Back to vendors
W

WotNot

Also known as: WotNot Inc, WotNot Solutions

Visit site
Entry price$29 per monthFull pricing detail

WotNot combines a visual conversation builder, knowledge grounded AI agents and human live chat. Teams connect customer support and booking workflows to external APIs, choose models from several providers and deploy the same conversation design across messaging channels. Agent paths handle branching and escalation, while execution logs expose API attempts and errors. Plans start with basic chat automation; function calling and live chat require Premium. Enterprise adds identity integration, custom models and private deployment options.

WotNot Inc operates a customer conversation platform that combines structured chatbot flows, AI agents and human support. Its terms also name WotNot Solutions Private Limited. Customers use a visual builder to define a conversation and place an AI Agent block within it. That block answers from business knowledge, invokes functions and selects an exit path when the visitor needs another process. The platform serves website and messaging conversations, including WhatsApp, Facebook Messenger, Instagram and SMS, with email, voice and API channels also offered. Lead collection and appointment booking use the same builder alongside customer support.

An agent's functions connect natural language requests to external APIs or JavaScript handlers. The HTTP block supports authenticated GET, POST and PUT calls, maps returned fields into variables and can use those results in subsequent steps. Native integrations include Zendesk and Freshdesk for support tickets, Google Calendar and Calendly for scheduling, and Salesforce, HubSpot and Zoho CRM for customer records. Teams can begin with support, booking, lead generation or property templates.

Knowledge bases import website pages, documents and text, with recurring refreshes for domains. Conversation variables reset when a chat ends; contact fields retain customer details across visits. Staff can join a conversation, receive a handoff or use a copilot draft before replying. Functions the agent calls run as soon as the flow reaches them, with no approval step for staff.

Lite costs $29 per month with 1,000 chats, 1,000 AI credits and one user. Starter costs $99 with 5,000 chats and credits, unlimited users and AI Studio. Premium costs $299 with 10,000 chats and credits, live chat, function calling and HTTP requests. Annual billing carries a 20% discount, and a 14 day trial is available. AI charges depend on the selected model and the combined input and output tokens; extra credits cost $5 per 1,000. Channel provider charges can add to the subscription.

Enterprise uses a quote for custom integrations, SAML identity integration, custom models and deployment requirements. WotNot holds its own SOC 2 Type 2 report and ISO 27001 certification and offers public cloud, private cloud and on premises installation. The managed service runs on Oracle Cloud Infrastructure, with data centers named in the United States, Europe, Abu Dhabi and the United Kingdom and stored data placed in Chicago in the encryption section of the same security page.

The product fits support teams that want deterministic process steps around conversational AI. Customers select OpenAI, Anthropic or Google Gemini models and tune response limits and temperature. HTTP retry logs expose requests, responses, failures and timestamps, while conversation logs show model and timing details. Preview testing can begin at a chosen block, but teams drive the test conversations manually. Developers can run deployed bot logic through authenticated conversation APIs. Website crawling supplies knowledge, and the agent has no browser or desktop control of its own.

Vendor details

Canonical URL

https://wotnot.io

Category

Customer support agent

Subcategory

Visual support agents with API functions and live chat

Funding status

Operated by WotNot Inc; its terms also name WotNot Solutions Private Limited.

Company status

independent

Use cases & customers

Primary use cases

Answering customer questions from company knowledgeCreating support tickets and retrieving statusBooking meetings and collecting customer detailsRouting conversations between AI and staff

Target customers

Support teams building multichannel automationBusinesses automating booking and lead captureEnterprises requiring private deployment

Deployment options

Public cloud on Oracle Cloud InfrastructurePrivate cloud deploymentOn premises installation on Enterprise

Integrations

Native integrations include Zendesk, Freshdesk, Google Calendar, Calendly, Salesforce, HubSpot, Zoho CRM, Slack and Zapier. HTTP requests and JavaScript handlers extend agent functions. Conversation APIs support custom interfaces.

In practice

A visitor asks about an order; an AI function calls an authenticated service and stores its result for the reply.

A booking path collects the required name and email before connecting to a calendar.

A support agent inspects a failed HTTP attempt and tests the changed block before deploying the revised flow.

Agentic Index coverage score

11.5 / 14 capabilities · 82%

Integrations & Tool Calling Full

The HTTP request block calls external REST services with GET, POST and PUT methods, authentication headers and JSON or mapped form bodies. Response fields can be saved into flow variables for later steps. AI functions choose an API or JavaScript handler from its description and collect missing parameters before execution. Native connectors also create helpdesk tickets and calendar bookings. Function calling and HTTP requests require Premium or Enterprise.

Sourcehelp.wotnot.io/build/integrations/http-request-apiread 2026-10-20

Workflow Orchestration Full

The visual builder combines an AI agent with action blocks and intent based exit paths. An agent normally loops inside its block until a configured condition selects another path; Execute Only Once instead continues immediately to the next step. A demo booking path can require name and email before entering the calendar flow, while a request for staff routes to live chat. Functions supply live data or external changes within that conversation process.

Sourcehelp.wotnot.io/build/action-blocks/agentread 2026-10-20

Knowledge Grounding & RAG Full

Knowledge bases combine website domains, sitemaps, selected URLs, uploaded files and pasted text. Domain imports support a refresh frequency that applies to child pages, with individual overrides available. Documents are divided into retrievable chunks supplied to the selected model when answering. Supported uploads include CSV, spreadsheets, text, PDF and Word files up to 5 MB. Operators can inspect source details and remove irrelevant pages from the maintained collection.

Sourcehelp.wotnot.io/build/knowledge-baseread 2026-10-20

Human Oversight & Guardrails Partial

Staff can inspect an active chatbot conversation and use Join conversation to take control and reply. Agent paths also route requests for a person into live chat, carrying the conversation to the support team. The Premium copilot drafts suggested responses for staff to edit and send. Oversight centers on transfer, manual takeover and response assistance; external function calls do not pass through a dedicated approve or reject queue before execution.

Sourcehelp.wotnot.io/monitor/live-chatread 2026-10-20

Security, Identity & Governance Full

Enterprise customers can use SAML 2.0 single sign on, and every account supports two factor authentication. Secret variables hold API credentials encrypted at rest and hide saved values from teammates, administrators included.

WotNot holds its own SOC 2 Type 2 report and ISO 27001 certification, separate from the accreditations of the Oracle data centers it runs on, and shares SOC 2 audit summaries on request.

Regular penetration tests by certified testers back the program, with findings summaries offered to Enterprise customers, alongside TLS transport, separate credential storage and handling of deletion requests.

Sourcewotnot.io/securityread 2026-10-20

Observability & Auditability Full

HTTP request logs record each attempt with its status, retry number, request details, response or error and execution timestamp. Operators open these logs from the HTTP Request integration to inspect calls made by their flows. Test Bot and Live Chat also expose chronological conversation logs with model details, latency and message flow. The function test view shows detected arguments and outputs, making failed tool execution inspectable alongside the conversation.

Sourcehelp.wotnot.io/build/integrations/http-request-apiread 2026-10-20

Memory & State Persistence Partial

Conversation variables hold selections, intermediate results and other temporary values until the session ends, when they reset. Contact variables preserve names, email addresses, preferences and lead status across conversations with the same visitor, always holding the latest value written. Variables can be archived, and a chat history variable carries the current transcript. What persists between visits is the contact record itself; the agent keeps no separate store of what it learned about a customer, with retention or deletion controls of its own.

Sourcehelp.wotnot.io/build/interface/variablesread 2026-10-20

Deployment & Data Residency Full

Customers can run the platform in WotNot's public cloud, in a private cloud or on their own premises, with on premises installation and a review of the deployment included in Enterprise. The managed service runs on Oracle Cloud Infrastructure, and the security page names data centers in the United States, Europe, Abu Dhabi and the United Kingdom, while its encryption section places stored data in Chicago. Static IP addresses let customers allowlist the service in their own firewalls.

Sourcewotnot.io/pricingread 2026-10-20

Prebuilt Agents / Templates / Packs Full

The builder offers editable templates for customer support, lead generation, appointment booking and property showcasing. Customers select a template by use case or industry and adapt its conversation blocks. The appointment template supplies date and timeslot selection, contact collection and confirmation. These are starting conversation flows; teams add their own knowledge sources, agent blocks and connected actions when they want open ended AI responses and business specific execution.

Sourcewotnot.io/blog/how-to-create-a-chatbotread 2026-10-20

Triggers & Channel Coverage Full

Inbound customer messages start the configured conversation flow through its Trigger block. Web deployments can initialize contact and conversation variables from query parameters or a JSON session payload, while messaging channels supply the incoming response. Outbound bots extend the same builder to proactive conversations. Customers select the channel and trigger conditions when creating the bot, and can also start a conversation from their own application through the API.

Sourcehelp.wotnot.io/build/action-blocks/triggerread 2026-10-20

Model Flexibility & Routing Full

The agent configuration lets customers select models from OpenAI, Anthropic and Google Gemini, with custom models available on Enterprise. Maximum response tokens and temperature can be set for the chosen agent. Billing varies by the selected model's credit rate and by input and output token volume. Function tests also permit a model selection, so teams can inspect responses from the configuration they intend to use before deploying the flow.

Sourcehelp.wotnot.io/build/action-blocks/agentread 2026-10-20

APIs / SDKs / MCP Extensibility Full

Developers can publish a bot on the API channel and run its conversation logic in a custom application. Authenticated REST endpoints start a conversation and submit subsequent visitor messages using bearer tokens. The start request accepts a bot key, an external visitor identifier and contact, system and conversation variables. Message contracts cover text, button selections, files and JavaScript results, allowing an external interface to advance the same deployed flow.

Sourcehelp.wotnot.io/deploy/publishing-agents/apiread 2026-10-20

Testing, Debugging & Optimization Partial

Test Bot opens the current conversation flow in a preview drawer, and a block level Test control starts from a selected point so builders can exercise changed steps. Functions have a Save and Run test view with detected arguments, generated output and a model selector. HTTP requests can also be tested before use. Testing remains a manually driven sequence of queries and inspections, without a stored test suite that replays and scores customer scenarios automatically.

Sourcehelp.wotnot.io/deploy/testing-the-chatbotread 2026-10-20

Browser / Computer-use Not documented

Website crawling prepares source material for the agent's knowledge base. Its settings can wait for page load, wait for a CSS selector or execute supplied JavaScript to avoid indexing unwanted popups. The support agent itself works through conversation blocks, API requests and connected integrations. An embedded chat widget provides its user interface; it does not give the agent a browser session for navigating websites, clicking through applications or controlling a desktop.

Sourcehelp.wotnot.io/build/knowledge-baseread 2026-10-20

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Pricing

$29 per month

Subscription with separate chat and token based AI credit allowances

Trial available

Included quota

Lite: 1,000 chats, 1,000 AI credits, 1,000 data sources and one user.

What is public

Lite $29, Starter $99, Premium $299 per month, with plan quotas and a 20% annual discount.

Billing mechanics

AI credits follow model-specific rates per 1,000 combined input/output tokens, in addition to the chat allowance.

Cost watchouts

Functions, HTTP requests and live chat require Premium. External channel charges and extra AI use add cost.

Variable cost rationale

Model and context length change credit use, and messaging or voice providers can charge separately.

Additional watchouts

A conversation may generate multiple responses and each response consumes tokens; included credits are not a guaranteed number of resolved tickets.

Overage / add-ons

Additional AI credits cost $5 per 1,000 and expire at the end of the subscription cycle.

Sales call required

Mixed (some tiers require a call)

Free / trial

14 day free trial

Lowest paid plan

Lite

Commercial notes

Enterprise uses custom pricing and adds identity, custom integrations and deployment options.

Key ambiguities

The exact combined bill depends on channel provider fees and the model and token volume of each response.

Cancellation / refund

Monthly or annual subscriptions renew automatically. Renewal can be canceled through account management or support. Refunds are discretionary.

Support SLA / resale

Enterprise includes priority support, professional services and InfoSec/legal review; white label arrangements are offered separately.

Missing data

A workload-specific quote for external channel costs and private deployment.

Agentic Index verified 2026-10-20

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.