Getting Weird with Squarely

Sanity Check • No. 018

I’m Ben, and welcome to Sanity Check. The newsletter for tips, stories, doodles, and even some questions about working in the data field. Glad you’re here.

---

FRESH FEATURE

Getting Weird with Squarely

A friend recently shared an article - “Do the weirdest thing that feels right.” The article is great, but the key takeaway is right there in the title.

A decision needs to be made. Thinking through the options, the responses range from “sounds good” to “that’d be okay.” No clear right or wrong. Just an array of equally reasonable paths forward. In those situations, what’s your tiebreaker? For the author, he chooses the “weirdest” thing.

I am faced with a decision. How can I best develop my craft, empower others to do the same, and use that power to grow their business?

My array of options seems equally reasonable.

• Generalize then share - take my experiences and strip them down to the root lessons. This is fine and works well in textbooks, but it lacks narrative and character to help the lessons stick with people.

• Set up fake scenarios then share - courses I’ve taken in the past do this well. Adam Wathan’s Test Driven Laravel imagined “Ticket Beast” and the Uplimit Advanced dbt course has “Bingeflix”. This approach lets you get into the details, but takes a lot of work to setup.

• Share an inside look at a real business - this would be ideal, but who would be willing to open up? I could use my own small business, but that feels too much like a crummy growth hacker on X.

Now my choice is clear weird… I’ll share the business of puzzles!
Not just any puzzle — but a new category of puzzles.
Squarely Puzzles!

This makes no sense!

• My dad created Squarely as a passion project, not as a business.

• There are no revenue expectations for Squarely — and that’s a good thing because the unit economics are bad.

• It will be complete overkill to build out an analytics suite this early.

Yet it does for all the right reasons.

• I have permission to share details that established businesses wouldn’t be publicly transparent about.

• The issues we come across with Squarely will be issues any eCommerce company will face. This makes for good learnings for you all 🍿

• If this works helps out in any way I know I am helping my family.

So stay tuned for this weird longitudinal case study on breaking into the puzzle market!

---

QQ: QUICK QUESTION

An issue I could use your help solving

How do you keep a data lake organized?

With the rise of lakehouse architectures, the starting point for analytics data is a filesystem — not a schema full of tables.

How do you get to that mind-blowing experience of select * from read_parquet(‘users/*’)?

---

CURATED COLUMNS

A few interesting articles, podcasts, or websites I recently came across

Becoming Pangea - I’m a loyal reader of the AE Roundup, but this past week’s was exceptionally thought-provoking.

• Hamilton’s 7-powers and which data companies have staying power

• Developing standards that become foundational to successive technologies

• The delicate relationship between hyperscalers, data platforms, and individual data products

This is all useful framing to see where the industry is going. If you can see ahead, then you can make informed bets that will pay off in your career and at your company.

The Road to Composable Data Systems - Wes McKinney’s article on Composable Data Systems covers several open-source data tools. Each tool is focused on opening up interoperability at different layers of data processing. Putting that into Hamilton’s 7-powers, these tools reduce switching costs between other data components.

The Open Data Stack - Pedram’s MDS Fest talk highlights just how good we have it. You can build out a powerful stack with a laptop and open-source software.

Now paying some money helps. You get to lift those processes off your computer and into a production-ready environment. However, given my interest in accessibility for new practitioners, I love to see this trend of open data stacks grow.

—

One more for fun…

---

Thank you for reading.

Let’s keep it going. 💜

If you enjoyed this edition, would you mind giving the heart below a click?