FE&R Glossary

This glossary provides clear definitions of the key terms used throughout the AB Tasty Feature Experimentation & Rollouts (FE&R) platform.

Whether you're a developer, product manager, marketer, or QA, understanding this terminology is essential for effectively managing flags, targeting users, analyzing experiments, and controlling feature releases.

The terms below cover core concepts such as feature flags, targeting logic, experiment analysis, rollout strategies, and platform components. Use this glossary as a reference to align your team’s understanding and ensure consistent usage across documentation, code, and campaigns.

Flag	A feature toggle used to control the activation or deactivation of a feature in real time without deploying new code.
Feature	A product capability or function that can be turned on or off using a flag.
Variation	One of the versions of content or functionality shown to users in an experiment.
Targeting	The process of selecting which users or segments will see a variation or feature based on defined conditions.
Context key	A key-value pair used to define targeting conditions or segmentation for users, such as location, device type, or subscription level.
User ID	A unique identifier assigned to a user, often coming from your application’s authentication system.
Feature Flag	A control mechanism used to enable or disable a specific feature dynamically without code deployment.
Flag Key	A unique identifier for a flag used programmatically within SDKs or API calls to reference the correct feature.
Campaign	A container for organizing experiments or personalizations, typically with targeting rules, metrics, and variations.
Scenario	A predefined testing or personalization use case that defines the logic and configuration for feature delivery or experimentation.
Server-side A/B tests	A/B tests executed entirely on the server before content is delivered to the client, used to test backend logic, APIs, or secure flows.
Server-side Personalization	The process of tailoring user experiences based on data and logic handled on the server, rather than in the browser.
Rollout	A progressive deployment strategy where a new feature or change is gradually exposed to users over time, often using flags.
Release management	The practice of using feature flags to manage the lifecycle of features, including phased rollouts, rollbacks, and testing.
Rollback	The action of deactivating or reverting a feature or variation that has caused issues.
Automatic rollback	A mechanism that disables a feature or variation automatically when a predefined failure condition is met (e.g. drop in performance or KPIs).
Rollback KPI	A performance metric used to automatically trigger a rollback if its value drops below a certain threshold.
Feature management	The practice of controlling the lifecycle of features through flags, rollouts, targeting, and analytics.
Panic mode	A safety switch that deactivates all active flags and features immediately, typically used during critical issues.
Conversion rate	The percentage of users who complete a desired action (e.g. purchase, sign-up) out of the total users exposed to a variation.
Chance to win	A statistical probability indicating the likelihood that a variation will outperform others based on current test data.
Campaign report	A summary of a campaign's performance, including variation metrics, conversions, and statistical insights.
Dynamic allocation	A system that automatically adjusts traffic distribution among variations based on their performance in real time.
Traffic allocation	The percentage of users assigned to each variation or experience in an A/B test or personalization.
Untracked traffic	The portion of user traffic that is excluded from analytics or experimentation metrics, either by configuration or due to consent settings.
Decision API	An API endpoint that determines which variation or feature should be shown to a user, based on targeting and context.
SDK (Software Development Kit)	A package provided by AB Tasty to integrate feature flags, personalization, and experimentation into your app.
Custom ID	A user-defined identifier used to track a visitor, often mapped from your backend systems (e.g. customer ID).
Slug	A human-readable, URL-safe identifier used to reference the custom ID
Dashboard	The main interface where you monitor, configure, and analyze campaigns, flags, and metrics.
Environment ID	A unique identifier representing a specific environment (e.g. production, staging) within your AB Tasty setup.
Folder	A way to organize flags, campaigns, and scenarios within the platform to reflect logical groupings such as product areas or teams.