Algunas cosas no cambian, incluso durante una pandemia. Como en años anteriores, en la Encuesta sobre el estado del CIO de 2021 del CIO, una pluralidad de 1.062 gerentes de TI encuestados eligieron “análisis de datos / negocios” como la iniciativa tecnológica número uno destinada a impulsar la inversión en TI.
Unfortunately, analytics initiatives rarely perform that well in terms of stakeholder satisfaction.
Last year, Mary K. IOC Associate Pratt provided an excellent analysis of why data analytics initiatives always fail, including poor-quality or isolated data, vague rather than focused business goals, and universal and rare feature sets. But a host of new approaches and technologies make these crashes less likely.
In this bundle of articles from CIOs, Computerworld, CSOs, InfoWorld, and Network World, you'll find tips and examples that can help you ensure your own analytics efforts are paying off. These initiatives tend to resemble development projects, even when commercial products are involved, and have the same well-defined goals and iterative cycles that distinguish successful software development outcomes.
To get a general idea, start with InfoWorld's guide "How to Excel at Data Analysis" by contributor Bob Violino. In this perfectly written article, Violino covers all the basics: establishing centers of analytical excellence; the benefits of self-service solutions (such as Tableau or Power BI); the exciting possibilities of machine learning; and the transition to cloud analytics solutions. Violino develops this last point in a second article, this one for the CIO: "Cloud analytics: the main challenges and how to overcome them." As you can see, the scalability of the cloud and the myriad of analytics tools can be overwhelming, but migrating large amounts of corporate data to the cloud and securing it can be an exciting adventure.
New technologies invariably bring new risks. No advance has had a greater impact on analytics than machine learning, from automating data preparation to detecting meaningful patterns in data, but it also adds unforeseen danger. As CSO editor Lucian Constantin explains in “How Data Poisoning Attacks Machine Learning Models,” deliberately biased data injected by malicious hackers can steer models toward a nefarious goal. The result could be, for example, manipulated product recommendations or even the ability for hackers to infer sensitive underlying data.
There is no doubt that analytics has a dark side, as Matthew Finnegan corroborates in the Computerworld article “Collaborative Analytics: Yes, You Can Track Employees. Ought? ”Collecting and analyzing metadata about user interactions on collaborative platforms has its legitimate advantages, such as the ability to identify communication bottlenecks or optimize the employee experience. But the same platforms can be used as employee monitoring systems that invade privacy and erode trust between management and everyone else.
On a lighter note, consider this great case study on analytics that increase user satisfaction: "Major League Baseball races towards network visibility." Writing for Network World, Editor-in-Chief Ann Bednarz examines how MLB uses network flow analysis software in its infrastructure to ensure players and fans alike have consistent network performance: end-to-end, Wi-Fi at headquarters and cloud services.
This effort to implement unified network analytics to optimize the user experience began just two years ago, primarily because MLB's new senior network automation software engineer saw the need to do so. Its realization broke through perhaps the most important obstacle to the success of analytics initiatives: cultural inertia.
Ultimately, the secret to successful analytics is not choosing and implementing the perfect technology, but cultivating a broad understanding that ubiquitous analytics leads to better decisions and superior results. You can usually fix technology issues or misunderstandings about the requirements.
<p>Copyright © 2021 IDG Communications, Inc.</p>