Skip to main content Skip to complementary content


Administrator, analytics

Analytics administrators are administrators who only have permissions to manage the Analytics service's user resources.

See:Permissions for analytics administrators

Administrator, audit

Audit administrators, when also assigned the Developer role, can access app feedback and usage metrics for Insight Advisor and Insight Advisor Chat.

See: Permissions for audit administrators

Administrator, data

Data administrators are administrators limited to managing data spaces and the data resources contained within those spaces.

See: Permissions for data administrators

Administrator, tenant

Tenant administrator are administrators responsible for managing users, system extensions, and system customizations. Tenant administrators have full access to the Management Console.

See:Permissions for tenant administrators


An aggregation is a calculation using multiple records in the source tables. Often it is a single field aggregated with a function such as sum, count, min, max, or average. For example, the sum of sales.

Similar terms: Calculation, Measure, Metric

See: Measures

Analyzer user entitlement

Analyzer entitlement is intended for users who only consume sheets and apps created by others. You need Professional entitlement to create, edit, or publish sheets or apps.

See: Assigning user entitlements


This term has several definitions:

  1. A Qlik Sense or QlikView app: Apps are task-specific, purpose-built applications. Apps contain data loaded from data sources that is interpreted through visualizations.

    Similar terms: Document, workbook

    See: Understanding apps

  2. The Qlik Sense Mobile app: A mobile app for iOS and Android devices. In the mobile app, you connect to and interact with your cloud data. You can work with your available apps.

    See: Qlik Sense Mobile SaaS

Apply dataset

The dataset on which a trained machine learning model is making predictions. The output of a Qlik AutoML prediction is a new dataset with predicted values for a chosen target field.

See: Creating predictions on datasets

Area under curve (AUC)

A ROC Curve for AUC (area under curve) describes how good a machine learning model is at predicting the positive class when the actual outcome is positive. The closer the True Positive Rate is to 1.0 (the maximum possible area under the curve), the more deterministic the model is. The ROC Curve is useful for understanding if separation between classes is possible, thereby indicating if the data is good enough to accurately distinguish between predicted outcomes.

See: AUC and ROC curve

Associative insights

Associative insights is an Insight Advisor feature that compares your selections and the values excluded by your selections to help you find blind spots and hidden releationships.

See: Discovering your data with associative insights

Basic User

A user type with limited access. As a Basic User, you can view app content in managed spaces.

See: Managing user entitlements


Bookmarks let you save specific selection states so they can be applied again in an app later and shared with other users. Layout information can be stored in bookmarks, so users can be taken to the correct place in the app when applying the bookmark.

SeeBookmarking selections

Business logic

Business logic is a suite of features that allows you to customize the behavior of Insight Advisor. Business logic has two main features:

  • Logical model: The data model of the app used when generating visualizations.

  • Vocabulary: Alternative terms for natural language queries.

Similar terms: Logical model, vocabulary, domain logic

SeeCustomizing logical models for Insight Advisor


The uniqueness of data values in a column. When training a machine learning model, Qlik AutoML can't use feature columns with high cardinality (too many unique values) or no cardinality (column data consisting of one constant value).

See: Cardinality


The catalog is the area of the hub where you can find your apps and data sources. Catalog tools allow you to profile your data.

See:Understanding your data with catalog tools

CDC (Change Data Capture)

In CDC landing mode, changes to the data source are captured as they occur and applied to the target in near real time.

See:Landing data from data sources and Full load below.


Charts are objects where calculations, aggregations, and groupings can be made. Graphical visualizations, such as bar charts and pie charts are common examples, but also non-graphical objects such as pivot tables are charts.

A chart consists of dimensions and measures, where the measures are calculated once per dimensional value. If the chart contains multiple dimensions, the measures are calculated once per combination of dimensional values.

Similar terms: Visualization, Hypercube, Cube

See: Choosing the right visualization


Collections are a tool that let you group apps, charts, notes, automations, experiments and links in the hub for ease of access and sorting.

See: Collections

Confusion matrix

A confusion matrix summarizes the accuracy of prediction results in a classification model. The number of correct and incorrect predictions are summarized for each class. This gives you insight not only into the errors being made by your classifier but also the types of errors that are being made.

See: Confusion matrix


A process completed during experiment training in Qlik AutoML in which training data is split into five segments (folds), allowing each segment of the data to be used as a test on the other four segments. Cross-validation results in metrics that will show how well a model can make predictions on data it has never seen before.

See: Holdout data and cross-validation

Custom object

Custom objects are custom visualizations added to tenants by tenant administrators or included in one of the Qlik extension bundles.

Similar terms: Extensions

See: Creating a visualization using a custom object

Data asset

A fit-for-purpose collection of datasets.

See:Creating a data pipeline in a data project

Data catalog

The data catalog is a component in Data manager and Data load editor that enables you to select and load data from all the datasets to which you have access.

See: Loading data from the data catalog

Data connection

Data connections are used to let data tasks access data sources, and external storage and cloud data warehouses used in a data project.

See:Connecting to data sources

Data Gateway, Data Movement

Qlik Data Gateway - Data Movement allows you to move firewalled data from your enterprise data sources to cloud and on-premises targets, over a strictly outbound, encrypted and mutually authenticated connection. By eliminating the need to open inbound firewall ports, Qlik Data Gateway - Data Movement provides a secure and trusted means for accessing your enterprise data.

See:Qlik Data Gateway - Data Movement overview

Data Gateway, Direct Access

Qlik Data Gateway - Direct Access allows Qlik SenseSaaS applications to securely access firewalled data, over a strictly outbound, encrypted, and mutually authenticated connection.

See: Qlik Data Gateway - Direct Access overview

Data leakage

An undesired phenomenon in machine learning in which an algorithm is trained with data for which it will be used to generate predictions. One indicator of data leakage is unrealistically high model performance, which results only from memorization of target values (which were incorrectly provided in the training data, either directly or indirectly) and not from actual learning of patterns and associations.

See: Data leakage

Data load editor

The Data load editor is a script editor that allows you to build and customize the script that loads data into your app.

Similar terms: Script editor

See: Loading and transforming data with scripting

Data manager

Data manager is an app component that allows you to load and manage data sources in an app. You can also preview and associate the data and perform data transformations.

See: Loading and managing data with Data Manager

Data mart

Data marts are the part of your data pipeline that contain a subset of data from your Storage or Transform data assets. You can create any number of data marts depending on your business needs. Ideally, your data marts should contain repositories of summarized data collected for analysis on a specific section or unit within your organization.

See: Creating and managing data marts

Data model viewer

Data model viewer is an app component that allows you to view the structure of the data added to an app and view metadata about tables and fields.

See: Viewing and transforming the data model

Data pipeline

In a data project, a data pipeline is a set of tasks for integrating data. Onboarding moves data into the project from data sources that are on-premises or in the cloud and store the data in ready-to-consume data sets. You can also perform transformations and create data marts to leverage your generated and transformed data sets. The data pipeline can be simple and linear, or it can be a complex pipeline consuming several data sources and generating many outputs.

See: Creating a data pipeline in a data project

Data profiling

Data profiling displays statistics and information about your data sets.

See: Managing field-level metadata and data profiling

Data project

A data project is where you create your data pipeline, using data assets. The data project is associated with a data platform that is used as target for all output. You can create a simple linear pipeline, or a complex pipeline consuming several data sources and generating many outputs.

See: Introducing Qlik Cloud Data Integration

Data task

In a data project. you can create data tasks of the following types:

  • Landing - Move your data from the data source to a cloud data warehouse or Qlik Cloud.
  • Storage - Store landing data in read-to-consume datasets.
  • Transform - Transform your data according to target requirements and business needs.
  • Data mart - Create data marts from stored or transformed data.

You can also combine the landing and storage into a single "Onboarding" task.

See: Data task


The term dataset is usually synonymous with table. It can refer to the original source table, the table after undergoing transformations, or the fact and dimension tables in a data mart.

See: Managing datasets


In Analytics Services:

A dimension is an entity used to categorize data in a chart. For example, the slices in a pie chart or the bars of a bar chart represent individual values in a dimension. Dimensions are often a single field with discrete values, but can also be calculated in an expression.

Similar terms: Category, group

See: Data grouping with dimensions

In Data Integration:

A dimension is a dataset in a data mart that forms part of the star schema. Dimension datasets hold the descriptive information for all related fields that are included in the fact table’s records. A few common examples of dimension datasets are Customer and Product. Since the data in a dimension dataset is often denormalized, dimension datasets have a large number of columns.

See: Creating and managing data marts

Dynamic views

Dynamic views allow you to query and view relevant subsets of large datasets from another app in a chart. They can be dynamically refreshed as selections are made. Dynamic views are similar to on-demand apps.

See: Managing data with dynamic views


A fact table works with dimension tables. A fact table holds the data to be analyzed, and a dimension table stores data about the ways in which the data in the fact table can be analyzed. Thus, the fact table consists of two types of columns: primary key columns and measure columns. The primary key columns allows joins with dimension tables, and the measures columns contain the data that is being analyzed. For example, the Orders fact dataset will simply list that on such and such a day, a certain customer bought a quantity of a certain product whereas the related dimension datasets will provide more information about the customer, the product, and the region in which the transaction took place.

See: Creating and managing data marts


Favorites is a section available to all users where they can add apps, datasets, automations, notes, experiments and charts from the hub. Favorites are private.

See: Favorites

Feature (machine learning)

A variable in a machine learning problem that can potentially contribute to the value of the target column. In Qlik AutoML, features are columns in a dataset which can have varying levels of influence on the results in the target column.

See: Machine learning concepts


A field is a data asset containing values, loaded from a data source. At a basic level, a field corresponds to a column in a table. Fields are used to create dimensions and measures in visualizations.


Full load

In Qlik Cloud Data Integration, Full Load refers to the initial replication of data from the data source to the landing.

See: Landing data from data sources and CDC above.

Full User

The Full User entitlement is applicable to all use cases in Qlik Cloud. As a Full User, you can do anything that your permissions allow, such as create shared spaces, create, edit, and publish sheets or apps, or work with Data Integration.

See: Managing user entitlements

Holdout data

In an ML experiment, holdout data is a subset of the training dataset that is set aside and not used during the training of an algorithm and subsequently used to score model performance.

See: Holdout data and cross-validation


The hub is the central point of access for apps, spaces, and collections.

Similar terms: Access point

See The Qlik Cloud Analytics hub

Hyperparameter optimization

A process of fine-tuning the constraints, weights, and learning rates of a machine learning model to increase its performance in solving a problem. In Qlik AutoML, this is not enabled by default but can be applied during experiment configuration for potentially improved results.

See: Hyperparameter optimization

Impact analysis

Impact analysis shows the forward-looking, downstream view of a database, app, or field dependencies.

See: Analyzing impact analysis for apps and datasets

Insight Advisor

Insight Advisor is a suite of features. Insight Advisor can help build your data model, create visualizations, and analyze your data.

See: What is Insight Advisor and business logic?

Insight Advisor Analysis Types

Insight Advisor Analysis Types is an Insight Advisor feature that creates visualizations for you by letting you select the analysis type you want to see and the data you want used.

See: Insight Advisor Analysis Types

Insight Advisor Chat

Insight Advisor Chat is a chat-based interface for conversational analytics. Insight Advisor Chat enables you to ask questions from the apps to which you have access. Insight Advisor Chat then returns relevant visualizations.

See: Exploring apps with conversational analytics

Insight Advisor Search

Insight Advisor Search is a Insight Advisor feature that allows you to ask natural language questions in an app and receive relevant visualizations.

See: Insight Advisor Search


The process of moving data from a data source to a landing zone in a cloud data warehouse.

See: Dataset architecture in a cloud data warehouse


In Qlik Cloud Data Integration, landing can refer to one of the following:

  • A database - commonly known as a landing zone - in a cloud data warehouse where source data initially "lands" before being processed further down the data pipeline
  • The actual task or process that moves the data from the data source to the landing zone

The landing task in Qlik Cloud Data Integration controls the continuous or scheduled landing of data from the data sources to the landing zone.

See: Dataset architecture in a cloud data warehouse

Large app support

With large app support, you can work with apps that are larger than the standard app size. Administrators can assign large app support to specific spaces.

The standard app size is up to 5 GB for Qlik Sense Enterprise SaaS and up to 10 GB for Qlik Cloud Analytics Premium and Enterprise.

See: Large app support


Lineage tracks data and data transformations back to their original source, representing it as a lineage graph.

See: Analyzing lineage for apps and datasets

Live views

In Qlik Cloud Data Integration, Live views allow you to access both current data (ODS) and historical data (HDS). Live views include data from changes tables that is not yet applied to the current or prior tables. This lets you see data with lower latency without having to apply changed data frequently. The ability to delay the merge allows for reduced costs and processing requirements in the target platform.

See also: Live view

Load script

The load script is a sequence of statements that defines what data to load and how to link the different loaded tables. It can be generated with the Data manager, or with the Data load editor, where it also can be viewed and edited.

Similar terms: Script

See: Loading and transforming data with scripting

Logical model

The logical model is the underlying data model that tells Insight Advisor how to use data when generating visualizations.

Similar terms: Logic model, logic framework

See: Building logical models for Insight Advisor with Business logic

Master item

Master items are dimension, measures, or visualizations that have been saved so they can be reused in other visualizations or sheets. You can then make changes or updates to the master item in a single place and have it impact all objects that use the master item.

SeeReusing assets with master items


A measure is a calculation base on one ore more aggregations. For example, the sum of sales is a single aggregation, while the sum of sales divided by the count of customers is a measure based on two aggregations.

Similar terms: Aggregation, calculation, metric

See: Measures

ML deployment

A model, generated from a single algorithm within a single experiment version, which is deployed to generate predictions in Qlik AutoML. It is the result usually achieved after configuring the experiment over multiple versions to produce the required level of performance for your use case. An ML deployment is available as an asset which is accessible from the catalog.

See: Deploying models

ML experiment

An asset in Qlik AutoML which allows you to train a machine learning model using historical data, with the goal of deploying one of the final results for making predictions on new data.

See: Working with experiments

Model metrics

Details about how well an algorithm is performing during experiment training in Qlik AutoML. Model metrics display how accurately each model is learning to determine the correct outcome to the machine learning problem. This helps in deciding which model you will deploy for use in creating predictions.

See: Viewing model scores

Monitor in hub

Monitor in hub is a feature that allows you to add charts from sheets or Insight Advisor to the hub so you can monitor them, without opening an app.

SeeMonitoring visualizations


Notes let you quickly add text commentary to an app or chart. They can also contain snapshots of data. You can keep them private or share them with other users.

Similar terms: Comments

SeeCapturing and sharing insights using Notes


Notifications let you know when there have been changes to apps or spaces, or when alerts you set have been triggered.

See: Notifications

On-demand app

On-demand apps load selections of data from large datasets without loading the full dataset into the app.

See: On-demand apps


In a data project, the first step of creating a data pipeline is onboarding the data. This involves transferring the data continuously from the on-premises data source and generating datasets in read-optimized format.

See: Onboarding data


Permissions are the roles in data, shared, and managed spaces that determine what users can do in that space.

Similar terms: Rights, user roles, security roles


Permutation importance

How much the machine learning model performance depends on a feature. When iterating on which features to include in the model, permutation importance can be used to determine which columns to keep, and which columns can be dropped.

See: Permutation importance

Preceding load

A preceding load is a script construct that allows you to load from the following LOAD or SELECT statement without specifying that source. Preceding loads are often faster than resident loads.

See: Loading data from a previously loaded table

Prediction (machine learning)

An estimate, by a machine learning model, of the future value of a target column. In Qlik AutoML, predictions are generated by ML deployments as one or more datasets within a personal, shared, or managed space.

See: Working with ML predictions

Professional user entitlement

Professional entitlement is intended for users who will create content in Qlik Sense. A user with Professional entitlement can create shared spaces, create, edit, and publish sheets or apps, and can create and edit alerts.

See: Assigning user entitlements


When editing a sheet, Properties contains options for configuring and styling visualizations.

SeeProperties panel


A QlikView Data file (QVD) is a file containing data exported from Qlik Sense or QlikView. It is a native Qlik Sense format optimized for loading data quickly. You can also generate QVD files with Qlik Cloud Data Integration.

See: Working with QVD files

Resident load

A resident load is a script construct that allows you to load from an already loaded table. Resident loads are often faster than accessing the original data source again.

See: Loading data from a previously loaded table


In a data project, artifacts are generated in an internal schema and a data asset schema.

  • The internal schema contains the physical data tables.

  • The data asset schema contains the views that you can use to consume the data.

See: Dataset architecture in a cloud data warehouse

Section access

Section access is a section of the data load script containing a security table that defines which users can see what data in an app.

See: Managing data security with Section Access


Selections are values selected by a user in visualizations in an app used to filter data. When a selection is made, all associated visualizations are updated to reflect the selection. Selections can be saved as bookmarks, and shared with other users.

See: Exploring data with visualizations

Service account owner

Service account owners (SAO) are responsible for maintaining and configuring the Qlik Cloud subscription.

See: Service account owner

Set analysis

Set analysis defines an alternative set of selections in a visualization different from those in the current selection. This allows for comparative analysis.

SeeSet analysis

SHAP importance

A measurement of how much influence each feature in an experiment has on the predicted outcome of the target. Qlik AutoML creates aggregated, row-level SHAP importance ranking charts automatically during the training of binary classification and regression experiments.

See: SHAP importance


Sheets are components of Qlik Sense apps. They present visualizations to app users so they can explore, analyze, and discover data. Sheets can be public or private.

Similar terms: Dashboard, worksheet

SeeSheet view

Sheet objects

Sheets objects are components used to create an interface on a sheet. Not all sheet objects visualize data such as tables and charts. The can include other objects such as buttons, text objects, and extensions.

Similar terms: Visualization, chart

See: Sheet view


Snapshots are graphical representations of a visualization at a certain point of time. Snapshots are used to create stories.

Similar terms: Screenshot

SeeCollecting insights for stories using snapshots

Space, data

Data spaces are governed areas of your Qlik Cloud tenant that are used to create and store data projects. Inside the space, you can also create new data connections using connectors, and manage access to Data Movement gateways. All data assets will be created in the space of the data project that they belong to.

See: Working in spaces in Qlik Cloud Data Integration

Space, managed

Managed spaces are carefully controlled spaces used to share apps with a limited group of users.

SeeWorking in managed spaces

Space, personal

A personal space is a private space belonging to users where they can develop apps.

SeeWorking in spaces

Space, shared

Shared spaces are areas apps and data sources can be shared with other users for collaborative development.

SeeWorking in shared spaces


Storage is the part of the data pipeline that contains ready to consume datasets in a cloud data warehouse, or in Qlik Cloud, from the data copied from the landing zone. The datasets can be kept up-to-date with the landing zone data without manual intervention.

See: Storing datasets


A story is an app tool that allows you to share data insights and discoveries you have made in an app with other users, combining reporting, presentation, and exploratory analysis.

Similar terms: Report

SeeSharing insights with data storytelling


Subscription reports let you schedule recurring emails containing a PDF of your selected sheets or charts.

SeeScheduling reports with subscriptions

Synthetic key

A synthetic key is a composite key between two tables in the data model. They are created when two or more tables have two or more fields in common. If you receive a warning about synthetic keys when loading data, you may need to review the data structure in the data model viewer. A synthetic key is not necessarily a problem, but if you have synthetic keys based on other synthetic keys, there may be errors in your data model.

SeeSynthetic keys

Tables: ODS, HDS, and Change

In a data project, there are several types of table, which may or may not exist depending on the project settings:

  • Current table (ODS)

    This table contains the replica of the data source updated with changes during the latest apply interval.

  • Prior table (HDS)

    This table contains Type 2 historical data. It is only generated if History is enabled in the data task settings.

    When a source table record is updated, a new record is added to the prior table each time. The history record is a copy of the previous current record, which also includes what was updated, and when it was valid.

  • Change table

    This table contains all changes that are not yet applied to the current table. It is only generated if the landing mode Full load and CDC is used.

See: Dataset architecture in a cloud data warehouse


This term has several definitions:

  1. In the context of machine learning, a target is the information that a machine learning problem aims to predict. In Qlik AutoML, the target is a column in the dataset.

    See: Machine learning concepts

  2. In the context of data movement, a target refers to the destination or endpoint where data is intended to be transferred, stored, or loaded. It's the location or system that receives the data being moved from a source. The target could be a database, data warehouse, cloud storage, server, or any other destination where the data is meant to be utilized, analyzed, or processed. Data movement involves transferring information from a source to a target, often as part of data integration, migration, or synchronization processes.

    See: Integrating data


The tenant is your cloud deployment of Qlik Cloud. It is a container for items such as users, apps, and spaces.

Similar terms: Deployment

SeeCreating and configuring the tenant

Training dataset

The dataset used to train a machine learning model in Qlik AutoML. By allowing algorithms to learn patterns and associations in your data, the resulting model is equipped to make predictions on new data (the apply dataset).

See: Getting your dataset ready for training


A transform task is the part of your data pipeline that allows you to create reusable data transformations based on rules and custom SQL. You can perform row-level transformations and create datasets that are either materialized as tables, or created as views that perform transformations on the fly.

See:Transforming data

Type 1 - Operational Data Store (ODS)

In a Type 1 dataset, the new information simply overwrites the original information. In other words, no history is kept.

See: Dataset architecture in a cloud data warehouse

Type 2 - Historical Data Store (HDS)

In a Type 2 dataset, a new record is added to the table to represent the new information. Therefore, both the original and the new record will be present. The new record will have its own primary key.

See: Dataset architecture in a cloud data warehouse


A variable in Qlik Sense is a container storing a static value or a calculation, for example a numeric or alphanumeric value.

See: Using variables in expressions


Views are used in data projects. A view is a virtual representation of the physical datasets. As views result from querying the physical datasets, they will always pick the relevant data from the underlying dataset. Views have several advantages over physical datasets: they are able to produce a single result set from joins between multiple datasets; they control access to underlying data; and they do not take up any significant disk space.

There can be several types of view in a data project. Which views are created depends on if you have enabled live views and history, and if you are using change processing.

See: Dataset architecture in a cloud data warehouse


Visualizations are charts, extensions, and other objects that visualize your data for exploration in a sheet.

Similar terms: Chart

See: Working with visualizations 


Vocabulary is a business logic feature that allows you to add synonyms and custom analyses to Insight Advisor Search and Insight Advisor Chat.

SeeCreating vocabularies for Insight Advisor

Did this page help you?

If you find any issues with this page or its content – a typo, a missing step, or a technical error – let us know how we can improve!