It goes without saying that for the past few years, Artificial Intelligence (AI) has been the hot topic in the world of technology, and it’s hard to spend even a few minutes on LinkedIn or at a conference without encountering conversations on how it is going to change the world. However, for all of the times I’ve encountered the term “AI” in passing, mentions of “Citizen Data Science” pale in comparison. I believe that understanding and embracing Citizen Data Science will be essential for industries like hospitality as they work to successfully integrate AI and leverage it for effective business decision-making.
We at Revenue Generation have spent the past year engaged with a portfolio of five independent hotels, assisting them with their Citizen Data Science journey. We’ve developed fully automated tools to organize and synthesize their data, recreate manual reports, and develop analysis that quickly delivers insights in a repeatable fashion. We’ve empowered team members within their commercial organization to be data “citizens,” pairing their business expertise with these analytical tools to address their most important problems. Crucially, they’ve been able to leverage their current technological and statistical skillsets to accomplish this, without the need for more training. While its journey is still ongoing, this organization has focused on incorporating data-driven decision making into its culture, and it shows in the results.
Citizen Data Science Defined
So what is Citizen Data Science, or similarly, a Citizen Data Scientist? There are various definitions floating out there, but the one I like the best comes from Gartner, the global research and advisory firm: “a person who creates or generates models that leverage predictive or prescriptive analytics, but whose primary job function is outside the field of statistics and analytics.” This person likely still needs to possess a high comfort level with technology, a decent understanding of mathematical and statistical concepts, and certainly (and arguably, most importantly) the appropriate subject-matter expertise. An advanced degree in a quantitative field is not required, and this distinction may be what sets the Citizen Data Scientist apart from their “professional” counterparts.
The Citizen Data Scientist sits at the important intersection between information technology and operations, which has historically been occupied by business managers, analysts, engineers, product designers, “professional” data scientists, or some combination of the above. Many industries face the ever-growing challenge of keeping pace with rapidly expanding data, while making business decisions at an exponentially faster rate and greater scale; remember the “three V’s” of Big Data from that conference you attended ten years ago – volume, velocity, and variety.
The human capital to support this massive effort inevitably falls short given the difficulty and expense associated with finding and/or acquiring employees with the correct skills and education. Fortunately, many advancements have been made in AI-based technological tools that bridge this knowledge and experience gap, in the form of large language models (LLMs, like ChatGPT) and automated machine learning (AutoML). LLMs, in this context, enable business users to build and support tools and applications, and AutoML helps to synthesize datasets into predictive and prescriptive insights that drive decision-making, neither requiring coding skills or knowledge. These tools have been made increasingly easier to use by less technically savvy users. Coupled with finding the right Citizen Data Scientists to deploy them toward addressing the right business problems, they help to overcome these challenges.
Citizen Data Science in the Hospitality Industry
As innovative yet practical as it may sound, Citizen Data Science is not the proper fit for every vertical. Removing or reducing the reliance on “professional” data science expertise can result in lack of proper oversight, data quality issues, and ultimately incorrect or poor decision making that can have very adverse consequences. So why might it be just the right fit for the hospitality industry, which is not typically known for advanced, data-driven decision making? I believe it boils down to three things:
The opportunity that more basic data analytics and automation provide
The current and future career trajectory of team members tasked with supporting data-driven decision making
The “lower risk” nature of the decisions themselves
First, there are two potential precursors to true Citizen Data Science that are immensely valuable by themselves and can form a basis for gradual evolution into Citizen Data Science. In his book All Hands on Tech, Tom Davenport describes “Citizen Data Analysis” and “Citizen Automation,” concepts that are within the realm of Citizen Data Science but are a step less sophisticated and don’t come with all of the same caveats. Citizen Data Analysis, often known historically as Business Intelligence, equips and empowers business users with the ability to synthesize available data into KPIs, dashboards, reports, and simple predictive models, often within a “self-service” paradigm. Business users can quickly derive actionable insights from the data, allowing data analysis professionals to focus their time on other activities aside from development, including scaling existing efforts. Similarly, Citizen Automation tackles routine, time-consuming, and hard-to-scale business processes by using accessible, user-friendly technology.
I work closely with Commercial Strategy teams within hospitality, including Sales, Marketing, and Revenue Management, and see firsthand how much time is spent just gathering and compiling data into reports, then distributing them to stakeholders on a regular basis. Likewise, strategy meetings require hours to prepare for, and this effort increases linearly with the number of properties that need to be covered. Recent industry surveys have revealed that these teams tend to spend over half of their time on such manual tasks, and all of this occurs before even a single insight is discovered, a decision is made, or a revenue opportunity identified. Through the combination of automation and Citizen Data Analytics, commercial teams in hospitality can reduce manual data preparation efforts, empower stakeholders to quickly find and act on insights, and more easily scale their efforts as their organizations grow.
Citizen Data Science also meets the needs of the hospitality industry given the personas of the individuals that would engage in these initiatives. Earlier I mentioned that an aptitude with technology and a familiarity with math and statistics are prerequisites, but not formal training in data science, programming, or computer science. A Citizen Data Scientist should also possess an intimate knowledge of the business problems that need to be addressed in the organization, and should have the motivation to solve these problems. I think again about Commercial teams and the types of problems that need solving – segmentation, targeted marketing, offers, and promotions, forecasting, pricing, CRM, personalization, and so forth. The nature of these problems and the ability to address them through quantitative methods forms the right foundation upon which teams can build their technological and quantitative skill sets. So, even if team members don’t currently check all of the required boxes, there is a natural career progression that begins with business expertise and, with the help of the right Citizen-facing tools, can guarantee professional growth and success for both the individual and the organization.
Finally, Citizen Data Science is most appropriate for industries where the outcomes of decisions are not “life-or-death.” In All Hands-on Tech, Davenport cites failures at a nuclear energy plant, potentially adverse medical implications, cybersecurity-related matters, and outcomes with substantial financial impact as those that would require the more scrupulous attention of trained data scientists. Hospitality does not involve such outcomes, and short of a calamitous and sudden emptying of occupied hotel rooms and/or a massive shift in strategy, the financial risk is relatively small, and usually outweighed by the upside potential. Of course, in all applications of Citizen Data Science, when indeed there are more significant consequences on the line, consultation with a trained data scientist is essential.
Steps Toward Adoption
Like all technological growth, adoption of Citizen Data Science and the corresponding organizational change cannot and should not happen overnight. An organization should implement basic tools and processes that drive immediate value, and then iterate, refine, and evolve. Hospitality organizations can take a few practical steps in parallel as they embark on their journey.
The first step is educating team members. This might include upskilling on tools, understanding available data, improving the ability to frame questions, and for those with less quantitative backgrounds, incorporating foundational statistics. Given different learning styles, this can take a variety of forms from online training classes to hands-on “learning by doing.” I have found that anyone that is well-versed with Microsoft Excel is likely ready to expand their knowledge to some of the common Citizen Data Analysis and Data Science tools. There are many excellent sources for learning statistics that are tailored more closely for a business audience.
Second, organizations need to evaluate their available data, ensuring that the right citizen analysts have access to the right data elements at the right times. This includes careful evaluation of the business processes that typically produce data internally for an organization, as poor processes often result in poor data. Metrics need to be well-defined and consistent, to strive for a “single source of truth.” For example, if Operations calculates occupancy based on physical rooms, while Revenue Management excludes rooms out-of-order from the denominator, two conflicting figures will emerge, both claiming to represent the same metric. This disparity can lead to confusion and skepticism from stakeholders. As commercial organizations in hospitality continue to centralize, the risks associated with siloed and inconsistent data are mitigated. I would personally advocate for a focused, centralized data and analytics effort to accompany these coordinated Commercial Strategy initiatives, including a single, central data model accessible by all Citizen Data stakeholders.
Finally, hospitality organizations adopting the Citizen Data Science paradigm need to establish and foster a culture of data-driven decision making. As is the case with most corporate culture, this should be championed by senior management and should focus on consistent use of data towards decision making at all levels of the organization. Much more could be written on the topic of building a data-driven culture, but technology has set the stage for exciting opportunities for career advancement, greater efficiency, greater job satisfaction and a slew of other intrinsic and extrinsic benefits. Energizing teams around the concepts of Citizen Data Science should prove easier than ever.
In summary, the implementation of Citizen Data Science in the hospitality industry offers remarkable opportunities for empowering business users, enhancing efficiency, and fostering a data-driven culture. By equipping team members with the necessary tools and knowledge, organizations can streamline manual data preparation and enable stakeholders to derive rapid insights, ultimately driving better decision-making and growth. The focus on Citizen Data Science is particularly suited for hospitality, where the impact of decisions is significant, yet manageable and relatively low-risk. With gradual adoption and strong support from senior management, hospitality organizations can seamlessly integrate Citizen Data Science into their operations, ensuring a future marked by professional development, improved performance, and innovative problem-solving.
