In just a month, over 135,000 people read my article. That’s more than twice as many people who live in the county where I grew up. I am humbled and amazed at the response. Thank you to everyone who read it, shared it, and discussed it.
My article was about Spotify-the-model, not Spotify-the-company. Despite the subheadline “Spotify doesn’t use ‘the Spotify model’”, some people seemed unable to disconnect the two.
I intentionally did not discuss Spotify-the-company beyond Spotify-the-model in my article. That would be airing the company’s dirty laundry publicly, which I refuse to do. I will only say that few of my former Spotify coworkers who have experience working at other successful tech companies would describe Spotify’s actual way of working as ideal or desirable, but it is improving.
I will let Spotify speak for itself in describing how it actually works. I think Spotify has not shared its current way of working publicly because it’s still figuring it out and the methods that are working are refreshingly unoriginal. There is no opportunity for Spotify to be a thoughtleader on process again, but there is an opportunity for Spotify to set the record straight.
About the cover
I received 3 comments from people who disliked the cover image. They all seemed to miss the reference to Taylor Swift’s excellent Bad Blood music video about squad goals. They also seemed to not read the explaination at the end of the article.
Some people hated the editorial design
I believe in the value editorial design brings to the principal content. This was my first attempt to bring magazine-style layouts to the Web using modern CSS layout features. Some people did not like it. Some people did. I welcome constructive criticism. It’s ok if you don’t like it. I create for myself and not everything has to be for you.
Other Spotify employees react
Nice article! I was at Spotify for almost 7 years, and we worked in the same tribe :). I've had this chat many times before, with startup people in tech conferences who came to talk about the model with enthusiasm after seeing my Spotify t-shirt. 😁— Felipe Ribeiro Barbosa (@felipernb) April 27, 2020
I left Spotify in 2019—in large part—because of all the problems identified in this article.— Allan (@ayyyylo) April 24, 2020
To describe Spotify’s product development as “glacial” would be an insult to glaciers.
Well worth a read. https://t.co/k6WjZn5MMJ
Я работал там 6 лет и могу подтвердить что чувак чертовски правhttps://t.co/0gf8AHq8uD— Supremacy of Pure Feeling (@GoblinGame) April 25, 2020
(Translation: “I worked there for 6 years and I can confirm that the dude is damn right”)
People shared how “The Spotify Model” failed them too.
Zalando also played with similar model in 2016 but (relatively) quickly found it doesn’t work well.— Boring humorist (@chilicoder) April 23, 2020
Much of this resonates, from a period when Typeform tried to emulate the Spotify model, and we ended up meeting many other companies who had tried and failed...— J(a)son Harmon (@jharmn) April 20, 2020
Enjoyed this article on the drawbacks of the Spotify Model by @JeremiahLee, we've tried this at BT with mixed success so I found the recommendations valuable.— James Trendell (@jamestrendell) April 29, 2020
Insightful requiem for the #Spotify model - autonomous teams did not work at #volvo experiment at Udevalla either - my reflection on visiting Spotify was there was no thought to capturing knowledge and learning - but a thought provoking visit!— Daniel T Jones (@DanielJonesLean) April 26, 2020
In one of my prior roles they were running the full Spotify model when I joined. I pushed us to kill for it two of the reasons in this article. 1) I am not a fan of matrix management. It makes ownership unclear and makes manager-employee relationships difficult. It’s hard to have meaningful conversations when your manager really isn’t focused on your actual work. 2) the assumed collaboration/coordination capability. Coordination and collaboration are hard - one of the reasons I’m a huge supporter of the Technical Program Manager role. These folks provided the dedicated ownership of cross team activities to make sure stuff gets done - so it isn’t left to good intentions.—Asanka Jayasuriya, CTO at SailPoint April 20, 2020
I worked at a startup that switched to #squads pretty soon after Spotify blogged about them. This article is entirely consistent with that experience. 1/5— Braedon (@braedon) May 1, 2020
It was a disaster. Months were lost. We ended up with mostly what we had before, but with shiny new terms, less accountability, poorer decision making, and a ton of relational capital burnt, both between individuals, and between engineering and management. 2/5— Braedon (@braedon) May 1, 2020
I have talked to a lot of people who were burned by VPs of Engineering reading Spotify blog posts and deciding they could fix internal orgnization. (Me included)— Tim Hopper (@tdhopper) April 20, 2020
If you're ever tempted to try out the Spotify "squads" model make sure to read this first - it's uncanny how much it matches experiences from people I know— Simon Willison (@simonw) April 24, 2020
People noted the harm the idea had done to the industry
The fact that the vast majority of both technical and organizional blog posts by startups are "aspirational" rather than reflecting reality is part of the reason our industry is so full of bad ideas and bullshit https://t.co/Yy7OqxTsSR— Camille Fournier (@skamille) April 24, 2020
Anyway, whenever you read a company blog post about the amazing new process/methodology/structure they're using (or any company blog really), head to your nearest ocean, and evaporate yourself the largest pinch of salt you can. 5/5— Braedon (@braedon) May 1, 2020
Read this before cargo-cutting into your own organisation’s Spotify model. I was amazed when a tribe lead from Spotify said: “You guys seem to be using the Spotify model, don’t! At Spotify we moved away from it.” After a year I understand why they did and this post explains it. https://t.co/jPeVQZlbeW— Alla Babkina (@allababkina) April 23, 2020
- Marty Cagan agreed with me and then took a hard turn into crazy town
- CTO of Groupe leboncoin shared he thinks the model can work, even though they don't use part of the model anymore