The book I will be reviewing today is called Analytics at Work: Smarter Decisions Better Results by Thomas H. Davenport, Jeanne G. Harris, and Robert Morison. The book was an interesting read although it felt like it was missing deepness and actual concrete examples. Don’t get me wrong! The authors are obviously professionals in their fields of analytics, but except giving advices left and right, it didn’t leave me an ever lasting impression. Hence, instead of judging on the quality of the book, I will focus on the main points it was trying to deliver to the reader.
I am saying this post is a book recap because I will go over the main points the book highlighted. I will neither compliment nor bash the book because I thought it was a great read, but that had not many dimensions for an interesting analysis.
Why use analytics:
Everyone knows that analytics is a science of the future. For companies, most debates can end with concrete numbers. The books highlight the importance of properly managing and understanding the changes in your business, know what is really working. Analytics is hugely dependant on technology so know and leverage the previous investments in IT the company is making. The goal is to cut costs and improve efficiency using data. Hence, analytics help to manage risk, anticipate changes in market conditions and have a basis for improvement decisions over time.
Some questions to ask ourselves:
- Time frame: Are we working with past, present or future?
- Innovation: Are we working with known information or gaining new insights?
- Insight: How, Why did it happen? What would be the best next action? What could happen next?
Typical Decision-Making Errors:
Now, data analytics is not a perfect since and so are the analysts/CFO/CTO/scientists working on them. Data is so large as a field and you can find interesting correlations between any two variables. Hence, be ready for logic and process errors. Logic errors include not using the right questions, making incorrect assumptions and failing to test them, using analytics to justify what you want to do and failing to take the time to understand all the alternatives or interpret the data correctly. Process errors include making careless mistakes (transposed numbers in a wrong spreadsheet or a mistake in a model), failing to consider analysis and insights in decisions, failing to consider alternatives seriously, using incorrect/insufficient decision-making criteria, gathering data or completing analysis too late to be of any use, and postponing decisions because you’re always dissatisfied with the data and analysis you already have.
Hence, it is always good to start off by defining properling the problem at hand before jumping to conclusions. Then, acquire and extract the data that is needed and define the proper model to approach the say problem. Implement the analysis and come up with a conclusion. Manage the ressources accordingly (for example. prepare budgets) and present the results to the management team.