Work-Bench Built a Venture Fund Around the Part Founders Dread: Selling

Enterprise software has a peculiar last mile. A founder can build a clever product, hire excellent engineers and produce a demo that makes a room nod in unison. Then the product meets the customer. The security questionnaire arrives. Procurement wants another review. Legal edits the paper. The champion changes jobs. A pilot that was supposed to take three weeks develops its own weather system. Work-Bench, a New York venture firm founded by Jonathan Lehr and Jessica Lin in 2013, built its identity around this stretch of road.

The firm leads seed rounds in enterprise software, now writing initial checks of $2 million to $4 million from a $160 million fourth fund. Its categories are deliberately workmanlike: artificial intelligence and machine learning, cybersecurity, infrastructure and developer tools, and vertical or horizontal enterprise applications. The portfolio includes early bets on Cockroach Labs, Socure, Spring Health, Dialpad, FireHydrant, Arthur AI and AuthZed. Yet capital is only the entry ticket. Work-Bench's real proposition is that it can help a technical founder turn buyer curiosity into a repeatable sales motion.

A fund born in the CIO's office

Lehr worked in Morgan Stanley's Office of the CIO, helping internal technology teams evaluate and onboard young vendors. Lin worked at Cisco with engineering teams on Agile transformation, learning and culture. Neither came from the familiar venture pipeline. Their advantage was a view from the other side of the conference table: how a large company encounters, tests and sometimes rejects new software.

Before the fund, Lehr started the NY Enterprise Technology Meetup in 2012. It grew into a group of more than 10,000 people. That sequence matters. Work-Bench did not bolt a community onto a finished venture brand. The room came first, full of the people who built software and the people who might buy it. The first fund, $10 million, was a proof of concept. Fund II closed at $47 million in 2018; Fund III at $100 million in 2021; Fund IV at $160 million in May 2025.

The expanding capital base has not pushed the firm into every fashionable market. Fund IV is expected to back roughly 23 to 25 companies, keeping the portfolio concentrated. The customer is a seed-stage founder with enterprise ambitions, often someone who has lived inside the workflow being rebuilt. Work-Bench looks for domain knowledge, grit and the ability to absorb feedback without surrendering a point of view.

“Customers are oxygen.”Work-Bench's shorthand for the seed-stage constraint

The product behind the check

Work-Bench talks about go-to-market as a moat. In practice, that means small, specific interventions. Who is the ideal first customer? Does the discovery call expose a costly problem or merely collect compliments? Can the demo tell one story instead of twelve? Which proof-of-concept terms prevent a pilot from becoming free consulting? What will security, legal and procurement demand before the salesperson promises a launch date?

The firm has turned those questions into a public Enterprise Playbook Library covering discovery and demos, messaging, pricing, customer acquisition, SOC 2, fundraising, legal review and open-source monetization. Its GTM Weekly newsletter reaches more than 35,000 subscribers, according to the firm. Publishing the machinery is a smart form of distribution. A founder can use a template before meeting the investors; a buyer or operator can recognize the quality of the conversation; Work-Bench learns which problems keep recurring.

The network supplies the other half of the system. Work-Bench connects portfolio companies with corporate decision-makers and says those introductions have helped produce tens of millions of dollars in enterprise contract value. Its events mix startup founders, functional operators and Fortune 500 executives - the firm's “suits and hoodies” formulation. The Work-Bench Next NYC Network adds leaders in sales, marketing, product, customer success and engineering. #Womenterprise supports women across the sector. Cofounders.nyc, launched in 2025, tries to make the earliest piece of company formation more local and more human than swiping through profiles.

Why New York is part of the machinery

When Work-Bench began, a New York enterprise fund sounded like a geographical handicap. The traditional venture center was three time zones away. But New York was dense with banks, insurers, media groups, retailers, healthcare organizations and corporate headquarters. Those companies contained both old systems and people paid to improve them. Work-Bench recast proximity to buyers as an investing advantage.

That does not mean every portfolio company is local. The fund invests beyond the city, and recent companies span San Francisco, London and Tel Aviv. New York remains the hub because the community can serve several jobs at once: sourcing founders, testing theses, finding early hires, conducting diligence and creating customer conversations. This is the important difference from a generic promise of “access.” The network is not a list in a spreadsheet. It is a recurring calendar and a set of relationships built before a deal needs them.

AI arrives wearing work clothes

The recent portfolio shows how the thesis adapts without becoming “AI, broadly.” Work-Bench led Chamelio's $5.7 million seed round for legal operations and Flock AI's $6 million seed round for retail content infrastructure. It backed Artian AI for regulated financial workflows, Paratus for military readiness administration, Built AI for commercial real-estate underwriting and Runhouse for AI post-training infrastructure. Goodfire, an AI interpretability company the firm backed at seed, announced a $150 million Series B at a $1.25 billion valuation in February 2026.

The common unit is not the model. It is the workflow. Enterprise AI must fit permissions, audit trails, domain language, legacy systems and a buyer's tolerance for risk. A polished demonstration can win attention, but production software must make fewer mistakes, explain more of its behavior and survive the organization's existing process. Work-Bench's corporate-IT roots are particularly relevant here: the distance between “the model can” and “the company will buy” remains wide.

What a founder can use before the check

Work-Bench's public material is useful even if the firm never appears on a cap table. A founder can download an ideal-customer-profile template, pressure-test a discovery call, build a value-based pricing calculator, study open-source monetization and prepare for the paperwork that follows a successful demo. The library's strength is sequence. It asks teams to identify a painful problem and a narrow buyer before expanding channels, headcount or product scope. That order sounds obvious only after a company has burned months doing it backward.

For portfolio founders, the service becomes more personal. The team helps rehearse demos, refine outbound messages, think through the first sales hire, navigate security and legal review, and find relevant operators or customers. Later, it helps prepare for a Series A. The aim is not outsourced selling. It is to leave the founder with a process that another salesperson can eventually run. That distinction is central to Work-Bench's place in the market: accelerators offer a broad curriculum, consultants sell projects, and large funds maintain platform teams; Work-Bench concentrates a smaller enterprise toolkit around a smaller seed portfolio.

The prospective buyer gets something too. Corporate executives can meet filtered young vendors, compare notes with peers and shape research around problems their teams already face. Operators can share tactics, meet future colleagues and see companies before the job listing exists. The community works because each group arrives with a different need, but the same conversation can serve all three.

Where it sits in venture

Work-Bench competes with specialist seed investors such as Boldstart Ventures, Bonfire Ventures, Point Nine and Heavybit, along with generalist seed firms, accelerators and multistage funds willing to write early checks. Many can offer recruiting, introductions and brand. Work-Bench's defense is specificity: enterprise only, seed first, buyer-informed research, and practical help from the first discovery call through procurement and the Series A process.

Its business model is conventional venture capital. Limited partners commit money to funds; Work-Bench buys equity in private companies; returns depend on those stakes gaining value and eventually finding liquidity. The unusual part is the service layer surrounding the investment. Community and playbooks make the firm easier for founders to discover. Buyer conversations sharpen diligence. Portfolio support can improve the odds that a product becomes revenue. Successful companies make the network more useful to the next founder.

There are limits to the approach. Warm introductions cannot rescue weak retention. A playbook does not remove the long cycle or political texture of enterprise buying. Concentration makes selection more consequential. And the more venture firms promise operational help, the more founders must ask who actually does the work. Work-Bench's answer is a small team that publicly revels in unglamorous detail - and, less relevant but more memorable, karaoke, Settlers of Catan and Olive Garden breadsticks.

That mix of rigor and odd specificity may be the brand. Enterprise software is built from chores that matter: incident response, identity checks, permissions, credit review, employee health, legal intake, military readiness. Work-Bench funds the people trying to redesign those chores, then helps them sell the redesign to institutions trained to say “not yet.” The clever move was recognizing that the dreaded part was not adjacent to the investment thesis. It was the thesis.