We use data
to tell new stories.
How do we gain new insights from our partners’ data? Good data science involves several steps:
- Condition the data. We clean and organize the data, and decide what to do about incomplete data.
- Understand the data. We learn what each data element means in the context of the business, and how data elements relate to each other. We also identify and work to understand outliers in the data.
- Combine datasets. Most companies maintain multiple, disconnected databases designed to support separate processes or business lines. We combine datasets to reveal new insights about customers, businesses, and industries.
- Gather new data. We might collect external data and match it with corporate data. Or, we could gather new data with sensors, Internet tools, and scientific experimentation.
- Ask the right questions. Good data scientists don’t pan for gold; their work is targeted. We identify questions that are important to our partners, and we analyze the data to answer those questions.
- Distinguish statistical trends from random variation. To quote Mike Loukides, an expert on data science, “Statistics is the grammar of data science. It is crucial to making the data speak coherently.” We use proven statistical techniques to identify meaningful data trends.
This process gives us a thorough understanding of our partners’ businesses and customers. It helps us see connections that our partners haven’t identified. In this way, data science helps us tell new stories about our partners and their industries.