Natalie Osborn

Are you ready for a data scientist, or are you building a culture of analytics?

You might be forgiven for thinking a data scientist is a mythical creature that is rarely found in the hospitality world. Part mathematician, part computer scientist and part trend-spotter, data scientists are analytical data experts who have the technical skills to solve complex problems – and the business acumen to explore what problems need to be solved. Intrigued? In this article we will explore what data scientists do, how they do it and whether your organization is ready for a data scientist skillset.

Data scientists possess a unique blend of skills. Not only do they have to solve problems by squeezing information out of data, they also communicate results and persuade others to apply that information to their decisions. To achieve this, statistics and machine learning are important technical skills for data scientists, but they are not the only skills required. Data scientists must be able to apply their technical skills to solve business problems.

And finally, even the smartest statistician in the room will fail if she cannot explain the relevance of her results. Data scientists need to fully understand their data and be proficient in explaining its significance to the problem at hand. So, once a data scientist has these skills, what do they actually do on a daily basis?

What Do Data Scientists Do?

  • Collect data and transform it into a format that is usable for analysis.
  • Use data-driven problem solving to address business-related problems.
  • Perform statistical tests and distributions.
  • Use machine learning, deep learning and text analytics techniques to find answers or evenopportunities in data.
  • Seek out order and patterns in data, as well as spotting trends that can help a business’s bottom line.
  • Communicate and persuade IT and business leaders of opportunities and the need for a solution.
  • Collaborate with both IT and line-of-business departments to implement successful solutions.

How Do They Do It?

Many data scientists began their careers as statisticians or data analysts. But as data began to grow and evolve, these roles evolved as well. Data is now key information that requires analysis, creative curiosity and an affinity for finding opportunity and defining solutions. The following techniques are among some of those most commonly used by data scientists.

Data Preparation: Data scientists often know what data they want to profile or visualize ahead of time. The data preparation process involves determining what data can best predict an out come. And because more data generally means better predictors, bigger really is better in this case. But accessing all that data is challenging. One reason is that different data sources, formats and structures make it hard to bring the data together.

Data Visualization: A picture is worth a thousand words – especially when you are trying to understand and discover insights from data. Visuals are especially helpful when you’re trying to find relationships among hundreds or thousands of variables to determine their relative importance – or if they are important at all.

Machine Learning: This is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. Machine learning is a method of data analysis that automates analytical model building.

Deep Learning: This area of machine learning research uses data to model complex abstractions. Deep learning is a type of machine learning that enables a computer to perform human-like tasks, such as recognizing speech, identifying images, detecting objects or even making predictions. Traditional computing involves running organized data through predefined equations. In contrast, deep learning sets up basic parameters about the data and trains the machine to learn using layers of processing to recognize patterns.

Text Analytics: This is an analytics technique that allows a data scientist to identify patterns and clues in text data to help detect problems. This technique has been used in multiple industries to detect and solve crimes such as fraud. This can also be used to understand systemic problems in hospitality operations that may manifest themselves in customer complaints, survey data, ratings and reviews or even operations logs.

Is Your Organization Ready For a Data Scientist?

Let’s go ahead and admit it: gaining a data-wrangling, artificial-intelligence wielding, problem-solving magician on your team sounds fantastic, doesn’t it? But before you go ahead and write that job description, there are several key characteristics that your organization needs before employing a data scientist. Here are some things to consider:

How much data do you really have?
Does your organization deal with large amounts of data and have complex issues that need to be solved? Organizations that can make the best use of data scientists typically manage massive amounts of data on a day-to-day basis. As I mentioned earlier, data scientists need a lot of data. If you are a single hotel or a small chain, hiring a data scientist might not be the best use of your resources. Consider investing in better tools for your business analysts, such as self-service data management tools or data visualization, and you will get a bigger bang for your buck.

Is data important to you?
Do you value data? Do you use common data definitions across your organization? Do you have well-defined and measured key performance indicators? Your company’s culture has a massive impact on whether you should invest in data science. Do you foster an environment that supports analytics? Are your executives believers? If not, investing in a data scientist would be money down the drain. That said, developing common data definitions and key performance indicators are not required only for data science. These approaches will improve the efficiency of your business – even without a data scientist on staff.

Are you willing and able to embrace change?
Data scientists doggedly devote their time to finding ways your organization can better function. In response, your organization needs to be ready – and willing – to adopt and implement the results of their findings. If you are not open to re-evaluating how you do business, you will soon frustrate your data scientist.

Do you have the support mechanisms needed for a data scientist?
You read in the, “What do data scientists do?” section of this article that collaboration and communication with both IT and business departments of your organization is an important element of what to expect from a data scientist. These two departments are also critical to supporting the success of the data scientist. From IT, a data scientist needs easy access to data and efficient implementation support once a solution has been determined. From the business side, whether it is marketing, operations or revenue management, the data scientist will need assistance with understanding the business problem and identifying the change management needed for a successful solution. If you don’t have the full support in either of these areas, your data scientist will likely fail.

In addition to critical departments such as IT and business, you also need leadership support for your data scientist. There may be times when analytic results contradict current wisdom or the way things have always been done. Your organization will need strong leadership, and a leadership that understands the value of analytics, to help drive a counter-intuitive change in your organization. However, it is these changes that give your organization the most competitive advantage.

How Can I Start to Build a Culture of Data and Analytics?

Hiring a data scientist to guide business decisions based on data can be a leap of faith for your organization. Make sure that your organization has the right mindset and is ready to make some changes. Then, simply point data scientists at your biggest problems and get out of their way.

The only catch? Data scientists are in short supply. In May 2018, the McKinsey Global Institute Workforce Skills Executive Survey identified data analytics as the highest skills mismatch. And it’s not just folks with deep analytics skills that are lacking. An earlier McKinsey study identified a deficiency of 1.5 million managers and analysts with the know-how to use big data and analytics to make effective decisions. If your organization is not ready to hire a data scientist, or team of data scientists, you can gain significant benefits by building a culture of data analytics.

To get started, identify a common business language across your organization so that data definitions and key metrics match up with each other. For example:

  • Do you have different ways of thinking about market segments?
  • How do you define occupancy?
  • Do you think about the geographical break down of your properties or your customers the same way?

You may be surprised at how often different departments disagree on these common metrics. Once you have established common data definitions, look at how different departments can use each other’s data or collaborate on special projects.

Incorporating data into everyday decisions is a key step on the path to a data-science-ready organization. Whether you are on the brink of hiring a data scientist or starting to build a data analytics culture at your company, how you use data and analytics to drive business improvements will be a key indicator of your success.

Natalie Osborn is senior industry consultant for SAS Institute’s Hospitality and Travel practice, and an 18+ year veteran of hospitality and hospitality technology solutions development, specializing in analytics and revenue management. Prior to joining SAS, Natalie was the director, product marketing for Minneapolis-based IDeaS Revenue Solutions, where she worked from 2000 to 2011. She is a frequent contributor to industry publications, speaker at industry conferences and is co-author of the SAS and Cornell Center for Hospitality Research blog, “The Analytic Hospitality Executive.” Contact Natalie.