When Steph Locke first ran her Data Science Fundamentals with R class, I was amazed at the work she put into the hands-on labs for the students. They had real challenges to help them learn how to model different kinds of data and learn answers from it.
I immediately said to her, “Hey, I wanna give you money. You clearly know what you’re doing. I want to learn more about my own customers. I’ve got a few different piles of data, and I want to figure out who buys training from me. If I can target the right customers, then I can send less emails to people who WON’T buy training, and keep their inboxes clean. Can you help?”
Holy smokes, did she help – she gave me insights about my customers that seem obvious in hindsight, but I wouldn’t have figured out otherwise. It was just awesome.
Now, you can use our data to help you learn. Part of our agreement was that she could use my anonymized data in her own workshops. You get the same source data, and alongside Steph, you will:
- Consolidate multiple data sources into single dataset ideal for trying to predict who will buy training.
- Decide the right sampling strategies, so we can build predictions and test them.
- Build different types of models to see which things influence whether someone will buy training.
- With multiple models on our hands, which one is the best fit with reality? Use evaluation techniques to identify the model that best identifies Brent’s future customers.
- Use the model to make some predictions to improve Brent Ozar Unlimited’s marketing strategies.
This isn’t some abstract, meaningless data set – it’s real data, solving real business problems. I know you’re gonna love this – go check it out.
