Progressive Ventures bets on the first customer

The awkward moment in enterprise AI comes after the demonstration. The software has answered a question, drafted a document or completed a task. Everyone has nodded. Then someone asks who owns the budget. Progressive Ventures has built its investment proposition around that second conversation: finding young companies whose technology can become something a business repeatedly pays for.

That puts the Palo Alto firm in a crowded neighborhood of venture capital, but gives its pitch a practical center. AI is the investment territory. The more interesting question is how a startup crosses from an impressive capability to a useful business. Managing partner Mala Ramakrishnan’s published writing keeps returning to the people, costs and habits standing in the way.

An operator’s kind of evidence

There is an unusually literal connection between Ramakrishnan’s early work and that outlook. As a Stanford student, she drove a golf cart around campus repairing computers. Her first startup addressed IT service workflows, including the problem of routing technicians. The job came before the software. She had encountered the inconvenience herself.

Her subsequent career included product leadership at Meta, Cloudera, Oracle and Veeva. An archived Cloudera conference biography describes experience spanning product management, marketing and software development. That background helps explain the firm’s emphasis on operator diligence: examining what a company must build, sell and support, rather than stopping at what it can demonstrate.

For a technical founder, the proposed help is specific: enterprise introductions, commercial validation and guidance on selling into organizations. For a limited partner, the proposition is access to early companies, selection informed by operating experience and continuing portfolio reporting. These are two audiences looking through different ends of the same telescope.

The startup wants enough money and useful relationships to grow. The investor wants a reason to believe that growth will create value. Progressive’s job is to make the connection credible. Its business is managing venture investment capital; its founders’ portfolio companies sell the software. Keeping those roles straight matters when reading the revenue figures surrounding an AI fund.

Revenue gets a seat at the table

The firm’s current selection criteria make its preferences plain. It looks for $500,000 to $1 million in recognized annual recurring revenue during a company’s first year, a defensible advantage and founders with relevant experience. It also prefers teams capable of raising more than $3 million at seed. Early, in this vocabulary, still comes with evidence.

“Revenue speaks.”

Progressive Ventures’ published investment criteria

That offers a useful distinction for founders choosing where to pitch. A team still searching for its first problem is at a different point from one already persuading customers to pay. Progressive emphasizes accelerator ecosystems including Y Combinator, StartX and Berkeley SkyDeck. Passing through one of those networks is a starting point for selection, rather than a substitute for commercial scrutiny.

The portfolio makes the idea concrete. Crimson builds case intelligence for litigation teams: organizing documents, evidence, issues and deadlines around a live dispute. SigmanticAI addresses chip engineering, including hardware description code and verification. Neither job becomes simpler merely because someone puts a chat window in front of it. Each requires knowledge of the work underneath.

Those examples also explain what “enterprise AI” can mean beyond office assistants. The customer may be a litigation practice or a semiconductor team. The shared problem is specialized work that consumes skilled time. Progressive lists both companies in its portfolio; their products illustrate the breadth of that approach without making them evidence of investment returns.

The pipeline starts earlier

The scale of the fund changed as Ramakrishnan built it. Her VC Lab account describes an initial plan for $1 million, followed by a roughly $12 million first fund. She raised her ambition as commitments arrived. That capital belongs to the investment vehicle; it is a pool to invest, rather than a price founders pay for introductions.

The firm’s mission adds another selection question: who gets to build these companies? Progressive publicly emphasizes women and underrepresented founders. In a published VC Lab interview, Ramakrishnan described finding fewer women than she hoped in the accelerator pipelines she was using. That was her observation about sourcing, with implications for where help should begin.

Alongside the fund sits Founders Creative, a separate nonprofit community and pre-seed launchpad. Its role begins before a startup necessarily fits a seed investor’s requirements. The July 2026 investor update described 30 companies launched through its accelerator, with women representing half of its founders. The community gives Progressive’s mission a place to operate earlier in company formation.

There is a useful lesson here for anyone building a network. Selecting from an existing pool and helping enlarge that pool are different jobs. They need different activities. A fund can evaluate a company ready to raise money; a launchpad can help someone reach that point. Combining the relationships does not make the organizations interchangeable.

The meeting after the demo

Ramakrishnan’s essays offer a version of this approach that founders can use without receiving a check. Start with an industry you can reach. Identify a particular job, calculate its economic value and find an internal customer advocate. Make the buying decision easier through a manageable pilot and evidence the advocate can carry into the next meeting.

The approach has conditions. Her writing asks whether model costs leave room for a viable margin, whether customers must migrate too much of their workflow and whether an incumbent already owns the relationship. Introductions cannot answer those questions for a founder. A product that needs expensive customization for every customer also has a different economic problem from repeatable software.

Perhaps the sharpest test in her enterprise buying essay is to consider whether a customer would still pay with the AI switched off. It forces the conversation back to the work being done. Progressive Ventures’ appeal rests there: capital and connections aimed at turning technical possibility into a purchase someone can justify.