Conor's Newsletter: Issue 47
Fantastic startup failures, causal conversion prediction, and the rise of Reddit
Hey there šš¼
Weāre on Substack now! While I generally like Mailchimp, their business model was getting further and further away from what I want to do with this newsletter: share things I find interesting with like-minded readers.
Let's get to links that I thought were worth sharing this week. As always, enjoy the newsletter and let me know what you think! If this was forwarded to you, check out some previous issues and subscribe for future updates.
š What I Wrote
Iām back at it! One of my goals in Q1 2020 is to write more. I thought it was fitting to start with a look back at last year. I really enjoyed reflecting while I put this together. Donāt hesitate to reach out with any thoughts. 2019 Year in Review.
š What I Read
āAbout 15 months ago, I left my full-time job as a machine learning team lead with the goal of doing independent/freelance data science consulting. Since then, Iāve gotten a lot of questions about what that means and entails.ā This is the more comprehensive post Iāve found on the topic. Doing Freelance Data Science Consulting.
I always enjoy reading Andrew Chenās posts on product and growth. This collection of his tweetstorms and essays from last year had some pretty interesting points. Product Tweetstorms from 2019.
āAt a high-level, Wizard is a tool to help us understand what is driving changes in conversion this month. It is interpretable by attributing pp step changes in conversion to pp step changes in specific drivers.ā This makes too much sense. I want it. Wizard: Our ML Tool for Causal Conversion Predictions.
Taking a step back to the basics, communication around data problems is a tough nut to crack. This post looks at interpreting problems like a grade-schooler, and itās a really helpful reminder. How to Solve a Business Problem Using Data.
Making things people want isnāt enough. Your product has to benefit all parties involved in the transaction for it to work. This is a perfect example. Itās well-written and should get a chuckle or two. I Wasted $40k on a Fantastic Startup Idea.
š„ What I Found Interesting
It seems like a whole suite of productivity and enterprise tools have risen out of taking things that developers were doing, and making them more accessible or exposing them to the ānormal userā in creative ways. How We'll All Work Like Developers.
I love when makers share open stats about products, especially in relation to launch. Daniel Vassallo is a good follow if you enjoy the same. Check out his tweetstorm. Launching An eBook.
This visualization of social media platforms and their relative growth is telling. Reddit and Instagram are notably up and to the right. Social Media Rise and Fall.
š¤ Quote I'm Pondering
"If you are building the wrong thing, optimizing the product or its marketing will not yield significant results. A startup has to measure progress against a high bar: evidence that a sustainable business can be built around its products or services." ā The Lean Startup
Thanks for reading! Iāll see you next week at the usual time. If youāre finding this newsletter valuable, pleaseĀ share it with your friends.
Until next time,
Conor š§āš»