Overcoming distrust on the path to productive analytics
We outline the importance of overcoming distrust in data and analytics, with tips on how to align all stakeholders, being a data optimist, streamlining the process, and more.
By Dejan Duzevik, Chief Product Officer at Concentric
Where there’s data, there’s often skepticism. With the ability to distribute and access information more easily and at a higher volume than ever before, there’s an understandable need to receive information today with a discerning eye. But in the business world, at what point does distrust in analytics hold an organization back?
Recent research shows that business leaders trust only 38% of customer insights gleaned from analytics and just about one-third (34%) of business operations intel. These leaders receive reports and face a tough decision: do they trust data-driven insights, or do they make a decision based on instinct? Those that move forward with instinct justify ignoring analytics with common excuses like “this report didn’t answer the right question,” and “we don’t have enough data.” Or maybe the report took too long. Maybe the methodology wasn’t right—the list goes on.
With high levels of distrust in analytics, it’s no surprise that businesses around the globe report they aren’t getting a sufficient return on their analytics investment. In North America, only 18% report that they are getting a sufficient ROI on their analytics, and across Europe, the Middle East and Africa, that number is an even-lower 14%. As with any healthy relationship, the first step to success is a foundation of trust. In order to see a return on their analytics investment—and to avoid making expensive decisions with unintended consequences—business leaders must first build trust in data, analytics and the resulting insights.
Here are four key steps to start trusting your data and getting the most out of your investment in analytics:
Align all stakeholders
First, everyone who will benefit from an analytics solution must be aligned and on board with its role in day-to-day operations. Each stakeholder will have a different perspective, and all of their unique needs and concerns must be addressed and met from the get-go. For example, the CFO will be concerned whether the solution is valuable. Business users will want to ensure the solution can deliver growth. And the analytics user, the one who must deliver immediate, trustworthy answers, will want to ensure the solution is valid and easy to use.
Once all members of the team are satisfied with an analytics solution, then a company can move forward in implementing the technology knowing buy-in has been secured across the organization. This is an essential step. In order for an analytics solution to move a company forward, leaders across departments must work together to define strategic business questions and source the information that will ultimately generate insights for the benefit of the full team.
Be a data optimist
More than half (56%) of CEOs express concern about the integrity of data that’s used to drive decision-making. While it’s essential to confirm that the information you enter into a model comes from a trustworthy source, the data doesn’t have to be perfect—or in plenty. A high volume of perfect data isn’t necessary to drive strategic insight and action. While that may have been the case with time-series analysis, forecasting using newer methods like simulation allows companies to do more with less.
With simulation software, you aren’t constrained by the hard data points you have for every input. Instead, you can enter both qualitative and quantitative information, leaning on human intelligence to make some estimates that are later validated for accuracy with observable outcomes. Instead of asking whether data is invalid, consider how valid it actually is. How good is the forecast you generate with the information you already have? Most businesses already have enough collective intelligence within their organization to create a reliable, predictive simulation.
One of the top obstacles to generating actionable insights with analytics is the lack of effective and standard processes. In a recent survey, 40% cited process as the number one obstacle to great analytics. To streamline operations, start by automating connections to data sources where possible, minimizing manual entry and expediting the information gathering process.
Next, standardize modeling and reporting. Per the same survey, almost half (45%) of decision makers point to inadequate skills for interpreting and using analytics among business staff as the barrier to success. When it comes to reporting, support business users with simple workflows and visualizations, providing only the most essential information. Presenting business users with too much information can overwhelm, causing confusion and hindering adoption and trust. These measures will allow your team to generate and absorb more data-driven insights, faster.
Foster a culture of feedback
Finally, to make sure the needs of all stakeholders are being met, foster a culture of feedback. The most successful analytics processes will be powered by business leaders that harness collaboration and welcome all contributing team members to the table. Consider implementing an open forum for discussion about data, analytics and how the process could evolve to better suit your business. Ultimately, whether or not analytics is involved, the more collaborative the process, the smarter the decisions.
With a team fully on board with an analytics process, insights will be delivered in formats and through methodology that’s expected. Reports in familiar formats—ones that answer business questions that were collectively agreed upon as important—are less likely to be greeted by pause and skepticism. Instead, with interdepartmental collaboration, streamlined process and a foundation of trust, businesses will start to see an ROI on their analytics investment, all while reaping the benefits of strategic, data-driven action.
Bio: Dejan Duzevik works on industrializing organizations' analytics: standardizing and speeding up data-driven decisions using Concentric Market, a platform for simulating human behavior.
- On-line and web-based: Analytics, Data Mining, Data Science, Machine Learning education
- Software for Analytics, Data Science, Data Mining, and Machine Learning
- Where did you apply Analytics, Data Science, Machine Learning in 2018?
- Don’t do analysis in a vacuum
- The Analytics Engineer – new role in the data team