Kameleoon

Kameleoon is an experimentation and feature management platform that combines AI-driven test creation with robust statistical analysis and feature flagging. The platform's flagship capability, Prompt-Based Experimentation (PBX), lets marketing, product, and engineering teams describe changes in natural language and have an AI agent generate, configure, and launch experiments without manual coding. According to the company, Kameleoon has been operating for 13 years and serves over 1,000 brands.

What It Is

Kameleoon is a full-cycle experimentation platform covering A/B testing, multivariate experiments, feature flags, personalization, and product recommendations. Its core differentiator is PBX — a connected system of AI agents that handles ideation, variant building, audience targeting, QA, analysis, and shipping of winning tests. The platform targets marketing teams, product managers, engineers, and data scientists across industries including e-commerce, healthcare, financial services, travel, media, and automotive.

How PBX Works

PBX (Prompt-Based Experimentation) is Kameleoon's agentic AI layer built on top of the core experimentation engine. Users enter a URL and a natural-language prompt describing what they want to test; the AI agent then:

• Scans the page for friction points using Predictive Impact Scoring trained on historical experiment data
• Generates front-end and back-end variant code without requiring a visual editor or manual coding
• Configures audience targeting, traffic allocation, and success metrics automatically
• Runs a QA Agent that detects failures and routes them back for automated repair
• Analyzes results and surfaces next-step recommendations

PBX 2.0, released in 2026, added the ability to ship winning client-side tests server-side via an MCP integration, a Figma integration that converts winning variants into reusable design files, and a Master Prompt feature for aligning variations across teams.

Statistical Rigor and Data Architecture

Kameleoon's stats engine supports sequential testing, frequentist and Bayesian methods, CUPED variance reduction, Multiple Testing Correction, Sample Ratio Mismatch (SRM) detection, Contextual Bandit and Multi-Armed Bandit testing, and custom attribution windows. The platform's architecture is described as 99% local storage-based with CDN latency under 50 ms globally, a 28.4 KB cached async snippet, and real-time streaming for instant feature flag updates. It bypasses Safari ITP and ad blockers for full traffic visibility and includes automatic rollback based on KPI thresholds.

Platform and Integration Breadth

Kameleoon supports client-side, server-side, and mobile app testing, as well as Single Page Applications built with React, Next.js, and similar frameworks. Integration capabilities include:

• Two-way sync with data warehouses (Snowflake, BigQuery, Redshift, Databricks) for audience targeting and metrics ingestion
• Native CDP connections for advanced targeting
• Shopify extension, Microsoft Visual Studio extension, and a Chrome debug assistant extension
• JavaScript, Data, and Automation APIs
• OpenFeature SDK support
• Figma integration for design-to-test workflows

Update: PBX 2.0 and Recent Releases

The product roadmap page notes PBX 2.0 as the current major release, described as enabling teams to "Ideate, Build, Configure, Analyze AND Ship winning client-side tests server-side in minutes with an MCP." The changelog shows over 100 new product releases per year. Notable recent milestones include: a new all-in-one dashboard homepage (June 2026), Figma integration for PBX, Master Prompts for cross-team alignment, PBX Ideate for AI-generated test ideas, Contextual Bandit testing (June 2025), Real-Time Streaming Architecture (May 2025), and a Sense Analyst (ContentSquare) x PBX agentic AI workflow integration (October 2025).

Who It Is Built For

Kameleoon positions itself for all experimentation roles without requiring developer involvement for every test. Marketing teams can launch banners, pop-ups, and personalization campaigns without engineering support. Product teams can validate features before full development and measure adoption impact. Engineering teams get feature flags, safe release tooling, automatic flag cleanup for technical debt management, and environment configuration. The platform is ISO 27001 and SOC2 certified, supports GDPR, CCPA, and HIPAA compliance, and offers private cloud deployment with end-to-end encryption and SSO.