This guide will help you understand:

What are Analyses in Bloomreach Engagement

The Bloomreach Engagement platform contains several useful tools collectively called analyses. The main purpose of these tools is to allow you to effectively manage your data. Not only do they allow you to gather new insights into the behaviour of your customers, but they also help in presenting your findings by transforming raw data into comprehensible data structures.

The analytics can be found by clicking on the Analyses section located on the menu in the Bloomreach Engagement web application.

698698

By clicking on the button highlighted in red, the analyses menu will appear.

Analyses currently available in Bloomreach Engagement

Tool

What does the tool do?

When can it be useful?

1. Dashboards

Displays key metrics and trends.

Present your most important data in one place.

2. Trends

Plots occurrence of a given event on a timeline.

Visualise the frequency of events.

3. Funnels

Plots the number of customers successfully reaching an event from a sequence of events.

Observe the step by step journey of a customer towards a specific outcome.

4. Reports

Summarise your data on a grid table by a custom metric.

Visualise important relations in your data.

5. Retentions

Display a table view of timeline of occurrence of an event happening after an anchor event had already happened.

Provide insight into how often do customers return to interact with your business.

6. Segmentations

Displays a chart graph of specified subgroups of customers.

Group customers on the basis of their attributes and behaviour.

7. Flows

Displays a map of the possible journeys that a customer can take.

Describe how customers navigate your website.

8. Geo Analyses

Displays a map of locations where a given event has occurred.

Provides insight into the geographical distribution of your customer base.

9. Predictions

Evaluates customer's behaviour using artificial intelligence.

Predict the behaviour of your customers.

10. SQL Reports

Displays reports that are using data from 3rd party data sources written in SQL query or visual mode.

Show the results where your report uses the external data.

To find out more about specific analytical tools, please visit their dedicated page in the subsections.

How to interact with your analyses using Controls

Controls allow you to filter and interact with the data in dashboards, reports, and analyses. You can apply filters to all levels of underlying analyses (e.g. filter applies to a condition in an aggregate nested in an expression), making the best use of Bloomreach Engagement.

Controls are based on dynamic parameters which can be added to any analysis (e.g. report, metric, funnel, etc.). They don't require you to provide an exact static value for a condition; instead, the dynamic parameter serves as a placeholder. This allows anyone in read or preview mode for a particular analysis to provide the static value at a later stage.

How to later provide the static value for a dynamic parameter?

  • Manually by someone on the Preview screen for a particular report.
  • Values such as ‘campaign_id’, ‘banner_id’, ‘experiment_id’, or ‘survey_id’ can be provided automatically for a particular campaign if the dashboard is picked in the Evaluate tab.

How to make use of Controls

You can make use of Controls in multiple different ways, including:

  • filtering the data by a specific value of a condition in an aggregate
  • specifying campaign ID in multiple funnel steps or the running aggregate in an expression
  • changing the attribution window of a report (used in multiple underlying analyses)

How to set up Controls

  1. Visit the Edit mode of your report.
996996
  1. In any place where a static value is used (e.g. example.com), replace it with a dynamic parameter. Start by typing double square brackets, continuing with the parameter name, and ending with double square brackets (e.g. [[email_domain]] ).

  2. Once you close the brackets, the look of the input field changes to blue color, indicating that the application has accepted a dynamic parameter instead of a static value.

10801080
  1. A new Control appears near the bottom of your analysis, helping you specify a static value and check if your analysis is calculated correctly.
842842
  1. Try to input different values and check if the report displays correctly by confirming the value using the ‘Preview’ or ‘Refresh’ button.

  2. Suppose your analysis is used as part of another analysis (e.g. metric is used in a report or report in a dashboard). In that case, the Controls are also displayed in the read mode for the top-level analysis. This allows the end-user to specify values for the controls and interact with the displayed data.

Example of using Controls

Digital Analyst creates a set of advanced reports for better product analytics insights. These reports include multiple complex conditions nested in various reports and analyses. The analyst cannot predict all static values for conditions and filters, so they opt for the dynamic parameters instead. They include a parameter for color, size, and brand of the products.

Their colleague, a merchandiser in the same company, is interested in the performance of all products in "yellow" color. The merchandiser opens the dashboard provided by the Digital Analyst and can specify the color of the target products thanks to the dashboard control.

All places where the dynamic parameter [[color]] was used were replaced with the static value "yellow". As a result, all the reports in the dashboards only account for "yellow" colored products.

Limitations

  • You can only specify one value for each control.
  • There is no support provided in API for analyses with dynamic attributes.
  • Jinja cannot be used where parameters are used (e.g. in filter conditions).

🚧

Updating a definition

If you decide to update a definition, replacing static parameters with dynamic parameters, you will also need to update all places where the definition has been used to support parameters. For example, if you're updating a metric already used in an unparameterized dashboard, you need to update the dashboard first.

This protects other users from breaking any analyses by changing and adding dynamic parameters to underlying definitions.


Did this page help you?