Metric and Dimension Definitions

Prev Next

Overview

This page provides definitions for all metrics and dimensions used in Mapp Fashion reporting. Use it as a reference to understand how your data is calculated and how to interpret performance across different parts of your site.


Attributed Revenue — Total, Direct, and Indirect

Attributed Revenue is defined as the total order revenue — including tax and shipping when available — for orders placed within the selected date range, if any item in the order was purchased within 24 hours of a Mapp Fashion widget interaction.

Types of Attributed Revenue

  • Direct Attributed Revenue: Revenue from purchases of items that were clicked in a Mapp Fashion widget.

  • Indirect Attributed Revenue: Revenue from purchases of the PDP item when an Outfits widget was shown on that PDP. This only applies to the outfits widget.

Widget Type Constraints

  • Similar Items widgets can generate only direct attributed revenue.

  • Outfits widgets can generate both direct and indirect attributed revenue.

Order-Level Counting Rules

  • An order may qualify for both direct and indirect revenue.

  • For each order:

    • Revenue can be counted once toward direct revenue

    • And once toward indirect revenue

  • However, total attributed revenue does NOT double-count.

  • Even if multiple items in one order qualify, the order's revenue is included once in each relevant category.

Date Filtering

  • Based on order date, not interaction date.

Splitting Behavior

Splits change which dimension the order is associated with. Because full order revenue is assigned to each relevant segment, totals across segments can exceed the overall metric.

Device Class + Platform Split

  • Uses the Device Class + Platform where the transaction occurred.

  • A purchase may occur on a different device/platform than the widget interaction.

  • Each order is associated with one and only one Device Class + Platform, so no double‑counting occurs across platforms.

Department Split

  • Uses the Department of the first seed item used to generate the widget recommendations.

    • Usually the PDP item.

  • If multiple seed items are involved (rare but possible), each Department may receive the full order revenue.

  • When split by Department, the sum across all Departments can exceed total attributed revenue.

  • Assumes each product has a single Department.

Widget Type Split

  • Uses the type of widget that generated the attributed revenue.

  • If different items in the same order came from different widget types, each widget type receives the full order revenue.

  • Totals across widget types may exceed total attributed revenue.

Placement Split

  • Uses the location of the widget (e.g., PDP, Basket).

  • As above, full order revenue is assigned to each Placement that contributed.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Attributed Revenue % of Total Site Revenue (Direct / Indirect / Total)

Defined as:

Attributed Revenue ÷ Total Revenue

Uses identical splitting logic to Attributed Revenue.

Available Splits

Date, Device Class + Platform


Average Order Value (AOV)

Total Revenue ÷ Orders

Department Split Behavior

  • Uses Department(s) of items in the order.

  • Because full revenue goes to each Department represented in the order, sums across Departments may not match the total.

Available Splits

Date, Device Class + Platform, Department


Canonical User ID

Internal Mapp Fashion identifier that unifies user activity across devices.

Purpose

  • Enables cross‑device attribution.

  • Combines data Mapp Fashion believes belongs to the same user.

Available Splits

None


Click Through Rate (CTR)

Recommendation Clicks ÷ Recommendation Served Impressions

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Clickers

Distinct Canonical User IDs who clicked an item in a Mapp Fashion widget.

Date Filtering Behavior

  • Based on interaction date.

  • A user clicking across several days counts as one clicker per day, not cumulatively.

Split Behavior

Users can appear in multiple segments:

  • Device Class + Platform: where the click occurred.

  • Department: Department of the seed item of the widget that was clicked.

Therefore, totals across segments may exceed overall totals.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Conversion Rate

Converters ÷ Interactors

Interactors Definition

  • Similar Items widget → users who click an item.

  • Outfits widget → users who perform any interaction (clicks, hovers, etc.).

Converters Definition

Users who made any purchase within 24 hours of interacting with a widget.

Date Filtering

  • Based on interaction date.

Split Behavior

  • Users may fall into multiple categories when split by Device, Department, Widget Type, or Placement.

  • Therefore, totals across segments may exceed the overall totals.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Device Class + Platform

Definitions:

  • Mobile app → represented as mobile app

  • Otherwise → Desktop, Smartphone, Other

Splits: None


Interaction Rate

Recommendation Interactions ÷ Recommendation Served Impressions

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Converters

Distinct Canonical User IDs who made a purchase within 24 hours of interacting with a widget.

Date Filtering

  • Uses interaction date.

  • A user converting on multiple days is counted once per day.

Split Behavior

Users may appear in multiple Device Classes, Departments, Widget Types, or Placements.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Interactors

Distinct Canonical User IDs who interacted with a widget.

Date Filtering

  • Based on interaction date.

  • Counted once per day.

Split Behavior

Users may appear in multiple Device Classes, Departments, Widget Types, Placements.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Item Views

Count of item page view events.

Available Splits

Date, Device Class + Platform, Department


Orders

Count of distinct order IDs.

Split Behavior

  • When split by Department, each order is associated with the Department(s) of all items in the order.

  • Thus, totals across Departments may exceed the overall count.

Available Splits

Date, Device Class + Platform, Department


Placement

Location where a widget appears (e.g., PDP, Basket, Email).

Splits: None


Purchasers

Distinct Canonical User IDs who have placed an order.

Date Filtering

  • Based on the order date.

  • Counted once per day.

Split Behavior

Users may appear across multiple Device Classes or Departments.

Available Splits

Date, Device Class + Platform, Department


Recommendation Clicks

Count of recommendation click events.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Recommendation Interactions

Count of interaction events with recommendations.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Recommendation Served Impressions

Count of recommendation served impression requests.

Available Splits

Date, Device Class + Platform, Department, Widget Type, Placement


Revenue Per Visitor (RPV)

Total Revenue ÷ Visitors

Department Split Behavior

  • Full revenue from an order is assigned to all Departments represented by purchased items.

  • Totals across Departments may exceed overall totals.

Available Splits

Date, Device Class + Platform, Department


Session

All activity from a Canonical User ID on a specific Device Class + Platform.

Rules

  • Session ends after 30 minutes of inactivity.

  • Activity includes any page views.

Splits

None


Total Revenue

Total site revenue including tax and shipping.

Device Class + Platform Split

  • Uses where the transaction occurred.

  • Each order belongs to one and only one device/platform.

Department Split

  • Based on Departments of purchased items.

  • Full revenue is assigned to each Department represented.

  • Totals across Departments may exceed overall totals.

Available Splits

Date, Device Class + Platform, Department


Visitors

Distinct Canonical User IDs who viewed a tracked page.

Date Filtering

  • Based on page‑load date.

  • Counted once per day.

Split Behavior

Users may appear in multiple Device Classes or Departments.

Available Splits

Date, Device Class + Platform, Department