
You've built something that works, your first users are engaged, and now you need funding to grow. The minimum viable product (MVP) stage is where most seed founders find themselves, and what investors expect before writing a check looks different than it did two years ago.
Seed investors now look for real traction, not just a working product, with many expecting meaningful annual recurring revenue (ARR) before a seed round closes. This guide covers how to build and validate an MVP efficiently, the mistakes that waste months of runway and what investors need to see. For a deeper look at fundraising mechanics, see our guide to raising seed funding.
The MVP stage is when you build the simplest version of your product that delivers value to real users and generates real learning. An MVP has enough core features to deploy to early adopters, collect feedback and begin validating your riskiest business assumptions.
Your initial idea is almost certainly wrong in some meaningful way, and the MVP stage is how you figure out where. Many successful startups pivot from their first idea. Building an elaborate product before testing core assumptions is an expensive gamble, execution and learning velocity during the MVP stage matter far more than the elegance of your original concept.
The MVP sits between ideation and product-market fit. Seed startups are often still building their MVP and identifying their ideal customer, while Series A companies have a working product moving toward product-market fit. The MVP stage ends when customer demand becomes strong enough that users tell others about your product and you can barely keep up with growth. See our guide for a breakdown of how early stage VC firms evaluate companies at each phase.
Eric Ries popularized the Lean Startup methodology: figure out the problem, develop an MVP to learn as quickly as possible and measure how customers respond. Rather than spending years perfecting a product in isolation, lean methodology pushes founders to get something into users' hands and let behavior guide what to build next.
The Build-Measure-Learn loop is a continuous cycle: build a small increment, measure how users respond and learn whether to keep going or change direction. Each pass reduces uncertainty about your market, product and business model. The faster you cycle, the more assumptions you test before your runway runs out.
Launching a mediocre product and iterating on real feedback beats waiting to build something perfect. A useful heuristic: look for a way to accomplish most of what you want with a small fraction of the work (the "80/20" Pareto mindset). The only way to reduce startup uncertainty is through contact with real users.
Not every MVP requires writing code. Each type below represents a different way to enter the Build-Measure-Learn loop, and the right approach depends on what risk you're reducing: whether people want your product, whether they'll accept the experience or whether you can deliver the service at all.
A concierge MVP means delivering the service manually to customers who know they're getting a human-powered experience. You can watch how people use your product and adjust quickly without engineering. When early customers start pushing to pay or asking for more, that's usually the signal to automate.
A Wizard of Oz MVP looks automated to the user, but humans perform the work behind the scenes. Zappos founder Nick Swinmurn used this approach by creating a website displaying shoe photos, then personally purchasing and shipping shoes from local stores when orders came in.
A landing page MVP is a simple webpage explaining your concept with an email signup or pricing to gauge demand before writing any code. Dropbox founder Drew Houston created a three-minute demo video showing how the product would work, and the waitlist jumped from 5,000 to 75,000 signups overnight. This is one of the fastest ways to test whether a market exists.
A piecemeal MVP combines existing third-party tools to deliver functionality without custom software. This works best when value comes from coordination or curation rather than proprietary technology. Many founders underestimate how far tools like Zapier, Airtable (CRV led Airtable’s Series A in 2015) and Stripe can take them before custom code is necessary.
Building an MVP well means being disciplined about what you build, what you skip and how you'll know if it's working. The sequence below prevents the most common mistakes. Each step narrows what you need to build, so by the time you write code, you're building the smallest possible test of the thing that must be true for your business to work.
Before writing any code, talk to dozens of potential users about how they currently deal with the problem you're targeting. If you can't articulate the problem, the current workaround and why it fails in one sentence, you're not ready to build.
Your riskiest assumption is the single belief that must be true for your business to work, and your MVP should test it. Airbnb's biggest risk wasn't technology, it was whether strangers would sleep in each other's homes. Their MVP tested that behavioral assumption on a shoestring budget.
For every feature you're considering, ask:
Features that don't directly test the riskiest assumption can wait.
Your tech stack should prioritize iteration speed, not engineering elegance. Third-party frameworks handle authentication, payments and infrastructure so you don't have to. DoorDash’s founding team used simple, off-the-shelf tools to run early operations (including consumer apps to track drivers).
Define success metrics before you build. A commonly used product-market fit benchmark is the "very disappointed" question: if a substantial share of users would be very disappointed without your product and a strong majority activate, you're on the right track. If only a small fraction activates, it's time to rethink your approach.
Validation isn't about collecting compliments from friends. It's about observing real behavior from people who feel genuine pain around the problem you're solving.
An early adopter must have the problem you're solving, know they have it and be actively seeking tools. The best ones show up in online communities, Reddit threads, Slack groups and industry forums. MVPs tested with a small group of carefully targeted users produce stronger signals than broad launches, because narrow audiences give clearer feedback.
Specific tasks during testing yield better signals than open exploration prompts. Keep feedback forms to three to five questions maximum. Look for patterns across multiple users rather than reacting to individual suggestions.
Iteration is refinement; pivoting is a fundamental change in direction. If users return regularly, organic word of mouth builds and engagement trends upward, you iterate. Flat acquisition and customers describing your product as "nice to have" are pivot signals. Products typically need several months of genuine testing before the signal becomes clear.
First-time founders consistently make three mistakes that waste months of runway, all stemming from building in isolation instead of building in conversation with users.
The word "product" in MVP leads people to overbuild. The real goal is running the smallest possible test of your riskiest assumption, not shipping a polished product. The duct-tape prototype is what users should see first, and the cycle between build and feedback should be as short as possible.
A proof of concept validates whether a technical idea is buildable; a prototype tests user experience; an MVP tests market demand with real users. A proof of concept can succeed in a lab, but an MVP only succeeds when real users engage with it in their actual workflow. Most founders using established technology stacks should skip directly to MVP.
Even at the MVP stage, understand what it costs to acquire a customer and how much revenue that customer generates. If acquisition cost exceeds gross profit per customer, scaling only widens the gap. You don't need perfect numbers, but tracking rough acquisition cost and revenue per customer tells you whether your business model can work before you pour money into growth.
Iconic startups started with remarkably simple MVPs and studied what happened when real people encountered them. These three examples span different industries and MVP types, but all reached the same conclusion: the fastest path to conviction puts a real offering in front of real people.
In October 2007, Brian Chesky and Joe Gebbia couldn't make rent in San Francisco. When a design conference sold out local hotels, they rented three air mattresses at $80 per guest with breakfast included. All three spots filled, validating the premise that strangers would pay to stay in someone's home months before Airbnb's formal incorporation.
Drew Houston's three-minute demo video drove the beta waitlist from 5,000 to 75,000 signups overnight. It validated market demand, attracted co-founder Arash Ferdowsi and caught Paul Graham's attention all without building the complex synchronization infrastructure.
Nick Swinmurn photographed shoes at local stores, posted them on a website and personally bought and shipped pairs when orders came in. The manual process validated demand without inventory investment, and Amazon later acquired Zappos for $1.2 billion in 2009.
The MVP stage isn't about building a product alone, it's about building the evidence base that makes investors confident enough to write a check. The founders who raise most efficiently treat their MVP as a fundraising instrument from day one.
At CRV, an early stage venture capital firm, we evaluate MVP stage startups across four dimensions: team composition, market opportunity, product quality and traction. The revenue bar has risen, with many seed investors now expecting meaningful ARR. For business to business (B2B) companies, the baseline often requires demonstrated monthly recurring revenue (MRR) or a meaningful base of engaged users. Pre-revenue isn't necessarily a dealbreaker, but pre-validation is.
What matters more than where your traction is right now is where it's going. Investors want multiple months of trend data showing sustained momentum. Revenue leads the slide if you have it; numbers need clear context; and your traction section should stay at one to three slides.
MVP stage metrics should show that growth is plausible and efficient. Investors evaluate:
Together, these tell a coherent story about momentum, durability and whether growth will get healthier or more expensive as you scale.
Knowing when to stop iterating and start scaling is one of the highest-stakes decisions you'll make. Moving too early burns cash; waiting too long gives competitors time to take the market you've validated.
The "very disappointed" framework is a widely used product-market fit benchmark. Superhuman CEO Rahul Vohra applied this framework systematically over several quarters of focused iteration before scaling. When you're overwhelmed with usage to the point where you can barely keep systems running, that's product-market fit.
The transition from MVP to minimum marketable product is about hardening what exists, not adding new features. You're making the functional core reliable enough for a broader audience. The key indicator: customers can experience the core value proposition without being distracted by bugs or rough edges.
Premature scaling is a widely cited startup failure mode. Avoid aggressively scaling headcount before achieving product-market fit. The money in your bank account is the time you have to hit product-market fit, and every hire before that milestone shortens the clock.
At CRV, we've backed founders at exactly this stage. CRV led DoorDash's first financing round, and we were early investors in Mercury and Vercel. The founders who succeed launch before they're comfortable, talk to users more than investors and measure progress in learning cycles rather than feature count.
CRV led DoorDash's first financing round and backed the company through its Series A and B. The firm led Mercury's Series A and participated in its Series B and C and sits on Mercury’s board. CRV took the same approach with Vercel, leading the Series A and joining the board, while also continuing to back the company through its B, C, D and E rounds.
If you're an early stage founder looking for a venture partner who understands going from zero to product-market fit, reach out to us.
The MVP stage is where a startup builds the simplest version of its product that delivers value to real users and generates validated learning. It sits between ideation and product-market fit, and the goal is to test your riskiest assumptions with the least possible effort.
Most MVPs take several weeks to a few months. If your timeline stretches beyond that, you're likely overbuilding. Scoping ruthlessly around a single riskiest assumption keeps the timeline short.
After the MVP stage, startups transition toward product-market fit and eventually to a minimum marketable product. Once a substantial share of users would be very disappointed without your product, you begin scaling the team, the go-to-market motion and the infrastructure.
Pre-seed rounds (often high six figures to low seven figures) remain accessible for founders with strong MVPs and credible teams. Institutional seed rounds now typically require meaningful ARR or other strong traction signals. If you're earlier, a well-executed MVP paired with clear learning velocity can still open doors at the pre-seed level.