MBP: Revenue Is Evidence
When building a product takes days instead of months, the bottleneck in venture creation migrates from production to proof. The payment becomes the validation.
A product someone will use tells you the product works. A product someone will pay for tells you something fundamentally different: the value exceeds the price. Usage advances feasibility indicators. Payment advances demand and economics indicators simultaneously. Willingness to pay, customer acquisition cost, retention, margin structure. Each of these moves from inference to evidence in a single transaction.
This is why we frame our GTM activities with the term Minimum Buyable Product rather than Minimum Viable Product. MVP implies testing whether the concept functions. MBP implies a product someone will exchange money for. The distinction matters because the payment resolves uncertainties that usage alone cannot touch.
A protoventure generating $2K per month in revenue has not "succeeded." It has produced high-fidelity data that advances multiple indicators from Hypothesized to Tested in the epistemic framework. That data sharpens the probability estimates in the option value calculation, enabling a well-informed decision at the spin-out gate. The revenue is not just the objective. It is the evidence.
The Path to Proof Is Compressing
Building an MBP used to be the expensive part. Six months of development, a small engineering team, $200K or more in salary and infrastructure before a single customer saw the product. The "build phase" was a significant capital commitment that preceded any evidence of demand. The studio was spending to build before it could spend to learn.
Agentic product development has compressed this dramatically. For most software concepts, the path from validated concept to working product is now measured in weeks and thousands of dollars, not months and hundreds of thousands. The build phase is no longer the bottleneck. The binding constraint has shifted from "can we produce a product?" to "will someone pay for it?"
This changes the information economics of venture creation. The most expensive ignorance to maintain in any venture evaluation is demand uncertainty: does the problem exist, will someone pay to solve it, at what price, through what channel? When the build phase took six months, that ignorance persisted for six months because you could not test payment until the product existed. When the build phase takes two weeks, you can get to the determinative signal (someone actually paid) before you have spent meaningful capital. The resolution value of demand investigation increases because the cost of reaching the point where you can investigate has collapsed.
What This Does to Funding Stages
The traditional funding taxonomy is defined by production milestones. Pre-seed is an idea and a team. Seed is a product and early traction. Series A is product-market fit. These stages were calibrated to a world where each production milestone took months of work and significant capital. The funding stage was a proxy for how far along the production process the venture had progressed, and how much capital had been consumed getting there.
When a venture studio can produce a working product with paying customers at a total cost of $5,000 to $20,000, those proxies break. The entity is "pre-seed" by capital committed but presents evidence that traditional frameworks associate with seed or Series A: real revenue, real unit economics, real retention data. The epistemic quality of the evidence has raced ahead of the capital commitment.
This creates an interesting misalignment. An LP or VC evaluating a studio entity using traditional stage definitions sees a "pre-seed" investment (small dollars, early stage) but encounters "seed-quality" evidence (actual customers paying actual money). The valuation conversation becomes awkward because the reference class is wrong. The entity doesn't look like other pre-seed investments because it has resolved uncertainties that most pre-seed investments haven't addressed yet.
We don't claim to know how this resolves. The funding taxonomy may adapt, or it may persist as a convenient fiction that everyone adjusts for informally. What we observe is that the compression of build timelines has decoupled the relationship between capital consumed and evidence produced. A studio that can reach real revenue evidence at pre-seed capital levels is operating in a space that the existing categories were not designed for.
Revenue as Multi-Indicator Evidence
The power of revenue as evidence is that it resolves multiple uncertainties in a single signal.
A customer who pays has demonstrated willingness to pay at a specific price point (D1 Demand). The cost of acquiring that customer is now observable, not hypothesized (D3 Go-to-Market). If the customer returns, retention is evidenced (D1 Demand). The margin between revenue and cost of delivery is measurable (D4 Economics). Each of these moves from inference to evidence as customer transactions accumulate.
No other single activity in our venture creation framework resolves this many indicators simultaneously. A landing page test advances demand interest but not willingness to pay at a price point. A competitive analysis advances market structure but not customer behavior. A technical prototype advances feasibility but nothing about demand or economics. The MBP, by producing actual revenue, generates evidence across multiple dimensions in a single operation.
Prioritization of the MBP surfaces in our framework because it's how our Value of Information thesis plays out in real life. We're walking the walk. For software concepts where the build is straightforward, the per-dollar information yield of revenue evidence is difficult to match through any other investigation pathway.
The Founding Function as a Continuum
When the studio produces an MBP with paying customers, validated unit economics, and a documented evidence base across 33 indicators, the operator who arrives at spin-out is not arriving at the idea stage. The founding function has been substantially performed. The operator is not co-founding. They are scaling a venture where the core uncertainties have been resolved and the evidence supports advancement.
The venture industry tends to treat founding as a binary: you are a founder, or you are not. Equity structures follow the same logic, with "founder" allocations, "co-founder" allocations, and "early hire" allocations as discrete buckets. But the founding function (identify the opportunity, validate the thesis, build the first product, find the first customer) is a continuum, not a category. The studio that hands over a concept with no revenue has performed less of the founding function than the studio that hands over a product with $5K MRR and three months of retention data. The handoff point on that continuum should inform the equity conversation.
Agentic development pushes the handoff point further along the continuum than traditional models anticipated. When the studio can build the product in weeks and validate revenue before the operator arrives, the operator inherits a venture where most of the founding risk has been resolved. The remaining work (finding product-market fit, scaling the go-to-market motion, identifying adjacent customer and market profiles) is critically important, but it is operating work, not founding work. The equity structure should reflect where on the founding continuum the handoff actually occurred, not which bucket the operator gets slotted into by convention.
The conventional equity templates were designed for a world where the operator arrived earlier in the founding process and assumed more of the founding risk. As studios perform more of the founding function before the operator arrives, the relationship between contribution and equity allocation will need to evolve. The founding function and the equity allocation are continuous variables that the industry currently treats as discrete. The compression of build timelines is widening the gap between the convention and the reality.