IGT is a global leader in gaming technology and lottery solutions, providing innovative products and services to regulated gaming markets worldwide. As a public company operating in a highly regulated industry, IGT's internal audit department plays a critical role in ensuring compliance, risk management, and operational integrity across the organization.
Challenge:
Solution:
Results:
Atanas Bosakov supports IGT's internal audit team with data analytics and automation. When he joined, the global lottery and gaming company had a fragmented analytics environment. Auditors relied heavily on Excel and out-dated tools like ACL, with limited automation or integration capabilities.
The internal audit team struggled with:
"In internal audit, there are so many different needs and audits, and there are plenty of opportunities to provide diverse outputs," Atanas explained.
Reproducibility was particularly critical: "One of the big requirements in internal audits is to have reproducible work, so we can actually audit our own audits and show that it's actually verifiable."
While they had various data sources including Excel files, structured and unstructured databases, and API connections, there was "a big void in between" their data inputs and desired outputs.
“When I joined the internal audit profession, I couldn't believe that people were using tools like ACL. It's just as painful as getting the injury.” Atanas Bosakov, Analytics & Automation Manager, IGT
The lack of a single tool that could integrate data from all their sources created a fundamental barrier to achieving continuous auditing. "If you think about it, if you don't have a continuous data pipeline in an integrated fashion, how in the world are you gonna do continuous auditing?" Atanas pointed out.
While he had experience with statistical tools like R and Python, writing code was time-consuming and difficult to modify when requirements changed, making it unsuitable for the dynamic nature of audit work.
The department needed to move toward reproducible, end-to-end automation to support continuous auditing. Atanas had KNIME in his “digital back pocket” from previous experience and saw an opportunity to introduce it at IGT.
Atanas introduced KNIME Analytics Platform early in his tenure. KNIME allowed the team to:
"It enabled me to really start day one with the tool and showcase the power of workflows," Atanas explained.
The visual workflow approach immediately demonstrated value through practical proof-of-concept projects that showed management the potential for automation.
As success became evident, the team also adopted KNIME Business Hub. This enabled:
“KNIME Hub was actually a fundamental part of our success story because we were able to put things on the schedule. We can also freely distribute some of the workflows and enable collaboration within the internal audit team, and also not only within the internal audit team, but outside by partnering with various business units.”Atanas Bosakov, Analytics & Automation Manager, IGT
The KNIME Business Hub implementation addressed critical compliance requirements through version control and testing procedures:
IGT's KNIME implementation centered around connecting their primary data source, SAP, to multiple output formats. Working with KNIME partner DVW, they implemented SAP connectors for both KNIME Analytics Platform and KNIME Business Hub, enabling end-to-end automation.
Their workflows handled complex analytical requirements including Benford's law analysis for anomaly detection. "We can efficiently query more than eleven million records, go through the data wrangling," Atanas explained.
When requests came in for journal entry analysis, he queried the data and saved the workflow knowing there will be follow-up requests.
For Benford's law analysis, which helps capture potential fraud through anomaly detection, they can "hit all the cycles, do additional checks and also write not only the output but we actually save the input files after we query the SAP databases. So this way we can have full reproducibility."
KNIME’s integration with Tableau was especially valuable: “We can query our data very efficiently… and we write to Tableau Server but also have local backups,” Atanas shared.
KNIME also handled sophisticated statistical analysis including stratified proportional random sampling. Using flow variables, they orchestrated everything dynamically, satisfying external auditor requirements for specific sample sizes while maintaining reproducibility through automated file labeling and documentation.
IGT is using KNIME not just for classic analytics, but also to evaluate local LLMs (like LLaMA and DeepSeek) for internal use. Workflows were built to:
“We are literally using KNIME to see which model serves us best," Atanas explained.
Beyond advanced analytics, KNIME helped solve surprisingly common and impactful issues like Excel lookups that fail when duplicate entries are present. With KNIME:
These small but meaningful improvements reinforced KNIME’s role as a collaborative platform between audit and business users.
“If I take the most painful use case that we used to run using other tools, we were able to do it about eighty-five percent faster in KNIME.”Atanas Bosakov, Analytics & Automation Manager, IGT
By enabling end-to-end reproducibility, smooth integration, and collaborative development, KNIME has empowered the team to meet regulatory expectations, deliver more projects, and do so faster and with greater confidence. Some of the key outcomes include:
Workflows like anomaly detection (e.g., Benford’s Law on 11M+ SAP records) and proportional sampling became dynamic and reproducible.
“If a picture is worth a thousand words, a KNIME component is literally worth a thousand lines of code,” Atanas said. This flexibility helped bridge expert code and visual analytics.
The ability to integrate R and Python within KNIME also helped IGT preserve past investments:
"We wrote a bunch of special functions in R and are they lost? No. We can actually import them into KNIME because it allows for that level of integration and really takes advantage of the custom functions that are specific to our department and our data analytics needs."
Atanas concluded, "Going back to our initial focal point, a variety of output data sources, but the ability to also query data and achieve this end to end automation has enabled us not only to perform some business analytics, but also to perform more analytics in terms of the frequency and how many projects we execute. So it's been a very empowering journey with KNIME."
Learn more about KNIME Business Hub and how it can help your organization or schedule a demo with our customer care team to see how it works in practice.