John List
speaker
84 appearances
2 recordings
2 series
first heard Dec 2024
last heard Apr 2025
John List’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
an idea because in the past it was move fast and break things, throw spaghetti against the wall, whatever sticks, cook it, fake it till you make it. We've all heard these statements, but it's always art. And what I'm adding here is we need science to figure out which ideas are truly scalable.
So here's kind of what happened. There was an innovation of the smart thermostat. Put it in your house. It will moderate your energy use over 24 hour day. And the engineers estimated that we will have huge savings because of the smart thermostat. So here's what we did. We chose 400,000 households in California and we sent 200,000 of them the smart thermostat.
In the other 200,000, we didn't send anything and we just observed all 400,000 households. Guess what happens? There is zero energy savings. Now you can say, well, wait a second. Why did the engineers so dramatically overestimate what would happen? Well, here's why. They assumed that the end user was Commander Spock. This is a guy who never makes a mistake, and he is 100% rational.
That's not who the end users are of this product. The end users are more like Homer Simpson. Homer Simpson is exactly what I did. I got this new gadget. I did not read the 28 page manual, how to use it. And then I went in and fiddled with the new gadget and I undid all of the presets. I undid all of the defaults. That's what the 200,000 households did in California.
They undid them exactly enough on average to undo all of the good stuff with the technology. That's knowing the situation. Know who's going to use your product and give them technology that they can actually use that will help save the Earth.
That's why that idea did not scale from the Petri dish to the large, because we did not understand the end users are very different than Commander Spock from Star Trek.
And let me be clear, they're working for some people. So when I say on average, it doesn't work, that doesn't mean that it's not working for some people. And you're right. What happens now is you come to smart thermostat level two and then level three, and that product will evolve to be much more user-friendly. When we put, I used to be the chief economist at Lyft,
And when we put out products there, whether it's called walk and save or wait and save, there's always some beta testing and then some evolution to make sure people understand the new product. The general idea here, though, is you might have a voltage drop. if your estimates from the beginning were that all the users are going to 100% understand and use a technology as you estimated to start out.
And in many cases, that doesn't happen. And that's just a general idea of a voltage drop.
If you don't execute on the idea, if you're not a good manager and make good decisions, the best idea won't work. But on the other hand, if you're 100% at execution, if you're trying to scale an idea that doesn't have the good signatures, that won't scale either. So there are ideas all over the world that look good on paper, But because of bad execution, they just don't scale.
many restaurants. When you think about Jamie Oliver, that was really bad execution. So Jamie's restaurants in the UK were really good early on and the execution ended up faltering. And there are many cases like that. I think Sears and Kmart were a bit like that too. Look at the 1955 Fortune 500 companies. In 1955, if you look at those 500 companies, only 70 of those are around today.
Now, in many cases, those are firms like Studebaker, Zenith, et cetera. They just did not pivot. And when the times caused you to change, they refuse to change. Blockbuster is much the same way. And in many cases, when markets change, if you're not pivoting, this is a lack of execution. The idea looked good early on, but then the market changes and you're not pivoting with it.
So the world is replete with examples now where Folks are not using data or not using new elements of data science to help make decisions. And I would say that those firms are endangered. Blockbuster could have purchased Netflix for pennies on the dollar had they wanted to, but they thought that's not the future.
They could have easily pivoted in that direction and said, we're going to do both and we're going to be diversified, but they didn't. When you look at many firms, they start doing something very different than what they become.
Let's think about a pricing idea that I had, and this might be familiar to some of your listeners, and it's called left digit bias pricing. And the idea is that as humans, we engage in shortcuts. And when we see a number, we focus on the left most digit. So what does that mean? People open up their apps and say they get a price of $7.93.
And then they make a decision whether to take the trip in Lyft or in an Uber. Now, that decision is not very different than if they receive a price of $7.94. Okay. Because humans see 794 as the same as 793. Now let's say I change that just a bit and I give you a price of 799 versus a price of $8.
That's basically the same as the first example, but now because people are focusing on the seven or the eight, that penny difference makes a big difference in your decision. Okay, so I tested that idea using Lyft data and using Lyft, a big field experiment on Lyft. And I find that that one penny change makes a big difference. And I test it again and I find it's not a false positive. Okay.
I then go to the next step to say, who does this work for? It ends up working for everyone. Everyone has this left digit bias. Okay. Now let's go to the situation. You have business travelers, people travel in morning rush hour, afternoon rush hour, going to the airport. It works in all of those situations.
So now that's great because I can control in a better way using pricing in this behavioral bias. So I can scale that up and I can effectively impact people's choices because of this simple left digit bias rule, scalable, It works. It's not costly to do it. Because remember, we have to give you a price anyway. What does it matter if I give you a price of $7.93, $7.99, or $8?
So the supply side of it is great. And the spillovers are great too, because I can control whether you get it or not. So if I'm on the bad end of a market, what I mean by bad end is there are a lot more drivers than consumers. I can move consumption. Or if it's more consumers and drivers, I can move consumption.
Showing 61–80 of 84 · page 4 of 5
← Previous
Next →