LTV Analysis

LTV Analysis Overview

Lifetime Value (LTV) is the commercial value contributed by a user over their lifetime. LTV analysis is an analytical model for analyzing user commercial value, which can analyze the average value contributed by a user group that visits on a specific date over a certain period of time.

LTV analysis can help answer the following questions:

Which channels bring in users with higher average value per user? Does the value contributed by users exceed the cost of acquiring them?
Did the campaign generate significant revenue? Did the average user value increase? Was the campaign ultimately good or bad?
Can the game product generate enough per capita spending to cover development costs? How long will it take for player value to stabilize?
Which user group represents the high-value customers for our product? What are the characteristics of these customers?

Typical case: What is the average recharge amount (LTV 30) of a new user who registered on September 10th over 30 days?

Calculation process:

Targeting new users through an initial date (September 10th) and an event (registration);
Calculate the cumulative recharge amount of the user group within the LTV (Lifetime Value) period (30 days) from the initial time (September 10th);
The LTV is calculated by dividing the total recharge amount by the initial number of users (the number of users who registered on September 10th).

LTV Analysis Interface Function Introduction

Select initial event or initial time

In LTV analysis, you can target the initial user group by event occurrence time or user attribute. Clicking the "Toggle" button allows you to switch between event and user attribute.

When selecting an event to anchor the initial user group, you can add filter conditions to view more granular dimensions. For example, if we want to analyze the situation of users in Beijing after registration, we can define the initial event as "registration" and add the filter condition "city equals Beijing".

Select revenue events

Adding revenue events allows you to analyze the corresponding Lifetime Value (LTV) for each event. LTV analysis supports adding multiple revenue events and analyzing their percentage of total revenue. For example, a game might have two revenue events: "Membership Top-up" and "Virtual Item Purchases." You can add these two events under the revenue events category and select their corresponding revenue amount attributes. The results table will then display the percentage of each event in the total revenue.

Filtering revenue events

Sometimes it's necessary to analyze the revenue contribution of specific revenue events, such as the revenue contributed by a newly developed virtual skin in the purchase of virtual items. In LTV analysis, you can add filters to identify the corresponding revenue events. For example, adding the filter "virtual items equal xx skin" will show the average revenue per user contributed by the sales of this skin.

Set the profit ratio for revenue events

In some cases, distributors take a cut from electronic payments, such as game top-ups on iOS, where Apple might take a 30% cut. In this case, the amount recharged by a customer doesn't necessarily represent the average value contributed by the user to the business. To accurately calculate LTV under these distributor commission scenarios, LTV analysis allows setting the profit margin for revenue events. In the example, setting the profit margin for iOS game top-ups to 70% provides a more accurate calculation of the LTV for iOS user game top-ups.

View by attribute

LTV supports grouping and viewing based on initial event attributes, user attributes, user segments, or tags, and multiple groups can be added simultaneously. For example, if you select user attributes to group by registration channel, you can see the LTV information for different registration channels.

If you select a numeric or time-type attribute here, you can customize the grouping interval. If not set, the query engine will dynamically calculate the grouping interval. This setting only applies to the current query; saving the query as a bookmark will also apply it to the bookmarks.

Set user filter criteria

You can add global user filters to analyze the LTV of specific user groups. For example, you can add "gender equal to female" to analyze the LTV of female users.

Trend View

LTV analysis supports viewing the results in two views: a trend view and a comparison view. The trend view shows the growth trend of LTV over retention time; the comparison view compares the changes in LTV over the initial date.

Trend Chart

The trend chart shows the growth trend of LTV as retention time increases.

The trend chart is displayed in two ways: when viewed as a whole, you can see the growth trend of each revenue metric and the LTV of total revenue as retention time increases; when viewed by attribute group, you can see the growth trend of total revenue corresponding to each group in the chart.

Trend Table

The trend table displays detailed LTV (Lifetime Value) figures. When multiple metrics are displayed, it shows total revenue and the percentage of each metric within that total revenue.

In the table,

When viewing by group, the group column is sorted by initial number of people from largest to smallest by default. You can re-sort the groups by group name, initial number of people, and total revenue (LTV) value for each group. You can also filter groups by group name.
In the metrics column, clicking the "Expand" button will expand the initial date row to view the LTV details for each initial date.
The Initial Users column displays the total number of initial users after deduplication for the time period, as well as the number of initial users who accessed the site on each initial date. Clicking on the number of users allows you to view the user list and create user groups.
The LTV columns 0-7, LTV 15, LTV 30, LTV 60, and LTV 90 display specific LTV values, including the LTV for each initial date and the total LTV for the time period. When multiple metrics are involved, the summary row will show the percentage of each metric in the total revenue.

View user details

The numbers in the initial "Number of Users" column are clickable. Clicking on them allows you to view detailed information about these users and further explore detailed behavioral sequences of individual users.

Custom LTV period

Supports displaying continuous LTV data from LTV 7 to LTV 15 and from LTV 16 to LTV 30.
Supports custom LTV periods, with a maximum LTV of 365.
Custom LTV data can be saved to the overview.

2.5.4. LTV Prediction

The trend chart supports switching to a forecast view.

It supports predicting LTV for 30, 60, 90, and 180 days, and displays the predicted values, maximum and minimum values obtained by the intelligent algorithm.

LTV Forecast Notes

Only when a single indicator is not grouped and the earliest date of the selected time period is at least 7 days ago can it be switched to the forecast view.
The forecast is the sum of the LTV for each date within the time period. Dates with incomplete data are removed during the forecast summary calculation. Therefore, the predicted LTV may be greater than the sum of the LTV in the trend view.

Comparison View

In the comparison view, you can compare how LTV changes over the initial date .

Comparison Chart

Similar to trend charts, comparison charts are displayed in two ways: when viewed as a whole, you can see how each revenue metric and the total revenue LTV changes with the initial date; when viewed by attribute group, you can see how the total revenue LTV for each group changes with the initial date.

The top right corner of the chart displays the current LTV value, which defaults to LTV 7. Users can switch the LTV value to be compared in the drop-down list. The available LTV values include LTV 0-7, LTV 15, LTV 30, LTV 60, and LTV 90 .

Comparison Table

Similar to the trend table, the comparison table also displays detailed LTV metric values. Furthermore, when multiple metrics are displayed, the total revenue is shown by default; clicking the "Expand" button in a metric column will show the LTV value of each metric and its percentage of total revenue.

In the table, when viewed by group, the groups are sorted in ascending order by group name by default. Group filtering is supported, allowing you to re-sort groups by group name and the total revenue (LTV) value for each group.