2 weeks Ago
Josh Dunham is the cofounder and CEO of shipping data and analytics company Reveel . getty Within the tech community, AI was the topic in 2024. It was also a year in which people and organizations tried to make sense of the fast-evolving AI landscape, understand its utility and identify opportunities to apply it in business.
It can also be argued that 2024 was the year AI truly went mainstream. Capabilities unimaginable just a few years ago—writing novels, driving autonomous vehicles with precision, upending photography and controlling increasingly human-like robots—teased all and frightened many with AI’s potential. Such futuristic applications also dominated the headlines, with the ensuing hype setting the tenor of society’s ongoing reckoning with AI’s impact.
My company uses AI in our Shipping Intelligence Platform where it’s appropriate, but in the face of such attention-grabbing applications, it can be easy to lose sight of the fact that other fundamental developments in how enterprises use data also occurred last year. In particular, innovations in data science and business intelligence (BI) continue to transform industries and are directly pertinent to the efforts of enterprise IT leaders to ensure that their organizations can effectively manage, parse and use their data assets. Just as importantly, data science and business intelligence capabilities are imperative, as the failure to glean actionable insights from data often has significant business ramifications.
This was particularly evident last fall in the parcel shipping sector, where my company operates, when Macy’s made the headlines for all of the wrong reasons. Specifically, company leaders found that an employee intentionally hid shipping outlays over several years. The scheme amounted to flawed accounting for some $150 million, a sum that forced the company to push back its quarterly earnings announcement and retroactively adjust financial statements going back nearly three years.
This experience serves as a cautionary tale for any enterprise that lacks the data science and BI-enabled capabilities to make sense of its own data or to flag when data outputs don’t make sense. It also serves as a powerful reason for IT leaders to stay informed of relevant advancements in data science that all too often fail to garner attention because of the AI hype despite their direct impact on business continuity and operations. In a time of AI hype, innovations in data science and BI continue.
The data science sector as a whole is dynamic, with new developments occurring at a rapid pace. Notably, it is important to remain aware of how such innovations enable organizations to be more successful. Consider a few of the representative examples shaping data science and BI today: While advancements in data science initially focused on using data to react more effectively to business developments, today’s powerful modeling technologies and BI offerings empower users to pose powerful “what if” scenarios to see how various actions will impact their operations.
While this transition to predictive analytics is not new, advancements in data normalization have extended it into new industries ( retail , healthcare and parcel shipping data are just a few examples). Importantly though, innovation is not stopping there. Today’s tools also enable prescriptive analytics that recommend what actions to take when the very events that are predicted to occur are detected.
Data, of course, must be stored if it is to be analyzed and used to make data-driven decision-making a reality. Most enterprises today place their data warehouses in the cloud, but there are challenges involved and issues on the horizon, particularly as AI workloads push many data centers to the limit both for capacity and processing power. Innovative approaches to data offer a solution, with Atombeam—a new startup that has upended how computers communicate—being one that has promised to decrease the size of data by 75% while keeping it intact and searchable, which is something not possible with compression.
The old adage that you can't manage what you can't see remains just as viable today, but advancements are augmenting and changing the visual aspect of decision-making. Data governance and management are, of course, more difficult as organizations grow more distributed and complex, a reality that Tableau worked to address with its Tableau Cloud Manager. Provisioning data, while addressing sovereignty issues, is but one advantage.
Synthetic data is not without challenges, but the ability to create data that mimics real data is exceptionally useful, particularly in fields where the data being parsed by analytics and BI tools includes personal information and is governed by rightfully strict privacy regulations. In the coming years, data created by algorithms for the express purpose of running analytics that enable data-driven decision-making will make several new use cases possible. The flashy use cases associated with AI will likely continue to dominate headlines and many of the discussions that arise around data science in the coming months.
Even so, the continued emergence of innovations that enable organizations to get more out of the data they have—and, in the process, identify where problems or challenges lie—will continue to benefit enterprises. Such will be the case in 2025, when more companies will continue to benefit from innovations in data science and business intelligence even as the hype around AI grows. Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives.
Do I qualify?.
