Qlik Sense

Overview

Qlik Sense is built on an associative engine that indexes every relationship in the loaded data, so unselected and excluded values stay visible rather than disappearing from the view.

That is a genuine analytical difference rather than a styling one: you can see what is not related to a selection, which no other tool in this category surfaces by default.

Its pricing model is the headline change and dates most existing guidance. The cloud product is capacity-based, priced on gigabytes of data for analysis rather than on seats.

From the standard tier upwards it states plainly that there is no cost for additional users, so readership is free and data volume is what you pay for.

That is the opposite exposure to the per-seat tools here: it gets cheaper as the audience grows and more expensive as the data grows.

Application size is a licensed constraint rather than a technical one, rising by tier, which is an unusual thing to have to design around.

Its data model is built in a load script and held in memory, so there is no declarative modelling language; the power is real but it is scripting.

Its 2026 direction is explicitly agentic, with assistants over unstructured documents, agents built on them, and a protocol server included from the entry tier upwards.

Automated machine learning is a separate capability gated to the higher tiers, with only a limited taste below them.

Note two first-party inconsistencies when researching it: its own pages give different customer counts, and the chief executive changed, so older copy naming an interim leader is out of date.

Key features and capabilities

The same headings are used for every bi & dashboards entry, so two tools can be read side by side.

Data modelling and semantic layer

• The data model is built in a load script and held in memory by the associative engine
• There is no declarative modelling language; the power sits in scripting and an expression syntax
• A companion integration product moves relational, software-service, cloud and enterprise-application data in
• Live or direct query to sources exists, but in-memory is the default posture
• Application size is a licensed limit that rises by tier, so model size is a commercial as well as technical question

Charts, dashboards and interactivity

• Interactive sheets with the signature selection states showing related, selected and excluded values
• An extension model and development hub for custom objects
• Public or anonymous access to content is limited to the higher tiers
• Mobile applications exist and are versioned separately
• The exact built-in chart count is not published on the pages read

Calculations, AI and natural language

• Set analysis and an expression language are the calculation surface
• A natural-language assistant answers plain-language questions of the data
• Generative assistants answer over unstructured documents, which can be indexed where they already sit
• No-code automated machine learning is available on the higher tiers only
• Rules-based alerts with custom thresholds

Publishing, embedding and collaboration

• Managed and shared spaces provide governed collaboration from the standard tier
• Scheduled report generation and distribution in document, presentation and spreadsheet formats
• Delivery to email, cloud storage and a common collaboration platform
• Embedding through published interfaces, including into assistants and applications
• Agentic assistants can be embedded into applications and workflows

Security, lineage and certification

• Row and column-level security is applied through a section of the load script rather than a settings screen
• Managed spaces hold certified content separately from personal work
• Data lineage connectors are limited to the higher tiers
• Administration tooling varies by edition
• Version control and deployment pipeline specifics were not confirmed in this pass

Integrations

• Hundreds of data sources through standard connectors
• A companion product covers relational, software-service, enterprise-application and legacy sources
• Extraction from one major enterprise application suite is limited to the higher tiers
• A protocol server for assistants is included from the entry tier upwards
• A no-code automation builder triggers actions across systems, alongside an open published interface

Where it runs and what it costs

• Software as a service on one major cloud, with the region chosen at tenant creation
• Seven named regions across three continents, plus a separate government tenant in the United States
• Client-managed deployment on your own Windows infrastructure is available through sales
• Multi-region tenants are limited to the top tier
• The unit of pricing is gigabytes of data for analysis per year, described as capacity rather than consumption

Pricing

SubscriptionFrom $300 per month for 10 GB

Capacity-based rather than per seat, which is the thing to understand before comparing it with anything else here. Published tiers start at $300 a month for ten gigabytes of data for analysis and ten users, then $825 for twenty- five gigabytes, then $2,750 for fifty, with the top tier quoted on request from two hundred and fifty gigabytes. From the second tier upwards there is no charge for additional users at all, so a wide readership is free and data volume is the meter. Application size limits rise with the tier.

Vendor pricing page →

Demos and videos

• Semantic understanding demo Product demo · YouTube
• Cloud analytics capabilities and getting started Webinar or talk · YouTube

About QlikTech International

QlikTech International is a private company whose copyright line dates from 1993, with Swedish origins and offices including Philadelphia and San Francisco. It has been majority-owned by a private equity firm since being taken private in 2016, with a later significant investment led by a sovereign wealth fund alongside that owner. Its chief executive is Saugata Saha, so copy naming an interim leader is out of date; the wider leadership team and board are published. Revenue and employee count are not published, and its own pages disagree on customer count, saying more than thirty thousand on one page and over forty thousand on another, so cite whichever you use. Several acquisitions are visible in the product navigation, though their dates and terms were not confirmed in this pass.

Founded 1993 · qlik.com