How JBS Built Sticky, Collaborative Applications in Qlik

When it comes to building applications that people actually want to use every day, there's nothing quite like seeing what real customers have created. We recently hosted a webinar featuring three developers from JBS, who shared how they're using Qlik sense write back collaboration to solve real business problems within their Qlik environment.

Instead of the usual product demo, we handed the mic to Shawn Schou, Srujan Kalavagunta, and Scott Nelson from JBS to walk through the applications they've built and use daily. What emerged were three distinct use cases that showcase just how versatile collaborative applications can be when you have the right tools.

Training AI Models With Write Back Tables

Shawn Schou, a JBS Qlik admin who previously worked in procurement, tackled a challenge that many large organizations face: how do you classify millions of materials into meaningful categories?

At JBS, they needed to break down their materials into seven unique categories to track spending patterns year over year. Manually classifying everything was simply impossible. Their solution was an AI tool that predicts material classifications based on descriptions. But like any AI, it needed training.

• The Old Way: Massive Excel files with hundreds of thousands of rows, distributed across 30 analysts, each logging changes in their own spreadsheet. Version control was a nightmare, and ensuring consistency across all those files was nearly impossible.

• The New Way: Shawn built a write back table directly in Qlik where analysts could review the AI's predictions and provide feedback. If the AI correctly predicted the family category and subcategory, analysts only needed to confirm with a single selection. If it was wrong, they could make corrections directly in the app. There was even an "unsure" option to flag materials for specialist review.

The results impressed Shawn and his team. They hit their confidence targets (90% for family, 85% for category, 80% for subcategory) and were able to apply predictions across all their materials, giving them a much clearer picture of their spending patterns.

But Shawn didn't stop there. He also built a feedback system into every procurement app using Vizlib's sheet menu. Users could submit comments, questions, or issues directly within the application, complete with automatic email functionality for follow-up. This eliminated the need for users to track down the right person through a chain of contacts; they could just leave feedback right where they were working.

Managing QVD Files

Here's a use case you probably didn't see coming: using write back collaboration for QVD file management.

Srujan Kalavagunta, IT Analyst at JBS, created what he calls the "Qliktionary": a catalog tool for all the QVD files generated within their Qlik environment. The problem it solved was simple but important: teams would lose track of what data each QVD contained, where it was located, and what it was used for.

The write back table lets users select a QVD file, view all its field names and different versions, and then add or update descriptions directly in the app. Once saved, those descriptions become searchable, making it easy for anyone to find the data they need. The app even provides the exact file path, so users can copy and paste it directly into their file explorer.

Unlike the flashy, end-user-facing analytics dashboards most people think of when they hear "write back collaboration,” it's an unconventional use case. But it's incredibly practical for backend data management and helps teams stay organized as their Qlik environment grows.

KPI Tracking Without the Spreadsheet Chaos

Scott Nelson, Analysis & Billing Supervisor at JBS, built an application addressing a pain point that many managers will recognize, tracking individual KPIs across a team when everyone's using their own spreadsheet.

Previously, each person in their department maintained their own Excel file with their goals and performance metrics. These files would float around, get sent to different managers in different versions, and calculations wouldn't always match up. Come end-of-year review time, reconciling became a headache.

Scott built a centralized KPI tracking application in Qlik with section access so that each person can see their own data and their direct reports, but nothing else. The entire system is built on write back tables and users can input their KPI information either through a traditional table view or through a more user-friendly input form.

Managers can even override calculated results when they know someone's put in exceptional work but the numbers don't quite reflect it (for example, when someone hits 89.98% but clearly deserves credit for 90%).

One standout feature: Scott used Vizlib Finance Report, typically used for P&L statements and balance sheets, to create a KPI tracker with commenting functionality. Users can add notes throughout the year about why certain numbers were high or low, what challenges they faced, or what improvements they made. Instead of trying to remember in December what happened in March, everything is documented right where it matters.

The Unconventional Uses That Make These Applications Stick

What struck us most about these use cases was not only that they worked, but the creativity the JBS team showed by using Vizlib. Vizlib extends Qlik with 40+ visualizations, write back capabilities, and AI-powered analysis business users need.

In our webinar, the JBS team used Vizlib products in ways we don't always see:

• Sheet Menu as a Complete Navigation System: All three developers use the side menu to house filters, keeping them accessible on every sheet without cluttering the interface.

• Finance Report for Non-Financial Data: Scott's KPI dashboard uses Vizlib Finance Report not for financial statements, but for performance tracking with thresholds, variance calculations, and in-line commenting.

• Write Back for Backend Management: Srujan’s QVD catalog shows that collaborative features aren't just for customer-facing dashboards; they're just as valuable for internal data management.

• Form View for Better UX: Rather than having users scroll horizontally across massive tables, Scott's input forms present data in a clean, vertical layout that's much easier to work with.

Reading about these use cases is one thing, but seeing the actual applications and how JBS built them is another experience entirely. Watch the full webinar to see Shawn, Srujan, and Scott walk through their applications, demonstrate the features, and answer questions about their development process.