> For the complete documentation index, see [llms.txt](https://partner-docs.covergenius.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://partner-docs.covergenius.com/optimizations/glossary.md). # Glossary ## Concepts and terminology

Term	Definition
Control	The unchanged version in an Exploration. Serves as the baseline for all variant comparisons.
Bayesian probability	The statistical method used to calculate Chance to beat control. Asks 'how likely is this variant to be genuinely better?' rather than relying on p-values.
Chance to beat control (P(beat))	The Bayesian probability that a variant outperforms the control. Winning ≥ 95%, Losing ≤ 5%, Inconclusive between 5% and 95%.
Dimension Analysis	A breakdown of Exploration or Experiment results by user segment (e.g. region or device type).
Duration	How long an Exploration or Experiment has been running.
Experiment	A randomized controlled A/B test used to validate whether a specific product change genuinely outperforms the control.
Exploration	A multi-variant price discovery test that distributes traffic across a range of price points to identify patterns and opportunities.
Hard Gate	A set of automated checks a recommendation set must pass before being shown to a partner — covering sample size, data quality, operational risk, and business value. Configured per partner and per Exploration.
Inconclusive	A variant with a Chance to beat control between 5% and 95%. More data is needed before a decision can be made.
Leading variant	The variant currently showing the best performance relative to the control.
Losing	A variant with ≤ 5% Chance to beat control.
Optimizations	The section in your Portal that contains Explorations and Experiments.
Pooled	Combined across all recommendation sets shown together, rather than a single recommendation set on its own. Used to distinguish summary figures (e.g., Pooled projected GWP uplift) from the per-recommendation-set figures in the table below.
Primary metric	The key measure of success (typically GWP per order). Experiments are anchored to this metric.
Projected GWP uplift	An Exploration-stage estimate of a recommendation set's annual GWP opportunity. Unlike Projected uplift (Experiments), this figure has not been validated by a controlled test.
Projected uplift	Estimated monthly business impact of shipping the winning variant, calculated from the variant's lift and current traffic volume. Directional guidance only.
Recommendation set	A segment paired with a proposed treatment, surfaced during an Exploration as a candidate worth validating with an Experiment. Recommendation sets are directional estimates, not confirmed results.
Related Experiments	Experiments derived from an Exploration. Shown in the Exploration detailed view.
Sample	The number of unique users enrolled in an Experiment.
Scheduled	An Exploration or Experiment that is configured but has not yet started collecting data.
Segment	A specific combination of attributes that groups similar customers together (e.g., a trip duration range combined with a price range). Used in Dimension Analysis and in recommendation sets.
`subject_id`	The partner-supplied identifier in the XCover API request used to assign users consistently to experiment variants.
Traffic %	In Explorations, the share of enrolled users assigned to a given price variant.
Treatment	The specific product change proposed for a segment within a recommendation set. Once validated through an Experiment, a treatment becomes a variant.
Uplift range	The 95% Bayesian credible interval for the relative improvement or decline of a variant. A range that doesn't cross zero indicates a clear directional effect.
Variant	A modified version that is tested against the Control — a price adjustment, a different configuration, or an alternative feature or design. Explorations compare several variants for directional patterns; Experiments test them to identify a winner.
Winning	A variant with ≥ 95% Chance to beat control.

## Metrics

Metric	Definition
Avg. protection price	The average price paid by the customer for a protection quote package.
Core order conversion	The percentage of visitors who successfully completed a primary order.
Gross Written Premium (GWP)	The total gross sales value of all protection products sold.
GWP % of order	The total value of protection products sold expressed as a percentage of total order value.
GWP per order	The average protection premium generated for each of your successful orders.
GWP per visit	The average protection product sales value generated for every unique visitor session (OfferId or QuotePackageId).
Premium vs. order ratio	The cost of protection expressed as a percentage of the item's price (e.g. 10% of the flight cost).
Protection attach rate	The percentage of successful partner orders that included protection.
Protection conversion	The percentage of protection quote packages or OfferIds that resulted in confirmed purchases.
Total bookings sold	The total number of quote packages or OfferIds purchased by your customers.
Total order value	The total sales value of your products for customers who also completed a purchase in the checkout flow.
Total quotes displayed	The total number of distinct protection quote packages generated.
Unique visits	The total number of unique quote packages or OfferIds exposed to the experiment variants.

--- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://partner-docs.covergenius.com/optimizations/glossary.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.