CODEX

Start your data quality journey with this toolkit you already have

Adapt UAT to create trusting, cross-functional partnerships

If you don’t have a data quality program, you can get started now! A useful and familiar toolkit is user acceptance testing (UAT), a standard quality control phase within the software development lifecycle (SDLC). You should have UAT playbooks in-house, and you can adapt UAT to improve data quality.

Everybody wants trustworthy data, but the cross-functional nature of data quality encourages finger-pointing, blame games, and stalemate. Trusting partnerships grow from collaboration, and the UAT framework can help you start collaborating today. Vendor products can help, but nobody understands your data landscape better than your team does.

What is UAT?

UAT provides a framework for partnership between business and technology teams. The whole purpose of UAT is for the business (the “user”) to test the software hands-on before deployment. One or more test scripts are created to confirm that the business requirements were fulfilled.

Each script describes a set of procedures, data, and outcomes and requires collaboration across technology, business, and even analytics. If the data match the expected outcomes, the business “accepts” the software.

Why UAT?

Your most-recent system change, upgrade, or conversion should have a UAT playbook. Go find it!

The UAT toolkit is useful because (1) the script is designed to test data against a reference standard (the business requirements), (2) UAT engages business and technology in a cross-functional partnership, and (3) your team already has sample UAT playbooks — plus experience using them.

UAT sets expectations up front because each test script is reviewed with the cross-functional stakeholders — including technology and the business unit. The expected outcome, data sources, and success criteria are written and agreed upon.

How do I start?

It’s OK to start small. Seek agreement on a few “pain points” in your data environment. Maybe start with an unreliable report that should publish similar numbers each month but doesn’t. Or reduce data-cleaning time for analysts due to inconsistent data entry by customer-contact teams.

Now the partnership really begins. From one of the “pain points”, drill down until you can focus on a few data elements that will deliver value quickly. Ask what is the need for data quality in this process? For example, if you’re looking at the unreliable report, do root-cause analysis as a cross-functional team: together, write a script to follow each step of the data supply chain. Go as far back as necessary to find the source of the data, with the business, technology, and analytic teams represented. Who creates the raw data? Is there transformation of the data before it’s stored? Unpack the process that creates and distributes the report.

Stay focused on creating value together, collaborating to improve the process. Suppose the root cause is manual data entry at the source. Engage the data entry team to help design the solution. A systemic control could improve their efficiency while also reducing defects — but would require support from technology. Cross-functional root-cause analysis helps everyone understand the other teams’ work and roles.

Finally, validate that the solution remediated the defects. Identify any opportunities to avoid these defects in the future — hopefully with more collaboration.

After you’ve completed the first cycle, choose another “pain point” and repeat the process. The learnings from each cycle will build experience, confidence, and trust in the process.

This UAT “hack” is not perfect out-of-the-box. For example, out of your whole data landscape, you have to identify which process to test. (In UAT, you’d only test processes impacted by the new software.) And long term, a holistic data quality program requires focused oversight and dedicated resources.

But small steps lead to bigger steps, just like an exercise plan. Once you start the habit of testing, you’ll build trusting partnerships, and demonstrate the value of improved data quality. Together, data and collaborative partnerships are catalysts for both data quality programs and long-term transformation. Start yours today!

Tom Hirata is the Founder of Data Mandala and developer of the More Meaning and Less Cleaning framework for small-medium businesses to strengthen their data landscape for digital transformation and privacy compliance.

To schedule a Privacy Compliance Check-up call with Tom, where you’ll learn 3 capabilities you must have for consumer privacy regulations like the CCPA and GDPR, click here. You can also message me on LinkedIn.

I’m here to accelerate your data quality journey.