Skip to main content

Supported models + data basics

Trends AI data models

Written by Nico Londono

To power diverse ministry insights, Trends AI uses several data models. Whether you’re working with AI Analysis or the Chart builder, picking the right model for the task is the first step to gaining great insights.

Foundational Concepts

Data grain

Each data model in Subsplash Analytics is built around a concept called data grain. This simply refers to the level of detail that each model represents.

For example, the Giving model is structured around individual transactions, while the People model is built around individual profiles. Because of this, each model tells a different part of your ministry story, even if they sometimes include similar fields like names or emails.

Understanding the data grain is important because it determines how you can filter, group, and analyze your data in meaningful ways.

One model per chart

Each chart can only use one data model at a time.

However, a dashboard (board) can include multiple charts side by side. For example:

  • One chart can show giving trends from the Giving model.

  • Another can show attendance trends from the Event Attendance model.

If you want to compare different datasets, simply create separate charts and place them together on your board.

Imported tables

You can also upload your own data to create a custom table.

Imported tables function just like standard data models, meaning you can:

  • Build charts from your data

  • Use AI Analysis to explore insights

  • Apply filters and grouping just like native models

This makes it easy to combine Subsplash data with external datasets.

Exploring your data models

AI Analysis

AI Analysis helps you quickly understand what questions you can ask based on your data. Start with a role-based prompt:

“I’m a bookkeeper at a church. What questions can I ask using the Giving model?”

You’ll receive suggested questions tailored to your role and available data.

View column information

Within the chart builder, you can select any column to view its description in the Column Information panel. This helps clarify:

  • What the field represents

  • How it should be used in analysis

  • Whether it applies to filtering, grouping, or calculations

Giving Model

The Giving model provides a detailed view of all donation activity in your system. It helps you understand how giving is trending, who is giving, and how funds are being used across your organization.

Because it is built on transaction-level data, it gives you a very granular view of financial activity. In this model, you can explore questions like donor behavior, fund distribution, and overall giving trends.

Key fields

Some of the most important fields in the Giving model include:

  • Amount – Gross gift amount before fees

  • Fund Name – Designated fund or sub-fund

  • Transaction Date – Date the gift was given

  • End User UUID – Unique identifier for the donor

  • Instrument Type – Payment method (Card, ACH, etc.)

  • Is Recurring – Indicates if the gift is part of a recurring schedule

Best practices

To get the most accurate insights, it’s helpful to apply a few filters when analyzing giving data:

  • Use Is Successful to exclude failed or incomplete transactions

  • Use Is Tax Deductible to filter out non-donation activity

  • Use Is Non Traditional to identify non-standard giving types (e.g., special transactions)

Example questions

The Giving model is commonly used to answer questions such as:

  • How is giving trending this year compared to last year?

  • How many unique donors do we have?

  • Who are our top 10 givers this month?

  • What percentage of giving goes to each fund?

Recurring Gifts Model

The Recurring Gifts model focuses on scheduled giving activity rather than individual transactions. It helps you understand how many donors are giving consistently, when new recurring gifts are created, and what future giving may look like. This makes it especially useful for forecasting and donor retention analysis.

Key fields

Key fields in this model include:

  • Amount – Current recurring gift amount

  • Is Active – Whether the schedule is currently active

  • Interval – Frequency of the recurring gift

  • Next Gift Timestamp – Next scheduled gift date

  • Fund Name / Sub Fund Name – Assigned fund

  • Created At – When the schedule was created

This model reflects schedules only, not historical transactions. For past gift activity, use the Giving model.

Example questions

  • What is the trend in recurring gift creation over time?

  • What is the expected total recurring giving next month?

  • How many donors currently have active recurring gifts?

  • Which funds receive the most recurring gifts?

People Model

The People model includes all profiles in your system and helps you understand how your community is structured and engaged. It gives insight into who your people are, how they connect, and how your organization is growing over time.

Key fields

Key fields in the People model include:

  • Profile UUID – Unique identifier for each profile

  • Profile Created At – When the profile was created

  • Profile Created Source – How the person entered your system

  • Membership Status Name – Status (Member, Guest, etc.)

  • Total Profile Gifts – Number of gifts associated with the profile

  • Group Name – Groups the person belongs to

  • Custom Fields – Any additional fields configured by your team

Example questions

This model is commonly used to understand:

  • How many new profiles are created each month?

  • What percentage of members are in groups?

  • What is our demographic breakdown?

  • Where are our members located?

  • How many members have given financially?

Event Attendance Model

The Event Attendance model helps you analyze engagement across your events and services. It combines check-in data and headcount data to give a full picture of attendance trends over time.

Key fields

  • Event Title – Name of the event

  • Event Start At UTC – Event start date and time

  • Total Event Attendance – Total attendance count

  • Check In Count – Number of checked-in individuals

  • Session Headcount – Manually entered attendance count

Example questions

This model is often used to explore:

  • What is our total attendance over time?

  • Which events have the highest attendance?

  • Who attends most consistently?

  • What days of the week have the highest attendance?

Group Models

Group Membership

The Group Membership model shows who is currently connected to your groups and helps you understand overall group participation and structure.

Group Attendance

The Group Attendance model tracks participation in group events over time, helping you identify engagement trends and attendance patterns.

Example questions

Together, these models can help answer questions like:

  • Which groups are most active?

  • How is group attendance trending over time?

  • Which individuals participate in multiple groups?

  • What group types have the highest engagement?

Did this answer your question?