Status quo
The following status quo might scare and upset many people, if you just want to know how Summer does it (i.e. the happy path) feel free to skip ahead
Imagine you run an online ice cream shop and you want to answer some questions about your business.
- How much revenue did I have last month?
- What flavors are people who received my last email preferring?
- Is my checkout causing churn?
In order to answer these questions, your company needs a data stack.
What is a data stack?
A data stack is the combination of all the data tools you need to answer your pressing ice cream performance questions. The three most common data tools you’ll see are:
ETL (Extract, Transform, Load) — Move data from its source to your warehouse
Warehouse — Store data and process queries
Business intelligence — Write queries, build charts, make dashboards
In addition to these pillars, data stacks will typically have additional tools for governance, cataloging, reverse ETL, dedicated transformations, and more.
Shoestrings and bubble gum
Let’s now review the steps to build your data stack,
- Evaluate features, negotiate price, and select each tool
- Integrate all the tools together
- Train your team on how to use each tool
- Maintain, renegotiate annually, repeat step 1 as needed
Don’t know how to do this? That’s ok, there are a few very expensive engineers or consulting firms you might be able to hire who can do it for you. Oh whoops, you just learned that your ETL tool doesn’t support a source you need? No worries, see step 1.
Ice cream?
We’ve lost sight of what’s important, ice cream! Months of time and hundreds of thousands of dollars later, we’re finally ready to start speaking to our data.
If this sounds as gross to you as it does to us, check out how we’re making data a pleasure.