
We built three sports activations at Qlik Connect 2026, streamed the action into Apache Iceberg on Qlik Open Lakehouse, and turned every attendee’s stats into a personalized AI trading card, using an open, decoupled architecture designed to scale performance while tightly controlling cloud cost.
Qlik Connect 2026 in Kissimmee wasn’t just keynotes and coffee. Down the hall from the mainstage, the Qlik® Games were in full swing, and what looked like a show floor competition was something more important: a live, public demonstration of how modern data architectures perform under pressure.
The buzz on the floor
Three zones. Three sports. One very competitive crowd. The Qlik Cycle Sprint had you out of breath in seconds. The Qlik Slap Shot saw execs trade blazers for hockey sticks. The Qlik Precision Par turned a corner of the convention centre into a short-game clinic.
Every sprint, slap, and swing was captured live, streamed into Qlik Open Lakehouse, and pushed back to players on leaderboards they refreshed in real time. Big screens in each zone showed the live standings, the most recent attempts, and the telemetry behind them, all driven by Qlik Cloud Analytics reading from Qlik Open Lakehouse using Amazon Athena as our query engine.
At the AI Trading Card station next door, Qlik Automate pulled your scores the moment you finished, paired them with an AI-generated action shot created from athlete’s own headshots, and printed you a keepsake before the next attendee had finished warming up. The line for the Qlik Games did not get shorter all week.

The customers behind the games
Each sport was anchored by a real Qlik customer, not as a logo on a backdrop, but as a partner in how we help turn athletic performance into data-driven performance. We picked these three intentionally. Cycling, hockey, and golf produce three very different data shapes: continuous rider telemetry, event-based game sequences, and high-volume discrete swings. And every one is the kind of challenge Qlik handles for performance teams around the world.
Pinarello Q36.5 Pro Cycling Team use Qlik to analyse historic rider performance data to identify the most capable riders to sign and then for rider selection for upcoming races to maximise their chance of success on the road.
Malmö Redhawks, with the help of our partners at Stretch Qonnect, use Qlik to transform game-by-game events into the insights their coaching staff and front office rely on to build winning squads.
Topgolf uses Qlik to turn millions of swings into operational and guest-experience decisions at a scale few venues can match.
Three very different businesses, all leaning on the same pattern: unify the data, make it fast to analyse, and put it in front of the people making the calls.
The part the data leaders leaned in for
Here’s where the room got quiet. Every data point from the Qlik Games — every pedal stroke, every shot, every metric printed on every trading card — landed in Qlik Open Lakehouse. One open storage layer, readable by any query engine and visualised in Qlik Cloud Analytics, with no proprietary format and no lock-in.

Let’s go through the Architecture diagram above – it shows the full power of Qlik’s portfolio working together.
Sensor and score data from each zone streamed in through a combination of Qlik Open Lakehouse Streaming Ingestion, Talend API and Application Integration and CDC via Qlik Talend Cloud Pipelines, and landed it as Apache Iceberg tables in Amazon S3.
Qlik’s Adaptive Optimiser worked continuously to ensure the Iceberg tables were always providing the best possible query performance at the lowest possible cost.
Qlik Cloud Analytics was connected to those tables through Amazon Athena to power the interactive leaderboards.
There was no second copy of the data, no proprietary warehouse tier, no re-ingest. When the doors closed each evening, the compute scaled down. When they opened the next morning, it scaled back up.
That’s the pattern you get when storage and compute are fully decoupled and the format is open.
What this means for cost at scale
The cost story is not an add-on, it is a direct outcome of the architecture.
For Qlik Connect 2026, we partnered with AWS to run the Qlik Games infrastructure. Here is the cost breakdown:

The result was hard to ignore. We ran a live, multi-zone, crowd-driven data experience for two days on a cloud budget of just $3.02, less than the price of a single takeaway coffee.
That covered streaming ingestion and transformation into Qlik Open Lakehouse, optimisation compute, object storage, and querying the lakehouse to power the leaderboards.
Behind that number was real activity at scale, telemetry from two bikes streamed at four records per second throughout the event, data from 3,350 hockey shots, extensive metrics from 1,117 golf swings, all from 457 participants over the 696 games played in total.
The point is not that the Qlik Games were inexpensive. The point is that cost stayed predictable under live, variable demand.
And in the world of AI, that matters. Businesses need to support more users, more queries, more experimentation, and more AI-driven applications without duplicating data, locking into proprietary layers, or watching cloud costs grow out of control.
The Qlik Games made that visible. They showed how open storage, decoupled compute, and interoperable tooling can help organisations scale performance while keeping cost under control.
See you at the next one
The Qlik Games weren’t a gimmick. They were a live demonstration of what happens when open data architecture meets great storytelling. The same pattern (open storage, decoupled compute, interoperable tooling) is what we’re helping customers put in place to keep control of their data estate and their cloud bill.
Want to see what Qlik Open Lakehouse could do for your data? Let’s carry the conversation on before the next Qlik Connect. And if you participated in this year’s Qlik Games, take a minute to see where you finished on our leaderboards and start practising for next time. I will be working on my slap shots for sure!










