Whismer

Whismer provides a visual node-based conversation designer that allows non-technical users to construct multi-turn dialogue flows without writing code. The builder uses a canvas-based UI where users connect decision nodes, response blocks, and action triggers to define chatbot behavior. This approach abstracts away programming logic into intuitive visual blocks representing questions, branching logic, and responses, enabling rapid prototyping of customer service workflows.

Whismer uses keyword and pattern-matching logic to classify user inputs and route them to appropriate responses, rather than leveraging neural language models. The system matches incoming messages against predefined keywords, phrases, or regex patterns to determine intent, then returns corresponding responses from a curated knowledge base. This rule-based approach is lightweight and deterministic but lacks the contextual understanding of modern NLP systems.

Whismer provides a theming engine that allows users to customize the chatbot's appearance to match their brand identity through a visual editor. Users can modify colors, fonts, button styles, chat bubble appearance, and widget positioning without touching CSS or code. The customization is applied via a configuration layer that generates inline styles and CSS classes, ensuring the chatbot visually integrates with the host website.

Whismer generates a single JavaScript snippet that users can paste into their website's HTML to deploy the chatbot widget. The snippet handles script loading, widget initialization, and communication with Whismer's backend servers. This approach abstracts away the complexity of managing dependencies, API authentication, and cross-origin communication, allowing non-technical users to deploy a fully functional chatbot in seconds.

Whismer stores and retrieves conversation transcripts for each user, allowing businesses to review past interactions and maintain conversation context across sessions. The system persists messages in a database indexed by user identifier and timestamp, enabling retrieval of full conversation histories through the dashboard. This enables customer service teams to understand customer issues over time and provide continuity in support.

Whismer supports outbound webhooks that allow the chatbot to trigger external actions by sending HTTP POST requests to user-specified endpoints. When a conversation reaches a specific point or user selects an action, Whismer sends structured JSON payloads containing conversation context to configured webhook URLs. This enables integration with external systems like CRMs, ticketing platforms, or custom backend services without requiring Whismer to maintain native integrations.

Whismer offers a free tier that allows users to build and deploy a functional chatbot with limitations on monthly conversation volume and feature access. The freemium model uses a quota-based system where free users receive a monthly allowance of conversations (e.g., 100-500 per month), with paid tiers offering higher limits. This approach enables non-technical users to test the platform and validate chatbot concepts before committing to paid plans.

Whismer provides a mechanism to escalate conversations from the chatbot to human agents when the chatbot cannot resolve a customer issue. The escalation workflow captures the conversation context, customer information, and unresolved query, then routes the conversation to an available agent through an integrated queue or external ticketing system. This enables a hybrid support model where the chatbot handles routine inquiries and humans handle complex issues.

Whismer provides a dashboard that displays key metrics about chatbot performance, including total conversations, resolution rates, average response times, and user satisfaction indicators. The analytics are aggregated from conversation logs and presented through charts and summary statistics. This enables business owners to understand chatbot effectiveness and identify areas for improvement without requiring data analysis expertise.