1. Building a data-driven culture
Welcome back! Let's continue on our journey of Demystifying Decision Science
2. Unmet potential
Decision science has immense potential, but unlocking it requires more than just data - it requires the right culture.
A significant number of decision science projects fail to deliver on their potential. They might get stuck in development purgatory, produce irrelevant results, have insurmountable implementation issues, or lose the confidence of their customers.
Understanding these common pitfalls is the first step in avoiding them.
It is useful to break these common pitfalls into two categories: Data reasons and Business-related reasons.
3. Pitfalls
The data reasons include a lack of data, insufficient infrastructure, lack of skilled personnel, and poor quality control. These are the technical failures that data scientists constantly think about.
The business-related reasons are often broader. They can include failing to gain executive buy-in, poor project management, insufficient communication, inadequate investments, poor collaboration between the business and technical teams, and one of the most common issues, a lack of a decision science culture.
4. Five key steps
Investing time to ensure your company has a decision science culture has very strong returns. Five key steps to successfully developing a decision science culture are learning the basic vocabulary, identifying problems, asking good questions, challenging assumptions and embracing a test-and-learn culture.
5. One, two, three
Step one is mastering the basic language of decision science, which includes risk, bias, and experimental design. These aren't simply words to memorize for an exam but concepts that drive much of decision science. Business customers who understand the basic vocabulary of decision science can be partners in success.
Step two involves prioritizing problems. Decision science is a set of tools for solving problems, but you need to start with the business problem, not the tool. Prioritizing the problems means that those that are most critical to the business's success and most amenable to solutions are near the top of the list.
With the language and problem list in place, these business partners are ready to move on to Step Three, asking powerful questions. They can inquire about the tools, frameworks, data sources, and models. They can ask deeper questions about what the company is trying to achieve if they are measuring the right things, and if they have considered other options.
6. Four and five
In asking these powerful questions, the decision science customers can also embrace Step 4, challenging assumptions. The status quo isn't always optimal. Use your analytical skills to scrutinize assumptions and challenge conventional wisdom. Data can be a powerful ally in revealing hidden opportunities and risks but it can be imbued with biases. Challenging the assumptions made by the data scientists can drive innovation and create a culture of continuous improvement within your organization.
Lastly, a data-driven culture embraces a "test and learn" mindset. In the real world, not every experiment is a home run. A culture of continuous learning is what separates innovative organizations from stagnant ones. Embrace the process of testing, iterating, and improving.
Decision science isn't just about having the right answer; it's about empowering everyone to make better choices.
Practice makes perfect, so the more you can improve your language of experimentation, prioritize problems, ask powerful questions, challenge assumptions, and embrace a data-driven culture, the more your organization will be positioned to tackle the business-related reasons for success.
7. Let's practice!
Let's do some exercises now to get that practice!