The most expensive thing Emergence Capital gives a founder may not be the money. It may be a partner's calendar. The San Francisco venture firm makes only five to seven new investments a year. Each investing partner is expected to make roughly one. In an industry where a sprawling portfolio can turn board members into quarterly visitors, Emergence has built its pitch around the opposite idea: fewer companies, more involvement, longer memory.
That sounds almost quaint beside a $1 billion fund. Yet the tension is the point. Emergence VII, announced in March 2025, gives the firm considerably more capital without changing the constraint it likes to advertise. The fund is aimed at B2B companies reshaping work with AI, but the product being sold to founders is still human attention - product judgment, customer introductions, executive recruiting, fundraising help and a board partner who has room for the bad week as well as the victory lap.
Emergence was founded in 2003 by Jason Green, Brian Jacobs and Gordon Ritter around a then-unfashionable thesis: enterprise software would leave company servers and move to the cloud. Its early Salesforce investment gave the thesis a horizontal platform. Veeva showed that cloud software could be rebuilt around the grammar of a single industry. Zoom made a work tool feel like public infrastructure. The portfolio became a compact history of how business software escaped the IT closet and entered everyday work.
A boutique service wearing a fund's clothes
Emergence is a company, though not a software company. It raises pools of capital from limited partners, buys equity in private startups and seeks returns when those companies are acquired, sell shares or enter public markets. Like other venture firms, its economics are generally built on management fees and a share of investment profits, although its specific fee and carry terms are not public. The limited partners provide the capital. Founders consume the service.
That service starts before a term sheet. The firm researches markets in advance, a habit it calls developing a "prepared mind." Rather than wait for a pitch and then learn the category, investors form a view on a platform shift, meet the people building inside it and compare each company against a thesis. This is how specialization compounds: every customer call improves the market map; every portfolio executive becomes a reference; every board lesson sharpens the next diligence process.
Prepare
Research a narrow change in how businesses work before the category becomes obvious.
Select
Make five to seven new investments across the firm, preserving conviction and ownership.
Stay
Use board work, recruiting, customer access and operating advice from first check onward.
The method solves a practical founder problem. Capital is plentiful in good markets and painfully selective in bad ones, but neither condition guarantees useful help. Early enterprise companies must find a repeatable sales motion, recruit leaders before the brand can do the recruiting and resist building features for one loud customer. Emergence's claim is that a focused B2B investor has seen enough versions of those problems to shorten the search for an answer.
“Some go wide. We go deep.”Emergence Capital's positioning, reduced to six words
The customers are founders - and their customers
The immediate audience is an early-stage founder building business software: horizontal platforms, tools for a particular industry, infrastructure and, increasingly, AI systems that deliver work rather than simply license software. The portfolio ranges from Salesforce, Box and Zoom to Gusto, Doximity and DroneDeploy; newer companies include Together AI, Bland, Unify, Genspark and Physical Intelligence. The common thread is not a department or a coding language. It is a bet that the product changes how an organization operates.
Sometimes the network becomes tangible. Emergence says it introduced Together AI to Zoom, which became the AI infrastructure company's first enterprise customer. The firm also helped recruit an engineering leader. At Unify, it says it supplied customer introductions, early hiring help and strategic advice. These examples are marketing, of course, but they reveal what founders should demand when diligencing any investor: names, actions and outcomes, not a slide filled with platform-team nouns.
Those four figures are Emergence's own measurement of its record, not an audited comparison with every rival. They are still useful because they show what the firm chooses to optimize and publicize: follow-on financing, billion-dollar outcomes, IPOs and cash returned. Venture results come from outliers. A concentrated portfolio raises the consequence of every miss, while a Salesforce, Veeva or Zoom can rewrite an entire fund.
The cloud playbook meets a stranger machine
Fund VII arrives at a platform shift that resembles cloud only from far away. Cloud replaced installed software with a continuously delivered service. Generative AI can replace pieces of the service itself. A conventional SaaS vendor charges for access to a tool; an AI-native service may charge for completed research, resolved support cases or closed books. Software revenue starts to mingle with labor economics. Gross margins, quality control and trust all need new playbooks.
Emergence has organized its research around that uncertainty. It studies vertical AI, open-weight models, inference infrastructure, AI-native services and what happens when tiny teams gain extraordinary output. Its recent writing argues that long-running background tasks will consume far more tokens than chat interfaces, creating opportunities in memory, throughput-first compute and context management. This is not a shopping list. It is a map of where the firm expects bottlenecks to become companies.
The size of the last bar creates the obvious strategic risk. A low-volume firm with more capital must either write larger checks, reserve more money for winners, expand into later rounds or quietly increase the deal count. Each choice can dilute the boutique promise. Emergence's 2021 solution was explicit: a $575 million early-stage fund beside a separate $375 million opportunity fund. Fund VII is described as a $1 billion commitment to B2B founders, but the enduring question is allocation, not announcement size.
Where Emergence sits on the venture shelf
A founder can choose a global multi-stage brand, a seed specialist, a former operator with a small fund, a corporate investor or a sector expert. Emergence occupies a narrower shelf: an established enterprise specialist with enough capital to matter across stages and a deliberately small annual intake. Its closest alternatives include enterprise-focused firms such as Scale, Wing and Bonfire, alongside broader competitors including Accel, Bessemer, Sequoia and Andreessen Horowitz.
The differentiation is easy to say and hard to verify before living through it. Emergence points to zero partner turnover, internal promotion and long board relationships. Continuity matters because a venture partnership can change faster than a startup matures. The person who wins a deal may not be the person still attending meetings eight years later. A stable team is therefore not office culture trivia; it is part of the product specification.
The firm's true inventory is conviction. Its true capacity limit is the number of founders a partner can know when the numbers turn ugly.
There is also a limit to expertise. The cloud era rewarded repeatable B2B playbooks: annual contracts, predictable renewals, inside sales, customer success and high software margins. AI companies may sell usage, outcomes or blended software-and-service work. Their costs move with inference. Their products can improve weekly and break unexpectedly. Buyers want automation and demand accountability. Pattern recognition helps until the pattern changes.
That is why Emergence is interesting now. The question is not whether Salesforce and Zoom were good investments; the market answered that. The question is whether a firm built around one technological transition can remain specific without becoming nostalgic. Its research suggests an answer: keep the operating discipline, replace the assumptions. Go deep on how work is changing, not on the word "SaaS." Treat AI as a new production system rather than a feature attached to an old seat license.
What a founder can steal
The most portable lesson has nothing to do with raising a fund. Choose a constraint that clarifies every decision. Emergence's five-to-seven rule narrows sourcing, diligence, time allocation and brand. A startup can do the same with a customer segment, a workflow or a promise it refuses to compromise. Specialization is not merely a smaller market; done well, it is a faster learning loop.
The second lesson is to turn assistance into evidence. “Network” is vague. An introduction that becomes the first enterprise customer is concrete. “Talent support” is wallpaper. A named engineering leader hired through the firm is a result. Founders evaluating investors should ask for references from the difficult middle: the missed plan, the executive replacement, the down round and the strategic disagreement. That is where high touch either becomes a practice or evaporates into branding.
Emergence has spent more than two decades arguing that being narrow can make a venture firm larger. Fund VII will test the clever inversion. A billion dollars can buy ownership, reserves and time. It cannot manufacture attention. If Emergence keeps the calendar scarce while the capital grows, its oldest constraint may remain its most valuable asset.