The Long Zigzag
The cleanest version of Matt Garratt's story starts in February 2023, when he walked into U.S. Venture Partners' offices on El Camino Real and became General Partner. But that version skips the aircraft designs. It skips the clean energy entrepreneurs he backed in Africa. It skips the UN policy papers. And it absolutely skips the fact that, when Salesforce Ventures was nothing more than an idea, Garratt was its first employee - the one who had to build the playbook from scratch.
His favorite maxim captures it cleanly: "The voyage of the best ship is a zigzag of a hundred tacks." For a man who started as a materials scientist, moved to renewable energy, created UN policy, and then somehow ended up as one of Silicon Valley's most consequential enterprise software investors - the zigzag is not a metaphor. It's a map.
Garratt's undergraduate degree from Purdue was in materials science and engineering. That's the discipline of understanding how physical matter behaves under stress - what makes metals fatigue, what makes composites fail, what makes a new material worth building with. He used it to design concepts for commercial aircraft. Then he went to Michigan Ross for a combined MBA and Master's in Natural Resources and Environment, became an Erb Fellow, did research for the United Nations, and helped stand up the Frankel Cleantech Fund - a student-run venture fund focused on clean energy.
The leap from clean energy policy to enterprise software venture capital is not obvious. But Garratt made it, joining Battery Ventures as a Vice President in 2012, where he focused on early-stage enterprise software and GreenIT companies. A year later, Salesforce came calling. They wanted to start a corporate venture arm. They needed someone to build it. They hired Garratt.
"The voyage of the best ship is a zigzag of a hundred tacks."- Matt Garratt's guiding philosophy on careers and investing
Building the Biggest in Tech
Eight years. One hundred-plus investments. A combined portfolio market cap that crossed $100 billion. When Garratt joined Salesforce Ventures in 2013 as its first employee, corporate venture arms were viewed in the industry with mild skepticism - a nice-to-have for big tech companies, not a serious player in the LP-backed fund ecosystem. By the time he left in 2021, Salesforce Ventures was the largest corporate venture arm in the technology industry. The skeptics had gone quiet.
The portfolio Garratt built during those years reads like a greatest hits of the cloud software era. Zoom. Snowflake. Twilio. MuleSoft. Anaplan. NCino. These were not lucky bets on obvious winners - they were early bets on category-defining companies when the categories were still being defined. Twilio was pitching developers on the idea that communication could be an API. Snowflake was telling enterprises they could run their data warehouse in the cloud. Zoom was promising that video conferencing could just... work.
Beyond the IPOs, the acquisition scorecard is equally striking. Gainsight - which effectively created the customer success management category - was acquired by Vista Equity Partners. ThousandEyes, the network intelligence company, went to Cisco. DocuSign went public and then some. Automation Anywhere became a cornerstone of the robotic process automation market.
What made Garratt effective at Salesforce was something that often gets lost in the corporate venture narrative: he was not investing to plug gaps in Salesforce's product roadmap. He was investing to build relationships, support founders, and generate returns - the same job description as any good VC. The strategic angle was real but it was the discipline of a financial investor that made the portfolio exceptional.
The Companies That Defined Cloud
"When you can measure what the software produces, the procurement conversation changes."- Matt Garratt on why AI-native software commands premium valuations
The Next Chapter
In June 2021, Garratt left Salesforce Ventures and joined CRV as a General Partner. The tenure was short - roughly eighteen months - but it produced results. He backed Power Finance, which was subsequently acquired by Marqeta, and Instil. Then, in February 2023, came the USVP announcement.
U.S. Venture Partners is one of the original Silicon Valley venture firms, founded in 1981. It has backed companies across four decades of technology cycles. Garratt joining as General Partner - the press release was headlined "USVP Doubles Down on Enterprise Franchise" - was both a signal about where he saw the market going and a significant bet by USVP on what the next wave of enterprise software would look like.
At USVP, the investment parameters are straightforward: checks ranging from $500K to $40M, with a sweet spot around $13M, focused on seed through Series B companies. The sectors are enterprise software, AI, analytics, climate tech, and fintech. The thesis is more specific than it sounds: Garratt is hunting for companies where the value delivered is measurable, quantifiable, and defensible against the next generation of AI-native alternatives.
The early USVP bets illustrate the thesis clearly. Clarify - an autonomous CRM that raised $22.5M in a Series A Garratt led - is not a better version of Salesforce. It is a CRM built from the assumption that AI agents do the work. Weave Bio - which raised a $20M Series A with USVP's participation - is building the regulatory workflow platform that pharma and biotech companies have needed since the FDA went digital. These are not incremental SaaS plays. They are companies that assume the AI is doing the work and build the infrastructure around that assumption.
The AI-Native Generation
Why AI-Native Wins
Garratt's core argument about the current technology cycle is not subtle. Legacy SaaS companies built software that organized and surfaced information. AI-native companies build software that acts on information. The difference is not incremental - it's the difference between a tool and an agent. And when software becomes an agent, you can measure what it produces. When you can measure outputs, you can justify prices that legacy SaaS could never command.
This is why his USVP portfolio skews toward regulatory automation, autonomous CRM, and agentic advertising technology. These are markets where the existing software is expensive, slow, and still requires enormous amounts of human labor to operate. An AI-native alternative that cuts that labor by 80% has a pricing conversation that looks nothing like a traditional SaaS sales cycle. It looks more like hiring.
His Weave Bio thesis statement is the clearest articulation of this view: "Few AI solutions deliver the value Weave provides for life sciences companies by automating the end-to-end regulatory workflow. The value created by this emerging generation of AI native companies far surpasses what legacy SaaS provided." In other words: the market is bigger than people think, and the winners will be determined by how much measurable work the software can take off the table.
The cleantech and climate thread that runs through Garratt's entire career - from Michigan Ross to Africa to Battery Ventures - has not disappeared at USVP. His investment parameters explicitly include climate tech and sustainability. The materials scientist who designed aircraft never fully separated from the engineer who wanted to measure physical systems. The framework transferred; the domain broadened.
The Details That Matter
Before his first VC job, Garratt invested in clean energy entrepreneurs in Africa and wrote UN energy policy. It's the Silicon Valley origin story that almost no one has.
As Salesforce Ventures' first employee, he didn't just join a program - he had to invent one. The playbook that made Salesforce Ventures the largest corporate VC arm in tech was largely written by him.
His newsletter, "The Powerloom," is named after the machine that mechanized textile production during the Industrial Revolution. The implication: AI is a structural shift, not a feature upgrade.
His undergraduate degree in materials science - the physics of how physical matter behaves - gives him a different pattern-recognition lens than the typical Stanford CS-to-VC path. He's been measuring systems his whole career.