c a n d l a n d . n e t

Experiments are good, bets are better?

Dusty Candland | | bets

I think most people would agree the best way to learn is through experiments, some might say the "hard way", but I'm starting to believe it's actually the only way.

Experiments produce experience and that's where the value comes from. I've spent way too much time reading how to do something but I don't feel like I've actually learned how to do something until actually doing it.

When people ask me how to learn to code. I respond with you just have to get started. Learn code the hard way by Zed Shaw is where I send them. It's called the hard way, because you have to start at the beginning and work through them by actually coding.

Assuming experiments are required, how do I prioritize and create experiments that are valuable? Or rather make sure the learning is worth the cost?

Bets. Everything has a cost. Costs are easy to overlook. Time is the one I overlook most often. When making bigger decisions I remember to think about opportunity cost, usually as time, but for smaller decisions it's overlooked.

Monetary costs are more visible, but I often struggle knowing if it's worth spending the money. I think that's because I haven't thought about it as an experiment, rather, just a cost.

I've said "I'm going to experiment with SEO", but I didn't define a test. Making it less of an experiment and more of an excuse to do what I wanted.

Bets, and experiments, need a test. How I'm testing the hypothesis? How do I know if the bet was successful or not?

A test implies measurement and a time limit. Measurement for success. A time limit to calculate costs and define an end point.

I can't bet that a sports ball team will win a game without some time frame. There has to be an end defined. A team can't win without a measurement or score.

Bets should be win-win. If the bet works, the upside should far outweigh the cost. And when the bet doesn't work, the downside should be worth more than the cost. The second outcome can be subjective and requires reviewing what happened and why.

Downsides can be minimized. For example learning a new skill or system can be valuable in addition to learning what didn't work.

The upside and downside needs to be defined so the cost can be weighed.

The lottery would be a bad bet. If it works, I win the lottery, it has a big upside. However, it's way more likely I lose. In that case, I've just lost money.

However, if I'm trying to learn how the lottery works, that can minimize the downside. And might make the cost worth it, turning a bad bed into a good one.

So I need a hypothesis, a test, an upside and downside, and a cost. Given that I can decide if the bet is win-win. Or said another way, is the downside worth the cost?

The meta bet(s)

Hypothesis: Working in bets will produce growth in my products.

Test: Work using bets for 2 months and measure product growth.

Cost: 90 minutes per bet. 60 minutes to define the bet and 30 minutes to review the results.

Upside: Working in bets dramatically improves growth.

Downside: I learn bets are not helpful for growth. I might find them valuable for other reasons. It will provide an interesting blog post.

Hypothesis: Working using bets in public will keep me accountable.

Test: Post the bets I'm working on to my blog, Indie Hackers, and a Medium blog for 2 months. See if I follow up/review each bet.

Cost: 2 hours per bet. 30 minute to write and post the bet. 90 minutes to write the review and results.

Upside: I'm accountable and get the most learning out of these bets.

Downside: I learn I need a different way to hold myself accountable. I force myself to post on progress more often (which is hard for me). Learn if there are other benefits posting about my work.


This is something I've been thinking about influenced by these books and posts.

Photo by Michał Parzuchowski on Unsplash


These are webmentions via the IndieWeb and webmention.io. Mention this post from your site: