Last Friday’s news that Nest CEO Tony Fadell would be leaving the company he founded with Matt Rogers and stepping into an "advisory" role seemed like the culmination of months of stories about Nest’s demanding culture — particularly the frank displeasure of former Dropcam CEO Greg Duffy, who openly regretted selling his company to Nest. These reports have largely focused on Fadell, whose management style has been polarizing. But another dynamic playing out may have been even more important, according to interviews with insiders: Google's restructuring into Alphabet last year, which placed new financial pressures on Nest to perform that some say limited its ability to innovate.

Fadell has said he made the decision to leave Nest last year, and sources tell me he wrestled with the idea of leaving Nest for weeks before telling Alphabet CEO Larry Page of his plans in December. The impetus, according to several sources close to Nest, was the increased pressure on Nest to deliver profitable results as a standalone unit inside the new Alphabet operating structure, instead of being safely ensconced inside Google and given room to grow. Fadell had been turned into a manager tasked with steadily growing his businesses instead of serving as a visionary CEO, and ultimately he walked away.

Fadell won’t say much about his decision to leave Nest, but he’s investing in over 100 startups, including the smart go-kart company Actev Motors. He told me that his goal now is "to find more disruptive technologies and entrepreneurs and support them with time and money, and my network and experience, and to start other companies."

"I want to leverage everything I’ve done and the expertise that I have. It’s not that I can do everything myself anymore. There are great innovations, and I want to help those innovations come to light, to really change the world in a great way. That’s what it’s really all about."

But Nest was Fadell’s baby, and his vision. So how did the switch to Alphabet change things so drastically?

Nest was among the first smart devices companies to make a splash. The company launched the Nest Learning Thermostat in 2011 to rave reviews — Fadell’s pedigree working on the iPod and iPhone lines at Apple shone through in the design and simplicity of the product, and his former Apple intern Rogers was an impressive young engineering talent bursting with enthusiasm and fresh ideas. And Nest as a company was early to the idea of meshing smart services with hardware: the thermostat was coupled to a sophisticated big data operation that continually made each individual thermostat even smarter, while amassing a valuable pool of information that Nest was able to turn into a business serving utility companies. In some markets, utilities give Nest thermostats away at steep discounts or even for free, because the energy they save and the data they collect is so valuable.

That combination of business models — hardware revenue plus service revenue — was at the heart of the Nest promise. "It felt like Tony and his vision could really create the next great hardware services company in the Valley," says Randy Komisar, the Kleiner Perkins Caufield Byers partner who led that firm’s investment in Nest and served on its board. "It was a platform that Tony and Matt could have operated for decades, like Amazon is for Bezos, or Apple was for Jobs, or Google is for Larry."

But in January 2014 Google came calling with what eventually became a $3.2 billion offer to buy Nest. The decision to sell was controversial; Komisar was against it, warning that big companies were liable to change strategy in a way the Nest team would dislike. But Fadell and Rogers were taken with the idea of using Google’s resources to accelerate their plans outside of the pressures of being a venture-backed firm. The main selling point, according to multiple sources close to Nest, was that Nest would be given a long runway to grow inside of Google, modeled on the company’s acquisition of YouTube, which took a decade to pay off. And Google’s purchase was widely seen both inside and out of Nest as Google investing in hardware to ultimately take on Apple. "I thought Google was genius to buy this company at the price they did," says Komisar. "They got Tony Fadell, they got Matt Rogers, and they got the DNA from Apple hardware. Short of buying Apple, you’re not going to get that DNA in one big effective group that was performing. They bought a terrific asset."

Initially, the acquisition went well — Nest began hiring at impressive rates. "My general impression was that Google was going to be extremely patient and allow them to broaden their reach," says Komisar. "Google was embracing what it took to be the YouTube of hardware."

A bump in the road was the acquisition of Dropcam in June of 2014, which happened at the behest of Google, according to a report in The Information. Dropcam was a small company that had attracted loyal customers to its elegant security camera products, and it had a roadmap to ship several other products. But integrating the team proved to be difficult; Fadell’s demanding style didn’t mesh well with Dropcam CEO Greg Duffy’s more laid-back manner. Fadell blamed Dropcam, telling The Information in March that "a lot of the employees were not as good as we hoped." Duffy fired back on Medium, saying that Nest’s management "seem to be fetishizing only the most superfluous and negative traits of their mentors," clearly implying that Fadell was emulating the worst of Steve Jobs. "I felt like I had failed all the people who worked for me and all the customers," by selling to Nest, Duffy told The Information.

These stories stung more because Nest’s wasn’t launching new hardware categories — although the company shipped new generations of each of its products and multiple updates to its app in the past year, the failure to launch obvious extensions of existing products like an outdoor camera or enter entirely new markets seemed to indicate that the company had stalled.

But Nest insists that the roadmap is strong, and that new products are due out soon. And Duffy left Nest in January of 2015 after just eight months at the company; various sources at Nest indicate that the Dropcam integration was a small issue amplified after The Information ran its story, and that Nest didn’t want to enter a PR war. And by that time Fadell had already decided to leave Nest. The far bigger change in Google’s strategy that Komisar had warned about had come about, as the company radically restructured itself into a holding company called Alphabet.

The Alphabet restructuring pulled several units out of Google and turned them into separate divisions under a single corporate umbrella. Google itself became the largest and most profitable division of Alphabet. Nest, reporting under a new division called "Other Bets," was the company’s other major consumer business, and the dynamics at Nest changed dramatically. According to Alphabet’s Q1 financial conference call, Nest forms the bulk of Other Bets revenue, along with life-sciences company Verily and Google Fiber. Google and Alphabet declined to comment on how these changes affected Nest.

Alphabet money is not like Google money, according to several sources at Nest. Whereas Google was content to float the company, under Alphabet Nest was tightly constrained and asked to demonstrate a level of financial discipline at odds with what the founders had expected when they sold to Google. For a five-year-old hardware startup in a still-unproven market, that meant Fadell’s role immediately changed — and that Nest’s conversations with their corporate parent went from being focused on growth and investment in technology to what one source called "effectively finance meetings."

That pressure may have led to the slow pace of new product introductions. "At any time in hardware I know how to get profitable, which is that I quit funding new products," says Komisar. "They went to Google to fund new products."

Ultimately the changing priorities led Fadell to leave, and Alphabet to install Marwan Fawaz as the new CEO of Nest. Fawaz worked as the CTO of Charter and Adelphia Cable before spending time at Motorola Home, and his reputation is for calmly running and scaling businesses. His job, according to sources, is to make hard decisions about Nest’s investments and refocus the company. No one seems to think Alphabet is planning to sell Nest or pull it back into Google; the plan seems to be to let Fawaz focus on sales and business, while Rogers takes over product development. Alphabet still intends to invest in Nest and for it to grow into a large, innovative company, but it will take a much different path than its founder anticipated.

"Nest has an incredible business," says Fadell. "It's growing really well, it's got an incredible team, it’s a leader in each of the categories it's in, and as long as the team continues, they’re going to be successful. They have a great roadmap that’s in place today for the next two years — they just need to deliver, and I’m sure they will because they have great talent."

But ultimately the question is whether the Alphabet experiment can allow other new bets to grow alongside Google when it’s clear the structure made it very difficult for Fadell and Nest to transition from being a fast-moving startup to a division of a larger corporate entity. And the integration between Google — which just launched the Google Home intelligent speaker — and Nest will need to get tighter, even as Nest remains distinct from Google inside the Alphabet structure. It’s going to be tricky, and demand a level of corporate savvy that doesn’t seem natural to the hard-driving Fadell.

"Tony at Alphabet was paying Google employees at Google rates with Google benefits. He was paying Google rents. This is not what a startup does," says Komisar. "The real story is Alphabet. This isn’t really about Tony Fadell. This isn’t really about Nest. Nest is still full of potential. This is about Google, and Google’s decision to build Alphabet."

"Larry and [Google CFO] Ruth [Porat] are smart people, and they know what they’re doing," Komisar continues. "It’s their prerogative. But Nest [inside Google] could have been a powerhouse and a half."

"Alphabet is a work in progress," says Fadell. "There’s many many great things about Alphabet and there’s a lot of learnings that are going on right now." He won’t say anything else.