Overview

Prescient’s Performance Measurement features give you a unified view of your marketing performance across all channels and campaigns. In this walkthrough, you’ll see:

Homepage Overview

Base vs. Halo Effect Rollup
MMM Fit
Trends
KPI Report

Performance Page Overview

Hero Metrics
Channel, Tactic, & Campaign Views
Insight Drawers
Saved Views, Columns, Filtering, & Exporting

For a deeper understanding of how Prescient’s MMM integrates with these features, refer to the Modeled Attribution article.

Homepage Overview

The homepage serves as the central hub for your account’s most important performance metrics, offering a snapshot of your marketing efforts across all connected data channels. Key components include:

📘
Info
Homepage visualizations are dependent on what revenue sources you have connected, and Amazon based views may only be available if you have an amazon storefront connected. Other views may be edited or reformatted if amazon is not available.

Base vs Halo Effect Rollup

At the top of your homepage, you’ll find the Base vs Halo Effect Rollup. This feature breaks down your base and halo effect revenue for each individual ad channel, helping you understand how channels drive both direct conversions (base) and indirect conversions (halo). A date selector lets you define the time window you’d like to analyze.

Depending on whether you have one or multiple storefronts (e.g., Shopify + Amazon), you’ll see different tabs within this rollup:

Shopify & Amazon Base vs. Halo Mix: Displays the distribution of base and halo effect value per channel, across each storefront.
Overall Base vs. Halo Mix: Shows the distribution of base and halo effect value per channel, summed across all storefronts. This helps you understand which channels lean higher or lower in the funnel—for example, Google and Bing often drive lower-funnel conversions, Facebook drives both awareness and conversions, and CTV channels may be purely top-of-funnel plays.
Base Only: Provides the split of base revenue across your storefronts.
Halo Effect Only: Focuses solely on revenue from halo effects per channel and identifies where those channels drive the most indirect traffic, such as direct sessions, paid search, or organic search/SEO.
Halo Effect Breakdown: Similar to “Halo Effect Only,” but also indicates which storefront the halo effect revenue ultimately went to.
Shopify Only: Isolates the channel base vs. halo mix specifically for revenue on your primary storefront. (Note that the name of this tab will match your eCommerce platform, e.g. Shopify, BigCommerce, etc.)
Amazon Only: Breaks down how much base and halo revenue is being driven to your Amazon store, per channel. Hover over the bars for exact dollar values and to see how halo effects are split by value.

MMM Fit Chart

Next to the Base vs Halo Effect Rollup is the MMM Fit chart. This visualization shows the share of your total modeled performance—either Revenue or New Customers—that is driven by paid media, seasonality, and trend across all connected storefronts.

When viewing Revenue, the legend indicates:

MMM Total The total forecasted revenue inclusive of paid media, seasonality, and trend.
Reported Total The actual revenue reported for your sales channels Prescient is modeling against (this may exclude non-transactional revenue such as subscriptions and/or wholesale).
R-MMM Fit (1-sMAPE)*100; This shows how well Prescient’s model fits actual reported revenue. The closer the value to 100 the better the fit.

Switching the chart to New Customers follows the same logic—just substitute revenue with new customer counts.

sMAPE stands for Symmetric Mean Absolute Percentage Error. It is a statistical measure used to assess the accuracy of forecasted values compared to actual values. It is especially popular in time series forecasting because it provides a percentage error, which makes it easier to interpret across different scales. One of the handy things about sMAPE is that the value is the same for over- or underestimation. Hence, if we predicted 1.2M and the actual was 1M, the sMAPE would be the same if we had predicted 800K.

We calculate sMAPE on a daily basis for our models, meaning that we calculate the under or overage differences daily and we take a simple average of the % differences over a specified time range (which is the most rigorous).

You can use this chart to quickly see where your media how well our models capture the signals that explain your performance or whether external factors (e.g., one-time events, on-site promotions, or word-of-mouth) are driving significant unexplained performance.

Trends

The Trends table breaks down all of the KPIs in the KPI Report based on a time frame you can select using a date picker at the top of the Trends report. It includes six columns:

Their total over the selected period (e.g. 30 days)
Their total over the prior period
The difference between the selected period and the prior period
A small graph reflecting that metric over the selected period
A % change between the selected period and the prior period

These trends are available in aggregate for your entire business or can be broken down for each connected ad channel.

KPI Report

This visualization lets you choose from various Ecommerce, advertising, and Prescient-modeled KPIs, all broken down by channel. Examples include:

MMM Revenue
Channel Reported Spend
Channel Reported Revenue
Ecommerce Revenue
Average Order Value (AOV)
Web Sessions

When you select a KPI from the left-hand panel, the graphs update to display your chosen KPI for each connected data source. For example, you can view Modeled Revenue broken down by ad channel or MMM % of Total broken down by revenue source.

You can toggle the visibility of each source by clicking the lines in the legend below. Above the graph, dropdown menus allow you to adjust the visualization with the following options:

View Type: Select Lines to compare individual channels or Stacked to show all channels combined for overall performance.
Aggregation: Group data by Day, Week, Month, Quarter, or Year, depending on the visualized time frame.
Comparison Mode: Choose Absolute for actual totals or % of Total to view each channel’s share of the metric.
Revenue Source: Display revenue from Shopify Only or Amazon Selling Partner Only.
Date Range: Define the time frame for the KPI report.

Performance Page Walkthrough

The Performance Page is where Prescient users can dive deeper into the metrics introduced on the homepage. It is designed to provide comprehensive performance tracking and enable granular analysis. Here’s a detailed breakdown:

Hero Metrics

The top of the performance page supports a wide range of customizable KPIs that report the overall performance of your business, along with a date picker to control the data shown on the entire page. Reported, Ecommerce, and modeled metrics are all available, and the metrics shown, along with their order, can be customized using the Metrics button located just below the date picker.

Modeled metrics derived from Prescient's attribution models are labeled with 'MMM' (e.g., MMM Paid Revenue, MMM Paid New Customers, etc.), while channel-reported metrics are labeled with 'Channel' (e.g., Channel % of Total, Channel Revenue, etc.). These hero metrics are numerous, and a full breakdown is available within the Glossary.

Channel, Tactic, & Campaign views

While the hero metrics at the top of the performance page always display aggregated totals for your entire business over the selected dates, the performance table can be aggregated at three different levels: Channel, Tactic, or Campaign.

All views display the date your models were last updated in the bottom-left corner, and views may include multiple pages depending on the Rows per Page setting in the bottom-right corner (default: 50). If you have more than 50 campaigns or tactics, only the 50 highest-spend rows will be displayed by default.

Channel Level View

The Channel-Level View displays performance metrics for all your ad channels, including Prescient-modeled metrics and channel-reported metrics. Prescient-modeled metrics are highlighted in green and include a Prescient logo at the top, while reported metrics are in white. Similar to the Hero Metrics, anything derived from Prescient's models are labeled with 'MMM' in the metric name.

You can filter the performance table down to a specific channel by clicking the channel’s name on the left-hand side of the screen.

Tactic View

The Tactic View leverages your created tactic groupings to report campaign performance metrics based on how you expect them to behave within your marketing funnel, strategy, or goal expectations.

For more details on tactic labels and guidance on creating your own tactics, visit Tactics Guidelines / Labeling.

Campaigns View

The Campaigns View is the most detailed view on the Prescient Performance Page. It enables you to see all of your advertising campaigns' metrics side by side. Any connected ad channel with active campaigns within the selected timeframe will be shown in this view. You can further filter, customize, and arrange the view using the controls at the top of the table. More details on creating saved views, controlling column order and visibility, filtering, and exporting data can be found in Saved Views, Columns, Filtering, & Exporting below.

Insight Drawers

Prescient's Insight Drawers provide more in-depth performance information for your campaigns. To open an insight drawer, click a campaign name in the Campaigns View. This action opens a new window with detailed data for your selected campaign, which is organized into several tabs:

Performance: This section includes two graphs: Return and Custom:
- Return: This graph allows you to compare two KPIs directly, focusing on the relationship between spend and MMM Revenue/ROAS, MMM New Customers/CAC, and Channel-Reported Revenue/ROAS. Dropdown menus above the graph let you choose KPIs, while the upper-right dropdown enables grouping by specific time intervals.
- Custom: This graph lets you visualize and compare metrics of a similar data type, such as Dollars (e.g., Spend, Revenue, CAC) and Numeric (e.g., Impressions, ROAS, Customers). Used alongside the Return graph, it allows for detailed performance analysis across multiple KPIs.
Attribution: This tab focuses on campaign-level halo effects, breaking down revenue into direct storefront impacts and secondary halo effects by source. More details on halo effects are available in our Halo Effects page.
Forecast: This tab displays saturation curves for revenue and new customers for your primary storefront and Amazon storefront (if connected). A table at the top provides modeled performance metrics for current spend and a 10% increase scenario. You can adjust inputs, and the saturation graph will update accordingly. Detailed information on these systems is available in Campaign Forecasting and Spend Optimization pages.

📘
Insight Drawer Tips
You can have up to two insight drawers open at once depending on the size of your screen.
You can also open insight drawers in their own window using the expand icon at the top of the drawer.

Saved Views, Columns, Filtering, & Exporting

In addition to customizing hero metrics at the top of the page, the Performance Page supports view saving, column editing, data filtering, and CSV exporting. Here’s a breakdown of these features:

Data Filtering: Data can be filtered in all views on the Performance Page using the Filter button. You can filter by specific channels, campaign names, or metrics. Prescient also provides predefined filters to easily find your best or worst campaigns. Filters can be removed via the filter menu or by using filter tags above the table.
Column Editing: Important columns are visible by default, but you can add and reorder columns using the Columns button.
View Saving: Filters and column configurations can be saved into easily accessible, repeatable views via the Saved Views dropdown above the performance table. Saved views are user-specific and retain all applied filters and column settings. Prescient provides two default views: Halo View for campaigns with halo effects and Has MMM Revenue for channels/campaigns/tactics with attributed revenue from Prescient’s models.
- To create your own view, apply filters or column settings, click the Saved Views dropdown (labeled Unsaved Changes or All campaigns, depending on if you already have filters applied), and select Save New View. You can access, edit, or delete views at any time using the dropdown menu, and these views can also be used when creating Spend Optimizations.
CSV Exporting: Data on the Performance Page can be exported for use in other tools. Clicking the Export Performance button exports visible columns and all available campaigns. You can also click the arrow next to the button to export all supported columns.

Using Performance Measurement to Inform Decisions

By leveraging these tools, marketers can:

Identify high- and low-performing campaigns: Use the Homepage's KPI Report to monitor performance across channels, such as Modeled Revenue or AOV, and the Performance Page's Campaign View to compare detailed metrics like ROAS and CAC side by side.
Optimize channel strategy: Leverage the Homepage's Base vs. Halo Effect Rollup to understand how channels contribute to base and halo revenue, and the Performance Page's Insight Drawers to analyze campaign-level saturation and impact.
Monitor progress toward goals: Set and track revenue, ROAS, or budget targets using the Homepage's Goals Section, ensuring marketing decisions align with broader business objectives.
Enhance collaboration and decision-making: Save and share custom views on the Performance Page to streamline reporting and ensure teams have access to actionable insights.