Find Your Degree






By Jean Thilmany

data science customer experience

You already know big-data analysts are indispensible to their companies. So much so, that there are currently (many) more jobs for them then there are data analysts to fill them. When you see the savings analysts can trim from company budgets and the other ways they can increase profits for their companies, the job divide makes sense.

“When we start to learn completely new things about our world by harnessing data, we realize we’ve been completely blind up until now as entrepreneurs and users,” says Mathias Lundø Nielsen. He founded Nustay, a hotel booking platform.

“But 90 percent of data we create is never sorted,” Nielsen says.

Data analysts help companies of all sizes increase their profits by shifting through that data in a measured way. They find ways that companies could trim costs, for example, or to increase the customer base, as well as myriad other profit-growing methods.

So let’s take a look at some of the actual ways data and analytics managers help their companies cut costs and increase value.

Bottom-line Savings

First up is Gestamp, which makes steel automotive components, such as the physical brake you press to stop the car. The company recently implemented a data analysis tool to help the automotive-component maker cut energy use. And boy did it succeed. Within one year, Gestamp cut its energy consumption by 15 percent across 14 European plants. That translates to $8 million and 45 gigawatt hours of energy saved and it removed 14,000 tons of carbon dioxide from the atmosphere, says Pablo de la Puente, Gestamp’s corporate information system director.

An energy monitoring system from Siemens now tracks the energy performance of equipment across those plants with analysts deciphering results. The tool also identifies and warns about energy malfunctions so they can be immediately addressed.

Before the tool, the company had no global approach to energy efficient measures. But Gestamp executives knew big data was important. In a conventional, low-resolution system, the company charted the energy consumption of one machine every 15 minutes–or even every hour–to measure energy usage, de la Puente says.
But with big data gathering and a resolution at every second in some cases, the new system provides a much clearer view of the equipment behavior.

To begin this study, analysts at Gestamp, worked with Siemens to identify the energy measurements and other information it needed to poll and pull together.

As the new Siemens system shows, data analysts must be ready to work with new data tools to get the most insight from the tool and (in this case) the equipment it monitors. They need to know how to turn that information into bottom-line savings via any number of company changes, updates, and other actions, de la Puente says.

In the case of Gestamp, the data showed that energy use spiked at some European factories at the beginning of the day. Data analysts were also able to determine the spikes happened when the plants were idle. What was going on? The analysts, working with maintenance personnel, discovered that furnaces are fired up at the beginning of the day.

“We had a big energy peak but no production. We were consuming energy without producing any part.” de a Puente adds.

“We talked with our maintenance teams to change the process, so furnaces are turned on sequentially before the beginning of a shift to try to avoid peaks,” he says.

Companies can pay more for their peak energy use. Limited production can also take place while the other furnaces fire up.

Next up, data analysts are looking to decrease Gestamp’s European energy consumption by an additional 5 percent, or 50 gigawatt hours, and reduce consumption by 10 percent at the company’s plants in the United States, Mexico, and India.

Booking the Best Hotel

Not every big-data analyst works with numbers returned at large, industrial companies. And not all works with numbers returned from equipment and systems already in operation. The service sector has plenty of opportunity for analysts whose bosses seek other ways to monetize big data.

Take Nustay, the hotel-booking platform founded in 2014 as a way to “personalize” customer experiences, in Nielsen’s words.

Big data is all about adding value to information. And that’s the job of the data analyst, he adds.

“So for oil, value would be in distribution; getting the oil efficiently to the people who need it,” Nielsen said. “But for us it’s personalizing; offering unique and tailored experiences to our users based on what they seek.”

Take the example, he says, of a man who wants to take his spouse to a great resort for an anniversary weekend. He doesn’t want a place with tons of kids swimming in the pool, and he wants the pair to be able to drive to nearby beaches.

Normally, the guy would spend half the morning looking up and comparing the amenities offered at different properties before booking a room, Nielsen says.

Nustay customers can search for exactly the type of hotel experience they want—say a golf resort with a white-sand beach nearby or a kid-friendly, small hotel with a restaurant that offers plenty of vegetarian dishes—and see potential properties ranked in order of preference. Big data operates behind the scenes, using metrics and numbers culled from social media and other sources, Nielsen adds.

Here’s where the analysts come in.

“We use a guest’s previous booking history to calculate how much money that person is likely to spend on a type of hotel and location, then we incentivize hotels to discount rooms further so their total revenue per stay is higher,” Nielsen says.

Because hotels themselves also benefit from use of the site, big data is, in this case, customer and supplier centric, he adds.

“We help out hotels solve more complex questions like can they optimize capacity all year round? How specialized can it become to attract more of a specific type of guest?” Nielsen says.

His final words on the subject underscore just how important big-data scientists are to all types of organizations, big and small.

“To be successful, you need to hire best data scientists you can get your hand on,” he says. “Find someone who understands architecture and is into the complexities you need. Then find someone who is visual and can turn it into something that can be understood by everyone else.”