includes prompts used across:

• Action Blocks (AI Agent, Answer AI, Set AI)

• Co-pilot on Live Chat

• Rewrite Messages

• Prompt Generator

This update gives you greater flexibility to fine-tune and optimize how your AI agents interact, ensuring a more consistent and personalized experience for your customers.

You can now view detailed event data — including user queries, responses, latency, and AI usage — for every interaction that happens with your bot.

🔍 Where to find it:

• On the Test Bot page → click “View Logs”

• On the Live Chat page → view logs for real-time conversations

Use this to monitor performance, troubleshoot issues, and gain insights into how your bot responds to users.

👉 Learn more

You can now connect your email account and let your AI agents automatically read, process, and respond to incoming emails.

🔧 What’s new:

• Connect your email provider (Gmail, Microsoft, TechAlpha, or custom IMAP/SMTP).

• Use your existing bots to handle inbound emails — just link your bot to a configured email address.

• Automate common email workflows like lead follow-ups, support responses, or FAQs.

👉 Learn more

It’s simple — share your unique referral link, bring in new customers, and get rewarded for every signup.

💰 Payouts: Commissions are paid out in the first week of each month, with a minimum payout of $100.

👉 Sign up here

For any questions or support, please reach out to hello@wotnot.io.

Once stopped, all contacts currently in the queue will no longer receive any future messages.

You can also update or remap these variables as needed before sending the campaign — ensuring the correct data is used in your WhatsApp messages.

With this new configuration, these chats will now be automatically routed to the next available agent as soon as they’re free.

This ensures customers get faster responses while agents handle conversations more efficiently.

Learn more

Now, you can see exactly which rules were applied and how an agent qualified for each assignment—making it easier to trace and resolve queries like “Why wasn’t this chat assigned to Agent A or B?”

Learn more

This upgrade not only improves response speed but also makes conversations feel more natural and real-time.

• Buttons: Provide quick, clickable options alongside the response.

• Carousel: Showcase items visually with images, descriptions, and actions users can scroll through.

• File: Share detailed documents, like PDFs, directly within the chat.

This flexibility transforms static answers into interactive experiences, making it easier for users to explore, decide, and act.

Learn more

This makes conversations more seamless and visually engaging for your users.

Now you can define keywords or phrases like “GET OFFER” or “START” to instantly launch a specific bot you’ve built.

Perfect for promotions, campaigns, or guiding users straight into the right flow.

Learn more

For example, you can set plan_type = Premium when a user selects a subscription plan and use it later for routing or CRM updates.

Learn more - https://help.wotnot.io/build/action-blocks/set-variable

Learn more

Learn more

Learn more

Functions can be used in the following blocks — AI agent, Answer AI.

Users can now speak directly to your AI agent and get responses in a natural, conversational flow. No typing needed — just a simple back-and-forth to get their questions answered.

This creates a faster, more intuitive experience for users who prefer talking over typing.

This ensures agents are always ready to take chats without needing to manually update their status—saving time and reducing missed conversations.

Goto Settings > Live chat settings > Human handover to enable it.

For businesses, this means you can automatically capture and store the location name, address, latitude, and longitude into variables, making it easy to personalize experiences or trigger location-based workflows.

This is what it will look like on WhatsApp👇

This seamless experience makes it faster and easier to create powerful, intelligent bots.

This enhancement allows you to go beyond standard message formats like reply buttons and lists, enabling support for media-rich carousels, image+text combo messages, and more — all within the same conversation flow.

This gives you more flexibility to deliver engaging, visually appealing experiences to your users.

For example, upload a picture of a food item and ask “What are the macros?”, and the AI will return information like calories, protein, carbs, etc.

Great for use cases in health, documentation, product support, and more.

You can also use it to improve and refine existing prompts with one click — ideal for building high-quality conversations quickly.

When a user messages in a foreign language (e.g., Arabic), the agent sees the message translated into their set language (e.g., English). The agent replies in English, and the message is auto-translated back into the user’s language.

This ensures smooth, multilingual support without switching tools or relying on external translation apps.

Languages supported: 79 (See the list)

When enabled, you’ll now see a “New messages” badge with a count—rather than being force-scrolled to the bottom every time a new message arrives.

This gives you more control over the conversation flow and ensures you never miss a message in between.

Better readability. Smoother experience.

Since it’s mostly a one-time setup, this change helps declutter the main nav and keeps all configuration-related options organized in one place—making it easier for you to manage settings without jumping around.

Same functionality. Cleaner navigation. Better experience.

This means your AI agents and chatbots can seamlessly access your availability and book meetings in real-time—no more back-and-forth scheduling.

Whether it’s sales calls, support appointments, or demo bookings, let automation handle it while you stay in control of your calendar.

Learn more

✨ What’s new?

• Sleek UI/UX redesign for a more intuitive interface

• New reports: AI performance & Outbound analytics

• Drill-in capabilities to dig deeper into your data

• Fresh set of KPIs to track what truly matters

This new update enables you to make every conversation count with smarter reporting.

• GPT-4.1

• GPT-4.1-mini

• GPT-4.1-nano

• o4-mini

• o5-mini

• GPT-4.5

• Claude 3.7 Sonnet

• Claude 3.5 Haiku

Start experimenting with these models to tailor your agents more precisely to your needs.

View the total number of campaigns run within a selected date range and assess the impact they’ve had on your business with ease.

• AI credits usage

• Upvotes & downvotes

• Conversation topics

• User sentiment

• Median messages per session

Learn more 👉 https://help.wotnot.io/reporting/ai-report

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

Learn more 👉 https://help.wotnot.io/bot-builder/action-blocks/agent

Check out how it works 👇

You’ll now find it under the Settings screen for easier navigation and centralized control.

Perfect for dynamic AI answers like:

• “How much balance do I have?”

• “List the top debtors in our company”

• “How much stock of Product ABC is available?”

This helps limit unwanted changes made by teammates.

The homescreen now shows templates created by the WotNot team, which can be directly used.

And the function detail screen shows a better segmentation of the requirements i.e. prompt and function handler in accordions.

We are also adding support for HTTP request as a option if you do not want to write a function handler code.

We’ve introduced a new widget style wherein you can include the icon and get together to provide more visiblity to users if an icon is not enough.

Some new features added:

• Expand mode - you can expand the chat widget to see it in a bigger view.

• Email transcripts - you can request for a email transcript of your conversation.

• End conversation - you can request to end the conversation at your will.

• Privacy banner - to meet compliance requirements state your privacy policy.

• Announcement banner - Let users know if there is anything important

• Privacy link - always give visibility to users on your privacy info.

Watch an overview -

You can now send MMS messages to your contacts with the new “Send MMS” action block.

It can be used in the inbound and outbound action blocks.

Learn more

Just open a “Closed” chat and click on the “Edit” icon to enter the value. Once entered, click on the “Tick” icon to save this value to that variable.

It is now a separate config that you can enable on all “Text” and “Collect” based action blocks — helping you offer more context to your users while collecting information.

List of actions that get a UI refresh:

• Action blocks

• Configuration panel of action blocks

• Add action block menu

• Test bot

• Utility tools

Goto Javascript > Enable “Execute asynchronously”.

Additionally, you can initiate a new conversation directly from the contact’s detail page.

Other users will be in read-only mode by default, allowing you to focus on bot creation.

They can request editing access when needed - to which editor can accept/reject.

Learn more

• exclude urls - define regex to exclude certain urls

• wait time - delay in crawling until your page loads.

• wait CSS - delay in crawling until a CSS selector is loaded.

• Execute Javascripts - to prevent any unwanted popups to be crawled instead of the actual webpage content.

Learn more

Why is it helpful?

• Helps you connect to channels, WotNot does not support yet

• Use your own chat widget UI, and WotNot for the bot logic

Learn more

With our new update, you can now see the real-time status of agents directly in the assignee list. This ensures that chats are transferred only to available agents, improving response times and ensuring a seamless experience for your users.

To access it:

Navigate to “Contacts” > Click on any contact > The Contact Detail Page opens.

Here you see details like:

• Attributes of the contact (like email, phone etc.)

• Past chats

Using these APIs, you can setup your live chat system such that WotNot’s chatbots do most of the talking; and when there is a need to transfer to an agent, the agents continue taking the chat from your existing live chat system.

APIs available are:

• to create a conversation

• to close a conversation

• to send an agent response

Learn more about these APIs

1. Create Tickets in Zendesk

   Automatically generate Zendesk tickets directly from your WotNot chatbot, ensuring no customer query goes unnoticed.

2. Fetch Ticket Status

   Keep your customers informed by fetching and displaying real-time ticket statuses directly within the chatbot.

3. Enable Live Chat via Zendesk

   Elevate customer interactions by routing chats from WotNot to Zendesk agents, enabling live support when needed through Zendesk comments.

This integration empowers your team to provide quicker resolutions and enhanced support, all while improving efficiency.

To learn more, visit: https://help.wotnot.io/integrations/ready-integrations/zendesk

Since then, its use cases have expanded significantly. We are now transforming it into a versatile feature that allows users to:

• Create contacts

• Edit contacts

This becomes helpful when your agents more often using the mobile app and require the same set of functions as available on the web app.

You can download the app at:

• iOS

• Android

Using the trigger action block, you can define which bot is to be shown on which website.

Note: The chat widget UI will remain the same on all the website the code snippet is used on.

This will allow for a engaging conversation between your agents and visitors/customers.

The count is fetched in real-time, so try filtering something and see the count.

This helps you block suspicious users from misusing the chatbots.

Learn more

This update allows you to tailor roles with specific permission sets, ensuring that every team member has access to the tools they need—nothing more, nothing less.

Custom Roles

• Easily create roles tailored to your organization’s unique needs.

• Name roles for clarity, e.g., “Marketing Analyst,” “Support Specialist,” or “Junior Admin.”

Granular Permission Sets

• Assign specific permissions to roles based on platform features, such as:

  • Access to analytics dashboards

  • Managing chatbot workflows

  • Grant view-only, edit, or admin rights for each feature.

Learn more

This functionality tracks events by firing the Google Analytics API when used on your chatbot flow, instead of the client-side Javascript fired.

With this support, non-web channels can make use of server side tracking.

Read help doc