Vertrauen in Daten aufbauen: Strategien und Erkenntnisse von Analyse-Profis

Strategy   |   Meredith Boll   |   Feb 16, 2024 TIME TO READ: 5 MINS

As companies strive to be more data-driven, having trust in your data is more important than ever. Data quality is also a costly challenge; Gartner research suggests messy and poor-quality data costs the average organization nearly $13 million per year.*

There is no single solution or strategy to guarantee trusted data in your organization. Our own Chief Data and Analytics Officer, Alan Jacobson, says:

I’m not sure I’ve found a company that has perfect data, with everything neat and tidy, and in the right place.

For most organizations, it’s a journey.

We asked data and analytics experts who play a role in building trust within their organizations to share what they’ve learned and the advice they have for organizations struggling with this common dilemma.

Complex Data and Disparate Sources

There are many reasons for poor quality data that can’t be trusted. But the growing volume and complexity of data — all coming from disparate sources — are common causes.

An example of this is new regulatory requirements for ESG reporting, says James Fernando, Regulatory Reporting Director at Societe Generale Global Solution Centre. “There’s a growing need to work with multiple, heterogeneous data sources and integrate them from different platforms,” says Fernando. “We need to check that the data being produced in different systems is consistent.”

“A growing challenge CFOs are facing is the sources of data are getting more complex,” says   Samir Jaipati, CFO Services Leader at Ernst and Young. “They need to get data from all of these sources into one place and then make sense of it.”

A Shift From ‘Need to Know’ to ‘Need to Share’

Forward-thinking companies are learning the value of getting data into the hands of more people. For them, the days of restricting access to only those who ‘need to know’ are over.

“What I’m seeing is, it’s no longer about giving people access to the smallest amount of data they need to have,” says Jacobson. “I’m now hearing there’s a greater risk of not sharing data across an organization.”

Jaipati agrees. “The whole idea of ‘I own this process, therefore I should own this data’ is gone,” he says. “We need to start sharing more data insights with more people.”

He’s encouraged to see more partnerships and data sharing among teams like supply chain, finance, technology, and customer support. “If you don’t share information with others, they will not be able to help you drive business value.”

Of course, the impact of good or bad quality data is amplified when more people are using it for insights and decision-making — another reason to focus efforts on cleaning up dirty data.

Improving Trust with Automated Analytics

Using manual processes and spreadsheets to analyze large data sets from multiple sources is bound to have data quality breakdowns. Upgrading to a modern environment with advanced analytics solutions helps prevent trust issues by reducing human errors and improving data transparency and consistency.

If you’ve heard of Big Lots, you can imagine how much shipping and distribution is involved in moving discounted goods to their retail stores nationwide. Big Lots uses Alteryx to automate the process of reconciling inconsistencies and ensuring quality.

“It was a challenge for us to match freight invoices with order details because of incomplete and inconsistent carrier data,” said Dan Yokum, VP of Finance and Ecommerce at Big Lots. “With Alteryx, we were able to match and join those disparate files depending on what identify shipment fields were available.”

Trusted Data Tips and Tricks

The experts we talked to agree that achieving data quality perfection isn’t happening anytime soon. In the meantime, they share small tasks and larger strategies you can consider to help improve trusted data in your organization.

  • Look at data freshness date: It’s a simple step but checking the ‘last refreshed date’ when you log onto your dashboard is an easy way to check the recency of the data you’re analyzing.
  • Make it easier to find data: Help build a community of practitioners by making data easier to find. It could be as basic as setting up a Teams or Slack channel to help one another find data.
  • Deliver results: Make sure you’ve established clear targets for your analysis and have a process for measuring toward those goals. Once you’ve achieved success, you’ll turn skeptics into believers.
  • Increase reusability: Building a standard library of reusable workflows improves the efficiency of reporting and trust in the outputs.
  • Develop partnerships: Lack of trust is often the result of siloed knowledge. Developing partnerships between teams such as finance, technology, and data architects promotes the ability to share data insights, challenges, and opportunities.

As analytics becomes more democratized and organizations train new models with their own diverse data, there will be increasing pressure to ensure the data has high integrity for accurate output.

You can learn much more from the experts we talked to about improving data quality in your organization by watching this on-demand webinar: The Power of Trusted Data in Modern Analytics.

*Sakpal, M. (2021, July 14). How to Improve Your Data Quality. Gartner.