Episode transcript The transcript is generated automatically by Podscribe, Sonix, Otter and other electronic transcription services.
Ricardo (4s): Hello, everyone. Welcome to the five minutes podcast Today I like to wrap up the podcast series about the Spotify model. And I want to talk about failure. If we are basing our Model in a cultural change in interest with a very strong autonomy. Do you agree with me that this lack of control will allow people to be more independent and naturally I need to accept failure, because people will fail and what Spotify did to sort this out, they created the concept of fail fast and learn fast.
Ricardo (44s): They said, look, we accept failure. And we are providing an environment that is psychologically safe for people to fail. What is important that if you fail you need to learn at the same speed, so you don't do the mistake again. And that's some point when I was watching and reading about them, there are some points that they take this to the extreme, they celebrate failure. And I need to be honest with you. I respect very much the celebration of failure, but I'm still struggling, you know, to see how people can celebrate. I think that a failure is, is wonderful as a learning opportunity too, but, you know, celebrate it's a little bit too much for me, but they do.
Ricardo (1m 28s): And they are very good in their business. So things are working there, but you know what they do, they create a fail wall and they use that concept of guilt I said last week about the community to share these failures, so people are not shy, people don't fear, punishment. People know that this is part of the process, but I have a question for you and Spotify address, the answer to this question in a very nice way. So imagine if you fail and you destroy the platform, or if you put Spotify out of air four a week, then, it's a big, a big mistake.
Ricardo (2m 12s): And imagine, look, let's forget. Spotify imagine that you work on defense or military software and you are developing a control system for nuclear devices and they fail. So this is not acceptable, right? This is not accepted. What they did. They create an off limited blast, means you can fail, but your failure will not jeopardize, the whole operation. So maybe it will affect a group of users. Maybe it will affect a functionality, but it will not destroy everything. And they use the concept of lean startup to do that.
Ricardo (2m 55s): So they create an environment where they are much more focused on recovery than in avoidance. They know that they cannot avoid if they want to innovate. So basically what they do, they take in, an idea. They create a narrative and a prototype, a minimum viable product, or what they call minimal lovable product. And they distribute this into a limited group of people. So they create these releases almost independently to a group of people or to a country or to a sector or to a feature. And they start testing. And with that, they have data to make two decisions, first is the product really adding value or not.
Ricardo (3m 39s): Should I invest more? So instead of creating massive releases, I released on beat sizes much more frequently. So I released today. If you remember software development in the past. It's something like you release a software. Then two years later, you release again. Then three years later, you release again. Now it's not that anymore. You release almost every day. And you don't need to wait for a long time to release things. And this is the process they do. And the test at the same time, the acceptance and potential problems. So when they finished the full release, they know exactly what is the success factor of that new feature or that new playlist story, the new search engine.
Ricardo (4m 26s): And they know also, do you have bugs? So they release, you know, they create this limited damage control because if something goes wrong, maybe they will make some people unhappy or some features unavailable, but they will not destroy everything because they built the platforms that allow you to do a coupled release instead of everything connected, you create the interface, but do you create a very high level of independence between the packages, right? They said that they have more than a hundred different features. And I think that a good Model to explain this is like Lego bricks. You know, you have the basic connections that connect the pieces, but each piece has a different shape and a different intent.
Ricardo (5m 11s): And that is, this is the way they built. Finally, I don't want here to convince you that the Spotify model is the best model of the earth, because I'm not competent enough to tell. And I don't work for Spotify to tell this, however, what I want to tell you that Spotify took the deliberate decision, to adapt the process too, their culture or the culture do they want to be built and this is what matters for me. So don't take a standard and say, Oh, I need to do exactly what is in this standard. this it's not adding value. So they, are ready to adapt and your company must be ready to adapt to combine agile methods, to combine with whatever that works.
Ricardo (5m 60s): And these for me is the magic of what is Spotify. They created their own vocabulary they created the concept of tribes squad, and they create the whole culture around failure, around motivation, around ownership that made Spotify be who Spotify is today. And this is exactly what you should aim for your company instead of just taking a book and implementing from page one to page a thousand. Think about that and see you next week with another five minutes. Podcast.