Because it’s related to data completeness, it’s great to look at another component of data quality – data accuracy. What determines the accuracy and completeness of data is different. Accuracy is about just how closely data reflects the real-world. Imagine selling tickets for a show in one of the many Fox Theaters, but not including the city or state where that theater is located. All the information about date, time and the seat itself might be captured, but the location is missing. This impacts the sales team’s ability to investigate how customers responded differently based on geography. Since knowing a customer’s location is important to understanding regional demand, the team doesn’t have a full and complete picture of their campaign – the data they do have is accurate, but it is incomplete. Your technical team will benefit from establishing a data quality framework or structured protocol that dictates how data should be collected, processed, stored, and validated.
As a Product Leader, what can you do to ensure the data you present across your portfolio is accurate? You guessed it, start by defining the ideal state of the data – a source of truth that is accepted as accurate. This step alone can be daunting, especially if you are the beginning of a data quality initiative, however remember that accurate data will enable your customers to make informed decisions, reduce their risks, optimize their operations and will be critical to achieving your product growth goals. Here are few of the techniques your team can use to increase the accuracy of the data: validate against trusted sources, sample testing to augment a full data set validation, leverage automated data validation tools and data consistency checks. One particularly powerful approach is to engage your technical team in defining measurable Key Performance Indicators (KPIs) for each type of accuracy technique. This engages the team in the business of data and can be very empowering. Also, don’t expect to implement everything at once. Celebrate each enhancement implemented as it happens.
As your team progresses, it is also critical to support them with the time to invest in robust techniques that improve the integrity and reliability of your data. I’ve found that it can be effective to concurrently add customer-facing features and implement techniques such as a establishing a data audit practice to identify inaccuracies, inconsistencies or anomalies. The same is true of automated validation checks to ensure data is in the correct format. When inconsistencies present themselves, don’t just solve the immediate issue, rather dive deeper and ensure that any systemic problem is resolved at the same time. Establish reconciliation processes when data comes from multiple sources, which will almost always be the case when you’re building out a valuable data solution. Advocate for automated data collection. Rolling up your sleeves to address issues manually is at the heart of a start-up, however it doesn’t scale as your business grows. By implementing these practices, your team can maintain and ensure data accuracy across the product portfolio, which is essential to running an efficient business. Fewer support calls, clear guidelines of what is presented across the portfolio offerings and data accuracy automation enable you to focus on innovation, bringing game-changing solutions to market quickly.