In baseball analytics, there’s a statistic called VORP: value over replacement player. Essentially measures how good a player is, relative to a player of “replacement level” talent that is freely available and in abundant supply. My version of that for the AI application layer: VOC - Value over ChatGPT (and/or Claude).
ChatGPT is essentially the “replacement player” of enterprise AI: it’s horizontal, does sufficiently well for a wide variety of use cases (and is getting better), and has essentially become the baseline when people want to use AI. When a customer is evaluating a series of vendors for a more specialized set of use cases (be it in sales, marketing, legal, procurement, finance, etc), there has to be a real reason for them to absorb the complexity of an additional piece of software versus just defaulting to a general model. I think of it as the VOC threshold: given the same prompt, how much better does your solution perform (for the specialized use case) than ChatGPT/Claude? Downstream of this is pricing power, faster sales cycles, and ultimately the usage/engagement that feeds into data moats and long-term defensibility.
Some concrete examples:
In sales: how much better is a sales agent’s research to qualify leads in comparison to what Deep Research could’ve done? How much more useful can the information surfaced be in actually closing customers?
In legal: how much better is the redlining of a legal-specific AI tool in comparison to feeding the document into ChatGPT? How much better can it understand and express your preferences?
In HR/recruiting: given a set of candidates, how much better is an AI-native platform at deciding who gets interviews compared to uploading a sheet into Claude? Is the platform able to pick up on undertones like culture fit?
Ultimately, the entire emergent field of AI engineering, from context management to evals and post-training, is in service of increasing an AI application’s VOC.
👋 I’m a Researcher at Work-Bench, a Seed stage enterprise-focused VC fund based in New York City. Our sweet spot for investment at Seed correlates with building out a startup’s early go-to-market motions.
