Sangeeta Krishnan is an engaging business intelligence and analytics leader who possesses a winning blend of subject-matter expertise and practical experience from a variety of industries. Most recently, she joined Bayer as North American Analytics Lead for mass sales. She has worked with Fortune 500 organizations, not-for-profits, and everything in between, helping various organizations build their operations and monetizing data products from the ground up. Krishnan is a public speaker, content creator having articles published in industry journals, and was recognized as a Finalist of the Women in IT Awards 2018 (USA) in the Data Leader of the Year category. She is the author of Thriving in a Data World (Business Expert Press, December 2022). Learn more at sangeetakrishnan.com.
Q: Can you tell us about your experience working with data analytics in various industries?
In recent years, the role of data analytics has expanded and grown in importance across various industries. This trend is a result of several contributing factors, like covid, digital transformation, etc.
Although use cases vary by industry, they all contain a few fundamental characteristics, including data-driven decision making, more personalized customer service, tapping into new opportunities or markets, gain competitive advantage and improving operational efficiencies.
Healthcare data is unique in that it is scattered, inconsistent, and complex. Also, extreme caution is required when dealing with PHI (protected health information). If your data is all about MRO (Machine Repair Overhaul), data deidentification may not be necessary. To be effective, one must be knowledgeable of regulatory and privacy rules particular to the industry you work in.
Q: How do you define data analytics and what are some common misconceptions about it?
Data Analytics is defined in a variety of ways. However, in a nutshell, it is the process of cleaning, analyzing, manipulating, and modeling data in order to discover usable information, draw conclusions, and support decision-making. It entails extracting insights and knowledge from structured and unstructured data using statistical, mathematical, and computational methods.
I'll highlight a few frequent misconceptions regarding data analytics:
- Data analytics is only for large companies: Companies of all sizes can get value out of data analytics
- Data analytics only requires technical skills: This is a widely held misconception. Communication skills and a data culture are required for success, and technical ability alone will not drive the adoption of technical solutions.
- Collecting data will lead to actionable insights: There is no automatic insight pipeline following data collection. It is critical to evaluate data analytics results with caution and to take into account elements such as data quality and completeness.
- Data analytics can be used to replace human judgment: It should be used to supplement, rather than replace, human decision-making. When evaluating results, it is vital to evaluate the limitations and biases of the data available.
- Can you discuss the different types of data analytics and their specific use cases?
Here are some instances of data analytics:
Descriptive analytics is concerned with summarizing and characterizing data in order to comprehend what has occurred in the past. Customer segmentation, sales analysis are common uses for descriptive analytics.
Diagnostic analytics is a step beyond descriptive analytics in that it aims to explain why certain occurrences occurred. Diagnostic analytics is frequently used for root cause analysis and determining the components that contribute to a specific outcome.
Predictive Analytics: This sort of analytics makes predictions about future events using historical data and statistical models. In financial services, healthcare, and marketing, predictive analytics is often used to estimate client behavior and influence strategic decision-making.
Prescriptive analytics goes beyond prediction and provides precise recommendations or actions to take based on data analysis. Prescriptive analytics is used to optimize operations and enhance outcomes in domains such as logistics, supply chain management, and healthcare.
Real-time Analytics: This sort of analytics is used to handle and evaluate data as it is generated in real time. Real-time analytics is widely utilized in industries such as transportation and e-commerce to monitor data in near real-time and respond quickly.
Cognitive Analytics: The use of artificial intelligence and machine learning algorithms to analyze data and reveal insights is referred to as cognitive analytics. This type of analytics is frequently used to automate mundane processes and to provide decision support. In banking, for example, cognitive analytics might be used to automate loan underwriting and detect potential fraud.
Q: How important is data storytelling in driving business decisions and why?
We humans respond more to stories than to data and numbers. Data storytelling provides a compelling and accessible means of communicating complicated data findings to a diverse set of stakeholders. By demonstrating the impact of data insights and highlighting opportunities for improvement, data storytelling can help inspire action. Data storytelling helps to make data insights more memorable by connecting them to a narrative that is easy to recall and understand.
Q: Can you talk about the process of building a data team and creating a data culture within an organization?
Find people with a strong combination of technical and soft talents to put together the finest data team possible. A varied team with diverse backgrounds and perspectives is also important since it can lead to more innovative and effective problem-solving. Furthermore, because data analytics is a constantly evolving, it is vital to foster a culture of continuous learning and development.
Culture cannot be created suddenly; it must be developed gradually. Recognize that it will take a lot of work to shift the mentality if using data has not been standard practice in your business.
Q: Can you discuss the potential dangers and ethical concerns that come with data analytics and how to mitigate them?
The increased use of data analytics comes a range of potential dangers and ethical concerns.
The following are some of the potential risks and ethical considerations related with data analytics:
Privacy concerns: Data analytics frequently involves the collecting and analysis of sensitive personal information, such as financial and health information, which can pose a significant privacy risk if the data is leaked.
Bias and discrimination: If data analytics algorithms are trained on biased data sets, they can perpetuate existing bias. This can also lead to lack of transparency and unintended consequences.
Organizations should create comprehensive data governance and privacy policies that comply to best practices and relevant regulations to mitigate these potential risks and ethical problems. Furthermore, firms should audit their data analytics methods and algorithms on a regular basis to verify that they are transparent, fair, and responsible. Organizations should also invest in data literacy training for their stakeholders and employees to ensure that they understand the limitations and potential risks connected with data analytics.
Q: How do you measure the return on investment for data analytics initiatives?
I will provide a few ways of measuring ROI for data analytics initiatives (which I touch on in my book Thriving in a Data World). An organization may have many fresh ideas, but it cannot implement them all. There must be a strategy in place to focus data team capacity on what is most important in terms of ROI. Based on the required KPIs, there are tangible and non-tangible ROI measurements. Another approach is to weigh two factors: the ROI of doing it versus the risk of not doing it.