Tom Conrad is sitting in Sonos’ Santa Barbara headquarters on a sunny Californian afternoon, talking about the worst year in the audio company’s two-decade history with the calm of a man who has already made peace with the mess he inherited – and started cleaning it up.
“The company would never have shipped the software if they’d known how it was going to perform in the real world,” he tells this masthead in an interview ahead of the launch of two new speakers. “And so, unfortunately, another misstep was that the company was not well prepared to roll back the decision once they realised, and so the only option they had was to try to fix it.”
In May 2024, Sonos pushed a disastrous app update that broke multi-room speaker systems across millions of homes. Features vanished, volume controls malfunctioned and sleep timers disappeared. Customers who had spent hundreds or thousands of dollars building out Sonos systems – the kind of loyal evangelists most brands would kill for – were furious. Social media turned toxic with the crisis quickly earning the label of the “App-ocalypse”, sales slowed and in January 2025, chief executive Patrick Spence stepped down.
Enter Conrad: a Sonos board member since 2017, co-creator of the streaming service Pandora, and a veteran of some of Silicon Valley’s most spectacular flame-outs, including a stint at Pets.com during the dot-com crash and a run as chief product officer at Quibi, the short-form streaming experiment that lasted barely six months. He is, in other words, no stranger to companies in crisis.
He took the interim chief executive role in January 2025, was confirmed as permanent chief executive in July, and has spent the past 14 months doing what he describes as “righting the ship”.
The turnaround has been methodical. Conrad made leadership changes, eliminated management layers, ended programs the company had spread itself too thin in pursuing, and delivered more than 50 software updates in 2025 alone, each one chipping away at the reliability problems that had plagued the platform. He restructured the product organisation so that, for the first time, there were dedicated, properly staffed teams focused purely on software.
“We brought a lot of science to it,” he says. “We deeply understand, quantitatively, how the software is performing in the field, and every week, we address the issues that are most impacting performance and reliability. One issue at a time.”
When asked what exactly went so wrong for Sonos, Conrad identifies three compounding mistakes: the company changed the app’s user experience too dramatically; it launched without feature parity, reasoning that only a small fraction of users relied on things like alarm functions; and – crucially – it simply did not understand how the software would perform in the messy, heterogeneous reality of customers’ home networks.
“They comforted themselves saying, ‘Oh, only 1 per cent of people use the alarm function, we’ll add it back in time’,” Conrad says. “Well, 1 per cent of 17 million households is 170,000 people, and it definitely takes less than 170,000 people angry on Reddit to make a lot of noise.”
Conrad is open and honest about the failings; it’s the kind of post-mortem most companies bury in internal reviews. The executive, who was on the board when the app shipped, is not trying to distance himself. “I feel absolutely responsible,” he has said elsewhere. “This happened, at least in part, on my watch.”
The good news, he says, is that the chapter is now in the rearview mirror. Sonos’ share price is up 16 per cent over the past year. Internal metrics show the platform performing better than it has in years, including before the botched update. Community sentiment on Reddit has shifted from vitriol back to users comparing speaker setups and congratulating each other on new purchases. And now Conrad is ready to start the next chapter: new hardware that reinforces what he believes makes Sonos unique.
This week, Sonos launched two new speakers: Sonos Play and the Era 100 SL. Conrad is pitching them not as standalone gadgets but as entry points into what he wants customers to think of as a single product: the Sonos home sound system.
“One of the first things I said to the team when I took this job is that I think we make really just one product, which is the Sonos sound system for the home,” he says. “Any individual device is just a way into the system, or a way to deepen your experience once you’re in. And I think, over the last handful of years, we kind of lost our way on that.”
Sonos Play is a deliberate callback to the Play:1, the speaker Sonos introduced 13 years ago that effectively invented the internet-connected home speaker category. Tens of millions were sold; nine out of 10 are still in use, Conrad says. The new Play is a stereo speaker with dual tweeters, a dedicated mid-woofer, Wi-Fi 6, Bluetooth, 24 hours of battery life, an IP67 waterproof rating, and a wireless charging base so it can live on a kitchen shelf and be carried outside for a dinner party. It is also portable in a way previous Sonos speakers were not – with a utility loop and a built-in power bank that can charge your phone.
The Era 100 SL, meanwhile, is a stripped-back, microphone-free version of the existing Era 100, priced at $289 in Australia – about $30 less than the current model. It is designed to be the simplest, cheapest way into the Sonos ecosystem.
Both products are currently available for pre-order, with general availability on March 31.
The strategy seems conservative, and deliberately so. Under former chief executive, Spence, Sonos had been branching out: headphones, a leaked-then-killed streaming box, a partnership with IKEA that was recently ended. Conrad is pulling the company back to its core.
“We really are a category of one,” he says. “There are not other companies that do these things. And so, as frustrated as customers were, the vast majority of them, even the ones who were the most acutely impacted, stayed with us.”
Still, the trust deficit is real. Conrad tells a story he has clearly been carrying around for months: a customer whose parents were celebrating their 50th wedding anniversary. The Sonos system did not work for the party. The customer was livid.
“I told him, I totally get it, and we just can’t let that happen,” Conrad says. “It’s not just a speaker. It’s a soundtrack.”
Looking further ahead, Conrad is thinking about artificial intelligence, including a potential role as a platform-agnostic hub for conversational AI agents in the home.
“Just like we were the level playing field for hundreds of music service providers when we launched in 2005, you can imagine we could become the same kind of agnostic platform for different kinds of voice agents in the home,” he says.
It is an intriguing pitch, and one that would position Sonos as a rare neutral player in the AI assistant wars.
But, for now, Conrad is keeping his ambitions deliberately contained. He has told his board he is focused on the next 18 to 24 months. He has plenty of ideas about where Sonos might go in five or 10 years, he says, but after the chaos of 2024, the priority is simple: make the thing work, and make it work every time.
“I feel really lucky to be so deeply woven into significant moments in people’s lives,” he says. “You have to take that really seriously.”
Get news and reviews on technology, gadgets and gaming in our Technology newsletter every Friday. Sign up here.
