Smarter, More Powerful, More Automated: ADP’s Path to Enterprise-wide Data Quality

How the Global Leader in Human Capital Management Drives Trust in Data using Anomalo and Databricks

It’s not easy serving 1.1 million clients across 140 countries. 

When 42 million workers are counting on your software to make payroll worldwide, including 1 in 6 American workers covering $3.1 trillion of U.S. payroll and taxes in fiscal year 2024, you have to lead with best-in-class technology, talent, and insights. This is the tremendous responsibility facing Anomalo customer ADP, which joined us last month in our second stop of the Databricks Data + AI World Tour in New York.

As a leading provider of human capital management (HCM) solutions, data is the lifeblood of ADP’s operations. However, the company’s traditional in-house approach to data quality was becoming a roadblock, threatening to stall its ambitious digital transformation initiatives. That was until ADP partnered with Anomalo to automate its enterprise data quality efforts, unlocking new levels of efficiency and insights.

Here, we recap my conversation with Kristin Hlavinka, Director of Enterprise Data Governance at ADP, to share best practices, process insights, and organizational learnings with data and business leaders at the Databricks Data + AI World Tour in New York.

This is the second in a series of updates around the world with Anomalo and Databricks this fall. For more in this series, check out our fireside recap from London with Matt Crawley, Chief Data Officer at Lebara. And be sure to stay tuned for an update from Chicago in November!

The Challenge: Scaling Data Quality in a Data-Driven World

Before Anomalo, ADP faced a common problem plaguing enterprises when it came to data quality: a manual, rules-based approach that simply couldn’t keep up with the pace of data production.

“We had 700 individual data quality checks that we created one by one,” recalls Kristin. “It was a tedious and inefficient process, with data stewards submitting rules, a centralized team programming them, and then going through an iterative review cycle.”

As ADP’s data volume and variety exploded, this manual approach became untenable. The company needed a more scalable and intelligent solution to ensure the reliability and security of its data–a critical requirement for powering its sensitive data-driven business initiatives.

The Solution: Automating Data Quality with Anomalo

Recognizing the limitations of their traditional data quality methods, ADP sought a new approach that could keep pace with their rapidly evolving data landscape. This led Kristin’s team to partner with Anomalo, a comprehensive data monitoring platform that leverages machine learning to automate the detection and resolution of data issues.

“We needed something smarter, more powerful, and more automated,” Kristin explained. “That’s where Databricks and Anomalo come into play.”

By integrating Anomalo’s automated data quality checks with their Databricks environment, ADP was able to scale their data quality efforts exponentially. Within months, they had expanded from 700 manual checks to over 16,000 daily machine learning-powered validations.

The Results: Efficiency, Trust, and Empowered Data Users

ADP’s data quality transformation has yielded significant benefits across the organization:

• Scalable Monitoring: Automated data quality checks have enabled ADP to proactively identify and resolve issues at a scale that would have been impossible with their previous manual approach.
• Increased Efficiency: By reducing the time spent on data quality issues from 70% to 30% of the centralized Data Governance team’s workload, ADP has freed up team resources to focus on change management across the enterprise.
• Federated Governance: ADP has adopted a federated data governance model, empowering data stewards across the organization to manage data quality rules directly within the Anomalo platform.
• Trust in Data: The visibility provided by Anomalo’s integration with the Alation data catalog has empowered ADP business teams to confidently rely on the data they’re accessing. Here, Anomalo also helps drive trust in data by inheriting redacted personally identifiable information (PII) from Alation while detecting anomalous sensitive data–resulting in a streamlined and secure experience for ADP data consumers. A single update is therefore propagated throughout ADP’s data ecosystem.

“We’ve gone from this manual process, one-by-one, where results were only seen by data stewards to automation and powerful machine learning checks,” Kristin said. “It’s much more effective in how we’re able to drive quality at ADP.”

Breaking it Down: Best Practices and Learnings for Data Leaders

ADP’s data quality transformation offers several valuable insights for other data leaders embarking on similar initiatives:

1. Prioritize Automation: As data volumes continue to grow, manual, rules-based approaches to data quality simply cannot scale. Embrace machine learning-powered solutions to keep pace.
2. Empower Data Stewards: Adopt a federated governance model that gives data stewards the tools and autonomy to manage data quality within their domains.
3. Foster Cross-Functional Collaboration: Encourage open communication between data producers and consumers to ensure data quality efforts are aligned with business needs.
4. Start Small, Think Big: Begin by focusing on critical data assets, but maintain a vision for comprehensive data quality management across the enterprise.

“Data quality is an art,” Kristin emphasized. “You have to be patient, iterate, and continuously refine your approach to get the right results.”

Unlocking the Power of Data-Driven Initiatives on Databricks

ADP’s data quality transformation has been a crucial enabler of the company’s broader digital transformation efforts, including the adoption of advanced analytics and AI/ML-powered initiatives with the power of Databricks and Anomalo combined.

“We couldn’t get to where we wanted to go on our data quality journey and unlock the power of our data for our data scientists and our Gen AI initiatives with the traditional data quality rule-by-rule approach,” Kristin said.

By automating data quality at scale, ADP has laid a solid foundation to drive innovation and stay ahead in an increasingly data-centric business landscape. As data leaders, the lessons learned from ADP’s journey can serve as a blueprint for organizations looking to elevate their own data quality practices.

To learn more about our integration with Databricks, connect to Anomalo on Partner Connect. For more information about Anomalo and to explore how data quality will drive your business forward, request a demo.