(US & Canada) | Data Tools Need Continual Iteration, Improvements, and Maintenance — Expedia Group, Director of B2B Marketing

Helena Cho, Director of B2B Marketing at Expedia Group, speaks with Diena Lee Mann, Founder and CEO of Spectio, in a video interview about her professional trajectory, her role in building prescriptive analytics toolkits, business challenges to make data-driven decisions, and delivering a successful data product.

Expedia Group, Inc. is an American travel technology company that owns and operates travel fare aggregators and travel metasearch engines.

Cho’s journey in the travel industry has been about constant learning and collaboration with the best minds. She began her career with Vrbo, a vacation rental brand, where she started the brand’s marketing analytics team.

The role gave her a comprehensive understanding of consumer behaviors and B2B interactions, says Cho. After joining Expedia Group, she shifted towards the B2B business side where she analyzed and provided recommendations across various supply partners. As a part of her analytics career, she and her team have developed toolkits and systems that move from descriptive to advanced prescriptive analytics.

Prescriptive analytics has constant data loops and insights for business leaders, partners, and travelers. Sharing some examples, Cho mentions building a self-serve data product and crafting a marketing technology infrastructure.

Further, she recalls engaging in various customer analytics, such as core analysis and lifetime value projections, and working on the capital allocation of different paid media channels and tactics. Adding on, Cho maintains that it required close collaboration with the machine learning team.

When asked why businesses face hurdles to making confident data-driven decisions, she states that the challenge is a result of a few things. First, Cho alludes to the expectation of speed versus the reality of analytic processes.

While decision-makers are in frequent need of quick insights, she notes that it takes significant time for the data analytics team to address concerns. The delay stems from the need to prioritize multiple ongoing requests on top of existing projects.

Cho further states that this misalignment between the speed of business needs and the pace at which data can be processed leads to frustration for both sides. She opines that the lack of a robust, user-friendly, self-service data platform is another challenge.

However, as organizations have started to take the route of developing the tools, many times these data products do not offer the flexibility to pull the necessary metrics specific to business questions.

Adding on, Cho addresses the common misconception that once a data product is built, everything can be made data-driven. Whereas, in reality, these data tools need continual iteration, improvements, and maintenance to address bugs and adapt to new requirements to support data-driven decision-making.

Commenting on developing the people and processes to deliver data products, Cho states that it is critical to understand the crucial elements of a good data product. Apart from that, it is necessary to share the roadmap and get buy-in from end users and involved teams.

Moving forward, Cho states that, while building the data loop, any part that must be implemented for the user or product experience without human decision should be automated.

Furthermore, she cautions against waiting for insights and then prioritizing with the product and tech teams, as it would slow down the process and frustrate the business and its users.

Therefore, it’s essential to automate parts of the decision-making process to ensure continuous improvement without human delay, even if it is not fully implemented from day one.

Secondly, Cho highlights the ability to manipulate data to test the end-user hypothesis. Often, data products come in fixed formats that have limitations and require constant updates.

In contrast, data products that allow end users to build their own dashboards or manipulate data tend to be more successful and continuously used.

Next, Cho emphasizes the importance of having a robust alert system as part of a data product. The alert system should notify both business end-users and the analytics and engineering teams when anything abnormal is observed in the data.

Sharing some examples, Cho mentions bot attacks that may not be noticed until someone reviews traffic data, among other such situations. An effective alert system can help catch the issues before they escalate.

Concluding, Cho states that adhering to these crucial data elements is essential to delivering a successful data product.

CDO Magazine appreciates Helena Cho for sharing her insights with our global community.