Analytics

Mind the gap: Fixing a broken decision-making process

By Brynne Henn - January 27, 2021

Today, the average employee in a modern company encounters dozens of dashboards, reports, and data-driven discussions every week. Despite the availability of data, the gap between data and decisions continues to be a challenge for even the most advanced enterprises. A recent Gartner survey shows that only half of the decisions businesses make today are influenced by data, and 73% of data leaders report that the data they’re collecting is not having the impact they expected.

In this data-rich environment, every employee, every process, and every decision has the potential to become more data-driven. For key performance indicators—from customer satisfaction metrics to engagement scores to revenue attainment—the path from observing a change in a metric to a data-driven response typically follows a decision-making process that looks like:

  • Observe performance. Every team has metrics they regularly track that indicate how a product or focus area is progressing. Monitored via executive dashboards and weekly business reviews, this is the most passive part of the process, but the real work starts when the “KPI owner” observes a change or adverse trend and requests a deeper diagnosis.
  • Diagnose change. At that point, the data teams swing into action, quickly diagnosing why the change occurred. Was it due to an action the company took, something a customer did, or something the competition did?
  • Recommend action. Decisions based on gut feel are the enemy of a data-driven organization. Ideally at this stage, insights are gathered, presentations made, and a course of action is determined. To ensure the success of this stage of the decision-making process, an analyst needs to be able to rapidly synthesize and communicate the changes they’ve observed and provide an informed course of action to the business.

When each step in this process runs smoothly, each division can be more accountable, more aware of shifts in the business, and more effective in their decision-making. But that’s all in theory. Too frequently, there are gaps in this decision-making process that block data-driven action.

Gaps in analysis lead to gaps in the decision-making process

In reality, the analytics skills needed to run a truly data-driven process are exceedingly rare. Furthermore, while the cost of data access has decreased by over 20x over the past twenty years, the cost to acquire deeper data skills has steadily increased. As a result, every business operator has to become more data literate out of necessity to keep up with data and the bottleneck of trying to scale people to do the job.

Suddenly, every business stakeholder or metric owner is expected to have two jobs — operator and analyst. This leads to three common gaps that slow down, or even break, the process of moving from data to decision making:

  • Missed observations. No matter how many data experts you have on your team, or how many dashboards you build, there’s no way to monitor all these reports and understand at a glance what’s important now. Often, users are unable to cut through the noise to answer which change is significant or understand what’s going on underneath the surface of a metric, leading to gaps in observation.
  • Gaps in diagnoses. When a meaningful change is observed, definitively answering why that change occurred is a slow, manual, and incomplete process. Data analysts (or in many cases determined product leaders) must dig in further, and it can take days, or sometimes weeks, for a useful answer.
  • Delays in action. A delay between observation and diagnosis naturally leads to a delay in action. Even when they’ve diagnosed why, an analyst will still need to connect the answer to possible actions and evaluate which decisions will have the best likely outcome — delaying a decisive response even further.

These delays in a business inevitably create poor, uninformed decisions based on an incomplete understanding of the data. But there are opportunities to augment the rote, repetitive pieces of analysis that often lead to these gaps with machine learning techniques — driving faster, more informed decisions.

Augmented Analytics eliminates the need for everyone to be an analyst

With the advent of more powerful data platforms and advanced machine learning techniques, we have the opportunity to augment this decision-making process with a new class of analytics. These Augmented Analytics platforms do not replace human intuition or ingenuity; rather, they augment our abilities and decision-making acumen by accelerating the rote, routine, and boring parts of analysis.

Augmented Analytics monitors a business user’s core metrics, checks every possible factor to diagnose the factors behind a change, and provides insights backed by data fast enough for business users to act. These facts may influence the decisions being made, but they are not marching orders, and they allow the business operator latitude in interpretation and execution.

With Augmented Analytics, organizations can close the gap between data and decision making. The average business user is freed from watching their dashboards for every significant change, from hours or days spent interrogating the data to ask “why” a given change occurred, from the task of gathering a collection of facts supporting biased hypotheses.

Augmented enterprises rise from Augmented Analytics

In a world where both business users and analysts can quickly understand and act on changes in their data, Augmented Analytics will naturally lead to an augmented enterprise. It will create an organization where every process and every decision is driven by people but informed with data. In turn, the decision process is continually supported by the technology handling the rote, repetitive work of analysis.

The information necessary to guide the thousands of decisions made daily within a modern organization are already captured within its data feeds and warehouses. We just need better tools to scan, filter, and explore this data to more efficiently identify the behaviors that matter, and to highlight them to the true masters of the enterprise universe – their human counterparts. In an augmented enterprise, these emerging leaders can now spend more time on high-level strategy and execution (the business cerebellum), and to make better-informed decisions based on data they already have.

It’s not uncommon for enterprises to have tens of thousands of people using business intelligence tools daily to access data volumes spanning every aspect of the customer lifecycle from acquisition to engagement and retention. The data required to feed such an augmented enterprise is already present, and companies across industries — like Samsung, Kitu Super Coffee, and Housecall Pro — are already using Augmented Analytics tools to build a better, data-driven, augmented enterprise.

Want to close the gap between data and decision and transform your organization into an augmented enterprise? Get in touch with our team to learn more about augmenting and accelerating your analysis with Sisu.


Read more

Solving the Data Analyst’s Dilemma:
Fast answers to more questions in a sea of data

Even with more data, analysts don’t need to burn the midnight oil to answer every question, faster. In this post, Callum breaks down how Sisu helps analysts quickly and confidently answer every question from the business.

Read more

5 Organizational Habits CDOs Must Break to Escape the Analytics Bottleneck

To build a truly data-driven culture that impacts the business and increases its value, CDOs must break the analytics bottleneck and address these five common habits within their organization.

Read more