Skip to main content
••24 min read•By Nathan Ibgui, fractional CIO

Qlik Sense vs Power BI: pay for readers or for gigabytes to run a small business on data in 2026?

Qlik or Power BI: one charges for readers, the other for uncompressed gigabytes. Costs from 10 to 150 people, and the traps of Qlik capacity in 2026.

Qlik is often the name you hear in a mid-sized manufacturer or distributor, sometimes for more than ten years, when Power BI arrives with Microsoft 365. Both build interactive dashboards, and comparisons usually pit them against each other on usability or charts. On October 5, 2026 we reread Qlik’s pricing page and the documentation behind it and found a much more structural difference: Power BI charges you for people, Qlik charges you for gigabytes. Beyond the entry plan, Qlik doesn’t count users at all. The whole calculation changes nature, and that’s what decides it.

This article compares Qlik Cloud Analytics (the online service that hosts Qlik Sense apps) and Microsoft Power BI from the point of view of a company with 10 to 250 employees that wants to track its sales, margin, inventory or production. Sources: the qlik.com pricing page (French and US versions, US version rechecked in November 2026), the Qlik Cloud online help (capacity subscriptions, the “Data for Analysis” meter, options by edition, overages), the Power BI pricing page (microsoft.com, French version at the time of writing, US version rechecked in November 2026) and Microsoft Learn. This is the fifth and final head-to-head in our series on reporting and business intelligence tools, after Power BI vs Looker Studio, Power BI vs Tableau, Excel vs Power BI and Metabase vs Power BI.

A note on method: no subscription was taken out, no price copied from a third-party comparison. Qlik publishes its prices in dollars, excluding tax and billed annually; Power BI prices are Microsoft’s US list prices, paid yearly. All prices are as of November 2026 (the French version of this article converted Qlik’s dollar prices into euros at the ECB rate of October 2, 2026, €1 = $1.1225; the chart keeps those euro figures). These are “starting at” prices: any discount is negotiated with the sales rep.

Two meters: people on one side, gigabytes on the other

Power BI counts people. Power BI Desktop, the authoring tool, is free on Windows. As soon as a report needs to be shared, every author and every reader needs a Pro license, at $14 per month paid yearly, or Premium Per User (PPU), at $24. Viewing without a license only exists from a Fabric F64 or Premium P1 capacity, out of reach for a small business. Ten more readers means ten more licenses.

Qlik counts data. The Qlik Cloud Analytics pricing page lists four plans. Starter starts at $300 a month for 10 users and 10 GB of data, with a note: additional user licenses are available for purchase, but “no additional data capacity available for purchase”. Standard starts at $825 a month for 25 GB of data, and its list of benefits begins with “no cost for additional users”. Premium starts at $2,750 a month for 50 GB. Enterprise is on quote, from 250 GB. The documentation confirms it: only the Starter plan is priced by number of users; the others are “capacity-based” subscriptions, measured by a meter called Data for Analysis.

Qlik’s FAQ presents this model like a cell phone plan: a fixed amount for a defined capacity, as opposed to an electricity bill that varies with consumption. The image is right for one reason and misleading for another. Right, because at constant volume, adding fifty readers costs nothing. Misleading, because capacity is measured in a way few people would guess, which we detail below.

So the first question to ask isn’t “how many users?”, but “how many people will read, and how much data will they read?”. A 40-person company that wants to open dashboards to all its sales reps, shop-floor supervisors and executives isn’t in the same situation as a 200-person company where only 12 managers look at the reports.

What drives the bill on each side Power BI: the meter counts people Power BI Desktop free on Windows: build, but don’t share Pro: $14 per person per month every author AND every reader PPU: $24 per person per month bigger models, 48 refreshes a day +10 readers = +10 licenses or +$1,680 a year on Pro License-free viewing: F64 / P1 capacity only Qlik Cloud: the meter counts GB Starter: $300/month 10 users, 10 GB FIXED (no extension) Standard: $825/month 25 GB, additional users at no cost Premium: $2,750/month 50 GB, then in 25 or 250 GB packs Enterprise: on quote from 250 GB US list prices as of November 2026, excl. tax, billed annually
Sources: Power BI pricing page (microsoft.com/en-us; French page: microsoft.com/fr-fr) and Qlik Cloud Analytics pricing page (qlik.com/us; French page: qlik.com/fr-fr), US prices as of November 2026. “Starting at” prices.

Two engines: associative versus star schema

Before price, there’s a difference in use that anyone who has worked with both tools recognizes. Qlik is built on a so-called “associative” engine. All the data loaded into an app is linked together in memory. When you click a customer, a region or a month, every list on the screen shows both the values associated with your selection and those that are excluded. It’s especially telling for exploration: you immediately see which products have not been sold to that customer, or which suppliers haven’t delivered a single order this quarter. Data loading goes through a script, which you write or generate with the wizard.

Power BI is built on a tabular model. You prepare the data in Power Query, link the tables (ideally in a star schema: facts in the middle, dimensions around them), then write measures in DAX. Cross-filtering between visuals exists, but it follows the relationships you defined. The approach is more demanding to design and more predictable to use. It also has a decisive advantage in a small business: it’s the same technology as Excel’s data model, as we showed in Excel vs Power BI. The skills often already exist in-house, or are quick to acquire.

In practice, Qlik appeals to financial controllers and supply chain managers who dig through data without knowing in advance what they’re looking for. Power BI appeals to teams that publish stable metrics to many readers, in the Microsoft 365 environment they already use. Neither is “better” in absolute terms: the question is which of these two uses dominates in your company.

Criterion Qlik Cloud Analytics Power BI
Billing unit Users (Starter), then data volume (Standard and up) Person (Pro, PPU), or Fabric capacity for very large deployments
Published entry price $300/month for 10 users (Starter), billed annually $14 per person per month (Pro), paid yearly
Additional readers Starter: price not shown; Standard and up: no cost One license per reader (except with F64 / P1 capacity)
Authoring tool In the browser (Qlik Sense in Qlik Cloud) Power BI Desktop, free, Windows only
Engine Associative, in-memory, script-based loading Tabular, Power Query + DAX, shared with Excel
Maximum size 5 GB app (Starter), 10 GB (Premium), 15 GB by default (Enterprise) 1 GB model (Pro), 100 GB (PPU)
Hosting regions North America, Germany, Ireland, United Kingdom, Australia, Japan, Singapore Home region of the Microsoft 365 tenant
Hosting on your own servers “Client-managed” version available, through sales Power BI Report Server, tied to specific licenses

What a Qlik gigabyte really counts

The pricing page says “data for analysis” and suggests a simple estimation method: estimate the size of the datasets you plan to analyze over a year. The help page dedicated to the meter Data for Analysis, updated on October 2, 2026, is much more precise. Five rules change the estimate completely.

1. Uncompressed size is what counts. The documentation says it in black and white: loaded data is counted at its uncompressed size when it enters Qlik Cloud. Many databases store their tables compressed. If your admin tells you “the ERP database is 8 GB”, that figure says nothing about what Qlik will count. Qlik itself suggests the right method: export the relevant tables to text or CSV and measure the resulting files.

2. The monthly peak is what gets billed. Every day, the system adds up stored files, data ingested by reloads and data created. The highest day of the month becomes the “monthly peak”, which is compared with the capacity purchased. A single day of full reloads at month-end is enough to set the figure for the whole month.

3. An app that doesn’t reload keeps counting. The volume ingested during the last reload is carried over every day, as long as the app exists and no smaller reload replaces it. An old app forgotten in a shared space weighs on capacity indefinitely.

4. The same data loaded twice counts twice. If two apps each query the ERP sales table, the table is counted twice. The fix, documented by Qlik: load the data once into QVD files (Qlik’s native format), then have every app read those files. Data loaded once into a tenant and reused is only counted once. It’s an architecture question, not a setting: it has to be decided with the very first app.

5. Some things don’t count. Apps and data in each user’s personal space, apps that only read another app or QVDs already stored, apps published to a managed space (only their reloads count) and data queried live (Direct Query) are excluded from the meter.

For a small business, the consequence is simple: the volume to buy depends as much on how the apps are built as on the size of the data. Two teams analyzing exactly the same numbers can consume 12 GB or 30 GB depending on whether they set up a shared QVD layer. It’s the same logic we advocate in our guide toERP data integration: a common layer, fed once, read by all.

Same numbers, two architectures, two bills (illustrative example) Each app loads the ERP ERP: sales 6 GB uncompressed Margin app Sales team app Leadership app Meter: 3 × 6 = 18 GB each load is counted separately One QVD layer, read by all ERP: sales 6 GB uncompressed QVD files loaded only once Margin app Sales team app Leadership app Meter: 6 GB apps that read QVDs don’t count Rules: Qlik Cloud help, “Data for Analysis”, updated October 2, 2026. 6 GB volume chosen for the example.
The “loaded twice, counted twice” rule and the exclusion of apps that only read QVDs are described in the Qlik Cloud help. The 6 GB volume is an example, not a measurement.

Starter: the small-business plan with no safety valve

The Starter plan is pitched as “cutting-edge analytics within reach of small businesses”. It’s priced like it: $300 a month, or $3,600 a year for 10 users. It includes interactive dashboards, access to Qlik’s answer agents and MCP server, automated report generation and distribution (up to 250 according to the documentation) and a no-code automation builder. Three limits to know before signing.

Capacity is fixed, and going over blocks you. The FAQ on the pricing page is reassuring: if you exceed capacity, “you keep full access”, and extra charges apply beyond it. But the documentation on overages specifies that this rule doesn’t apply to Starter. Its 10 GB capacity is fixed; once the limit is reached, you can no longer import or reload apps, or add files. To free up space, you have to delete apps or files; to get more, change plans. In practice, dashboards that no longer reload show numbers frozen at the last successful reload.

The price of the eleventh user is not published. The page says additional licenses are available for purchase, without giving the price, in either the French or the US version. A business with 15 readers therefore can’t price Starter without going through a sales rep.

App size is capped at 5 GB. That’s comfortable for most small-business uses, but a Qlik app lives in memory: a line-level sales history over several years can get close.

Starter is therefore a good fit for a team of 5 to 10 analysts and managers working on a controlled data scope. Beyond that, the next step isn’t “a little more expensive”: Standard costs 2.75 times as much as Starter.

The math: from how many readers does Qlik become cheaper?

Since the two tools don’t count the same thing, the comparison has to be made on a scenario. Assumptions: data needs that fit within the included capacity (10 GB for Starter, 25 GB for Standard), Power BI Pro licenses for everyone, published prices before discounts (US list prices as of November 2026). Over a year, Power BI Pro costs $168 per person. Qlik Standard costs $9,900 a year, whatever the number of readers.

Readers and authors Power BI Pro Qlik Starter Qlik Standard (25 GB)
10 people $1,680 $3,600 $9,900
25 people $4,200 can’t be priced (price of licenses beyond 10 not published) $9,900
60 people $10,080 can’t be priced $9,900
100 people $16,800 can’t be priced $9,900
150 people $25,200 can’t be priced $9,900

Three takeaways. Up to 10 people, Power BI costs less than half as much as Starter ($1,680 versus $3,600). Qlik Standard becomes cheaper than Power BI Pro from 59 readers, provided 25 GB is enough (at 60 people, $9,900 versus $10,080). And if your models exceed 1 GB or you need more than 8 refreshes a day, Power BI moves to PPU ($288 per person per year): Qlik Standard then becomes cheaper from 35 people. (With the French prices in euros, the thresholds were 61 and 36.)

The reasoning flips if the data grows. Each extra 25 GB block is paid for, and Premium ($2,750 a month, or $33,000 a year for 50 GB) is only justified with hundreds of readers or specific features: SAP extraction, machine learning, anonymous access. For a small business, the real contest is between Power BI Pro and Qlik Standard, and it comes down to two numbers only you know: the number of readers and the uncompressed volume.

Annual cost excl. tax by headcount, France list prices in € (data within included capacity) 0 €10k €20k 1,452 3,207 10 3,630 8,820 25 8,712 8,820 60 14,520 8,820 100 21,780 8,820 150 people (authors + readers) Power BI Pro Qlik Starter (10 users) Qlik Standard (25 GB, readers at no cost)
Calculation behind the chart (France list prices): Power BI Pro €145.20 excl. VAT per person per year; Qlik Starter $3,600 a year for 10 users; Qlik Standard $9,900 a year for 25 GB; €1 = $1.1225 (ECB, October 2, 2026). US list prices: Power BI Pro $168 per person per year. Published prices, before negotiated discounts.

Overage: flexibility you pay for every month

On Standard, Premium and Enterprise, going over capacity doesn’t cut off service. Qlik’s documentation calls it “overage”: usage can continue up to ten times the purchased capacity, and the overage is calculated and billed every month. Notifications appear in the admin console at 80%, 90% and 100% of capacity; the account owner receives an email and “may be contacted by Qlik” about an add-on offer.

It’s comfortable: a dashboard never stops because of an exceptional load. But it qualifies the promise of “predictable cost” highlighted on the pricing page. The plan is predictable as long as you stay within it; beyond that, the bill follows the monthly peak, and that peak is measured in uncompressed data, with the carry-overs from one app to another described above. The meter, at least, is visible: Qlik provides an app for detailed consumption monitoring. Assign someone to check it every month, just as you would monitor a company mobile data plan.

On the Power BI side, going over takes another form: a model larger than 1 GB can no longer be published on Pro, and a ninth scheduled refresh in a day isn’t possible. There’s no surprise bill, but there is a step: move the whole team to PPU, or move to a Fabric capacity. So both tools have a safety valve; it just isn’t in the same place.

Hosting, security and vendor

Where is the data? Qlik’s FAQ lists seven regions to choose from: North America, Germany, Ireland, United Kingdom, Australia, Japan and Singapore. There’s no France region, but there are two regions in the European Union. The platform is hosted on AWS, and Qlik states that its generative AI assistant processes requests in the same AWS region as the data. For Power BI, data resides in the home region of your Microsoft 365 tenant. If your customers or your industry require specific hosting, ask the question in writing in both cases, and record the answer in your data processing records (the GDPR record of processing activities, if you operate in the EU).

Who sees what? Both tools can restrict the rows each person sees, but not in the same way. At Qlik, the long-standing mechanism is called section access: it’s written into the app’s load script, which reduces the data to what each user is allowed to see. In Power BI, row-level security (RLS) is defined through roles in the model; Microsoft Learn specifies that it doesn’t apply to workspace members with the Admin, Member or Contributor role. In both cases, it’s design work, not a checkbox: test it with a real sales rep account before opening the dashboards.

Who is the vendor? Qlik is a company of Swedish origin, owned since 2016 by the US private equity firm Thoma Bravo; Qlik acquired Talend in 2023, hence the name “Qlik Talend Cloud” for its data integration line. In the French company registry, you find QLIKTECH FRANCE, an SARL registered in 2007 in Suresnes, and the parent company QLIKTECH INTERNATIONAL AB, registered as a foreign company. Microsoft operates in France through its Issy-les-Moulineaux subsidiary. So both vendors have a local contact in France; the difference will rather come from the integration partner who supports you.

What about hosting on your own servers? Qlik still offers a version installed at the customer’s site (“client-managed”), but it isn’t in the public price list: the pricing page points to sales. If it’s a requirement, Qlik remains one of the few major BI tools that allow it, just like Power BI Report Server on the Microsoft side. Our comparison on the cost of an ERP migration to the cloud covers what to price before choosing between on-premises and hosted.

When to choose Qlik, when to choose Power BI

Choose Power BI if you’re on Microsoft 365, if you have fewer than about sixty readers, if part of your reporting still lives in Excel, or if you want to be able to hire a skilled profile easily. That’s the case for the vast majority of businesses with 10 to 100 employees. The entry cost is low, the authoring tool is free, and the learning curve builds on Excel.

Choose Qlik Standard if you want to open dashboards to many people (shop floors, stores, sales force) without paying per head, if your data stays within a controlled volume, and if your users explore more than they consult. It’s also the logical choice if Qlik is already in place and the existing apps work: migrating to Power BI costs a whole rebuild project, which has to be weighed against the license savings.

Choose Qlik Starter for a small team of analysts and managers (10 people at most) that needs associative exploration, while keeping a close eye on the fixed 10 GB.

Be wary of both if nobody in the company has time to model the data. Neither Qlik nor Power BI fixes poorly structured data in the ERP; they display it. A reporting project always starts with data quality, as withcash flow optimization based on ERP data.

The method in five steps

1. Count people, not licenses. List who creates and who reads, today and in two years. Below about 60 readers, Power BI Pro is almost always cheaper; above that, Qlik Standard deserves a quote.

2. Measure the uncompressed volume. Export to CSV the tables the dashboards will use (sales, purchasing, inventory, production) over the history depth you want, and add up the file sizes. That’s the method Qlik suggests, and it’s the only number that matters.

3. Decide on the architecture before the first app. On the Qlik side, a shared QVD file layer avoids counting the same data twice. On the Power BI side, shared semantic models avoid multiplying refreshes. In both cases, the same principle: load once, read everywhere.

4. Ask for the three numbers missing from the page. At Qlik: the price of a Starter user beyond 10, the price of a 25 GB block and the overage rate. At Microsoft: nothing is hidden, but check whether your Microsoft 365 E5 licenses already include Power BI Pro.

5. Prototype the same dashboard in both. Qlik offers a free trial of its cloud plan, and Power BI Desktop is free. Rebuild the most-read metric, show it to its readers, and check in the Qlik console how many gigabytes the prototype consumed.

If you’d like an outside view of this choice, it’s the kind of assessment we carry out in our engagements on reporting and ERP data, or as part of a fractional CIO engagement.

Key takeaways from the “reporting tools” series

Five head-to-heads, five different ways of counting what you pay for. Looker Studio, now Data Studio, is free to read but the report belongs to its author. Tableau shows a teaser price for a role you can’t buy on its own. Excel doesn’t have the limit people attribute to it. Metabase is free until you connect Microsoft 365. And Qlik doesn’t charge for people, but for uncompressed gigabytes at the monthly peak.

The common lesson fits in one sentence: before comparing prices, compare the units. A BI tool is chosen by counting its readers, measuring its data and reading the licensing documentation, not just the pricing page. To place reporting within your overall software stack, see our overview of digital transformation tools for small businesses.

Frequently asked questions

How much does Qlik Sense cost for a small business?

Qlik Cloud Analytics Starter starts at $300 a month, billed annually, for 10 users and 10 GB of data, or $3,600 a year. The Standard plan starts at $825 a month for 25 GB of data, with no cost for additional users, or $9,900 a year (US list prices as of November 2026).

Is Qlik cheaper than Power BI?

Not for a small team: with 10 people, Power BI Pro costs $1,680 a year versus $3,600 for Qlik Starter. Qlik Standard becomes cheaper than Power BI Pro from about 59 readers, and from 35 readers against Power BI PPU, provided the data fits in 25 GB (US list prices as of November 2026).

How does Qlik measure “data for analysis”?

According to the Qlik Cloud help, it’s the uncompressed size of the loaded data, plus stored files and data created. The total is calculated every day and the highest day of the month is compared with the purchased capacity. The same data loaded into two apps counts twice, unless it goes through shared QVD files.

What happens if you exceed Qlik capacity?

On Standard, Premium and Enterprise, the service continues up to ten times the purchased capacity and the overage is billed every month. On Starter, the 10 GB capacity is fixed: once it’s reached, you can no longer reload or import apps; you have to delete content or change plans.

Where is Qlik Cloud data hosted?

Qlik offers seven regions to choose from: North America, Germany, Ireland, United Kingdom, Australia, Japan and Singapore, on AWS infrastructure. There’s no France region, but there are two regions in the European Union. A version installed on your own servers exists, on request through sales.

In short

Qlik and Power BI aren’t opposed first on charts, but on what they count. Power BI charges every person who reads a report; Qlik, beyond its Starter plan, charges for the uncompressed data volume at the monthly peak, and lets as many people read as you like. The first is unbeatable for a small business with fewer than about sixty readers on Microsoft 365; the second becomes rational when you want to open reporting to the whole company on a controlled data volume.

Before signing, measure your data in CSV, count your readers and decide on the loading architecture: those three numbers, not the pricing page, will give you the real cost. And if you’d like to talk it over, contact us.

Nathan Ibgui, IT & AI consultant

Written by

Nathan Ibgui

Independent consultant and fractional CIO, I help small and mid-sized businesses with IT strategy, artificial intelligence, automation and cybersecurity — from assessment to execution.

Ready to transform your business?

Describe your needs in a few lines: I reply within 24 hours with an initial analysis and concrete recommendations.