By Derek Slater
With big data driving known disrupters, such as Netflix and Uber, more and more organizations are jumping into analytics.
The proof is in the research. Over the next 12 to 16 months, companies will prioritize business intelligence and analytics above almost all other technology-driven strategic initiatives. This is according to 633 IT decision makers recently surveyed by Enterprise Strategy Group (ESG).
So-called optimization projects, says ESG senior analyst Nik Rouda, comprise a critical first step for many businesses breaking into big data.
| Related: Alleviating Stress Through Analytics
For example, freight specialists, such as UPS or FedEx, use optimization. “Route optimization—just using less gas, getting more deliveries per driver, per truck, per day,” he says, “can have an impact in the millions of dollars.”
That kind of increased operations efficiency, he says, “is probably the number one use case for analytics right now.”
But optimization isn’t the end game. Business transformation is. And creating a transformative analytics program requires support, focus, and action from the very top of the organization.
| Related: Using Data to Build Growth and Revenue
Here are five steps for bringing your analytics programs to the next level. According to Rouda, these steps take an organization from optimization to disruption. Then, ultimately, on to true transformation.
- Build the habit of asking data-oriented questions:
Projects that transform companies and even industries often start with a simple question: “What do the numbers tell us?”
Some companies “make the mistake of leaving this up to the data scientists,” Rouda says.
Leadership teams that develop the habit of asking this question, set themselves up for success. From there, you want to spread that habit across the organization.
Rouda also notes that by asking more data-focused questions, you get a cycle of collecting and integrating more data. In turn, creating more opportunities for transformative insight down the road.
- Make data highly accessible:
Once companies are habitually asking data-oriented questions, Rouda says, it’s imperative to make the answers very easy to find.
Rouda previously led the marketing at a technology company, where he says a lot of potentially valuable data wasn’t always easy to access or analyze.
“Once we started getting better tools in place and creating things like dashboards” for sales leads, he says, “we started to get value out of the data.”
Some employees equipped with simple, yet dynamic, analytics tools will roll up their sleeves and start directly interacting with big-data sources, but “you can’t expect everyone to know how to get started,” Rouda says.
“Find as many ways as possible to make it easy for people to find and use data in their jobs.”
- Charter for disruption:
The first two steps create a foundation for change, but to get truly transformative work from your analytics program, you have to clearly ask for it.
“Ask employees to throw out how you’ve always done it, and think about, ‘What else can we do with our competencies and assets?’” Rouda advises. “Give them the freedom to throw the crazy ideas around. Maybe nine out of 10 ideas never get beyond the whiteboard, but the tenth is something totally different.”
Rouda calls this a charter for disruption. And it requires mixing analytics with creativity.
“This can be hard for companies because they feel they’re disrupting their own business,” he says. “But if there’s not a charter for that, you end up only refining your current processes.”
- Work across disciplines:
To effectively infuse analytics with creativity, Rouda advises incorporating the perspectives of people with different backgrounds, skills, and vantage points.
Creative, disruptive, and potentially transformative analytics “has to be interdisciplinary,” Rouda says.
“Your data scientists may not fully understand the business. They can do all the math but don’t necessarily know what the problem is that HR is trying to solve in employee retention, for example.”
So don’t let all the creative brainstorming remain in disconnected silos. Bring personnel in human resources, sales, and marketing together with IT and whoever has in-depth knowledge of the data.
- Incorporate machine learning:
The high volumes of data that organizations collect will make so-called machine learning not only viable but also a critical part of their analytics portfolio.
Machine learning is form of artificial intelligence that allows devices to learn without being programmed. Computers programs change as they become exposed to more and more data.
| Related: Four Steps to Experimenting with Innovation
Organizations that embrace machine learning will increasingly lean on these systems to find patterns and outliers in their data.
“Machine learning lets you process huge volumes and pull out new nuances and correlations,” Rouda says.
It helps you “look at 100 percent of the data to improve your algorithms.”
The math behind these capabilities isn’t new, he says. But with improved economics, speed, connectivity, and software associated with data storage and analysis, machine learning is becoming increasingly accessible for real-world applications.
Wherever a company is in its analytics work today, it’s important to move steadily along this maturity curve, Rouda says.
Indeed, he adds, “There’s an argument to be made that the companies that don’t put forth this effort are the ones at greatest risk of disruption.”