Sections in this category

Sisu Glossary

  • Updated

The Glossary defines common terms used throughout Sisu.

 

 

Admin terms

Custom Role

A special type of role that has privileged access to a private data source. See Creating & Managing Custom Roles.

Data Source

A connection to a centralized data repository like Postgres, Amazon Redshift, Google BigQuery, or Athena. To access a data source, first create a connection using credentials generated by your database administrator. See Understanding Data and Data Sources in Sisu.

Project

A grouping of metrics and analyses on these metrics. Best practices are to make this either a personal project, e.g. John's Project or a group's project, e.g. Marketing leads project. See Sisu 101.

Workspace

The Sisu homepage where you can see all projects.

 

Analysis Terms

Combination

When Sisu runs your data to determine factors that impact your defined metric, it considers all possible combinations of your data. 

These combinations describe the number of unique factor-value pairs that Sisu has analyzed.

For example:

  • tenure_in_months = 13+ months 
  • region = Canada 
  • tenure_in_months AND region = Canada constitute three unique combinations.

Data Set: Table

Data set that comes directly from the data source without additional data modeling. See Previewing the Underlying Data Table.

Date Set: Saved Query

Data set that comes from a SQL query. See Previewing the Underlying Data Table.

Exploration

A type of Analysis in Sisu that enables you to slice and dice your data manually. You can select the metric or dimension you want to explore, add or pivot by dimensions, and define filters—all without any SQL. You can then run the Exploration and visualize the output in table format and as various graph types that can be added to a dashboard and shared with other team members. See Understanding Explorations.

Fact

A set of statistics describing a subgroup in the data, including its size, its metric value (e.g. average, sum, rate or count), and its lift or drag on the overall population metric. A fact can describe over-performing and underperforming segments with respect to a baseline. See Understanding Facts.

Impact

This measure indicates to what degree various Subgroups in your data affect (either positively or negatively) important KPIs (Metrics) across your organization. See Understanding Impact.

Key Driver Analysis (KDA)

A type of Analysis in Sisu that enables you to explore why your data changes by identifying what is driving the change. Sisu uses its proprietary algorithms to identify subgroups that drive metric performance and calculate relevant statistics for each subgroup if they are statistically significant.  See Understanding Key Driver Analyses

KDA Type: General Performance 

A type of workflow on a metric that looks at over-performing and underperforming populations relative to the overall metric value in the dataset. e.g. a marketing campaign generally has 2x the conversion rate versus the entire population. See Configuring a General Performance Analysis and Interpreting a General Performance Analysis.

KDA Type:
Group Comparison

A type of workflow on a metric that looks at over-performing and underperforming populations within one group relative to another group. e.g. a marketing campaign had 2x the conversion rate within the mid-market segment compared to its conversion rate within the enterprise segment. See Configuring a Group Comparison Analysis and Interpreting a Group Comparison Analysis.

KDA Type:  Time Comparison

A type of workflow on a metric that looks at over-performing and underperforming populations relative to the metric values in a prior time period. e.g. a marketing campaign had 2x the conversion rate compared to its conversion rate in a time period. See Configuring a Time Comparison Analysis and Interpreting a Time Comparison Analysis.

Metric

A metric or KPI is a unit of analysis for Sisu. Choose a metric by selecting the relevant column in your dataset and tracking its average, sum, or count. By adding a metric in the analysis page, workflows like general comparison, time comparison, and group comparison can be run. See Understanding Metrics and Defining a Metric.

Metric Type

Method of aggregating a column into a metric. For numerical metrics, the methods are average, sum, and count. For categorical metrics, they are rate and row count. See Understanding Metrics.

Column Type: Numerical

Numerical metrics derived from a column that is numerical in nature. For example:  order amount, item price, age (when analyzing average age), etc. See Understanding Metrics.

Column Type: Categorical

Numerical metrics derived from a column that is categorical in nature. For example, gender, age (when analyzing age “groups” as categories), SKUs, etc. See Understanding Metrics.

Size

The percent of total rows that a subgroup represents. This enables you to understand how big a population is, or how much it may have grown from one period to the other.

Statistical Model

A mathematical model that embodies a set of statistical assumptions concerning the generation of sample data. A statistical model represents, often in considerably idealized form, the data-generating process. Sisu uses three different statistical models to develop Analyses. See Understanding Statistical Models.

Subgroup

Set of column-value pairs (up to 3 pairs) that determines rows of data for analysis. Subgroups are displayed in the Fact Table in all Analysis types.