How Unbuilt Lab takes you from idea to MVP in 14 days

Most idea-validation guides on the internet stop at the validation. Unbuilt Lab does not — it goes all the way to a shippable MVP plan. This guide walks through every step of that flow: the free trial, the On-Demand bundle, the catalog of pre-validated ideas, the Idea Validation Report, the Blueprint Pack, and the optional "we can build it for you" engineering service.

The fastest path from a blank page to a buildable MVP plan on Unbuilt Lab takes one afternoon. The longest realistic path — full subscription, comparing dozens of ideas, generating multiple Blueprint Packs — takes two weeks. Both are designed around the same constraint: founders do not have months.

The five steps from blank page to MVP

Every Unbuilt Lab account, regardless of tier, follows the same five-step funnel. Each step exists to reduce one specific kind of risk.

1. Catalog scan — find ideas the market is already asking for, not ideas you have to convince the market about.
2. Idea Validation Report — turn a one-line idea into a 12-page document covering market size, demand signals, competition, monetization paths and feasibility.
3. Blueprint Pack — the six documents you need to actually start building (the Validation Report is one of those six).
4. Pick a build path — code yourself, hire a developer, use a no-code stack, or hand it to our engineering team.
5. Ship the MVP — typically 4-12 weeks from Blueprint Pack to live product.

The product covers the first three steps. The fourth and fifth are on you (or on us if you book the build-out service).

The 7-day Trial — $3.99, one-time, no auto-renew

If you want to scan the full catalog of validated ideas once without committing to a subscription, this is what you buy. It is intentionally not a free trial — we have learned that free trials attract people who are not serious about building. $3.99 filters that down to founders who actually look at the data.

What you get for $3.99: full catalog access for 7 days, full idea details (no teasers), 1 Idea Validation Report on the house, and 1 Blueprint Pack bundle. After 7 days the account reverts to free-tier. There is no recurring charge. You will never be billed again unless you explicitly subscribe.

Best for: career-switchers, side-hustlers, and first-time founders who want to know whether the product is worth a subscription before paying $14.99/month.

The On-Demand bundle — $7.49 for 3 reports + 1 bundle

If you already have three specific ideas in mind and want to compare them rigorously without subscribing, the On-Demand bundle is for you. $7.49 gets you three Idea Validation Reports (one per idea) and one full Blueprint Pack on whichever idea wins the comparison.

The 3-report design is deliberate. Single-report packages encourage anchoring on whatever idea you already love. Comparing three on identical scoring dimensions forces you to be honest. About 30% of our On-Demand buyers tell us their final Blueprint Pack was not on the idea they entered the funnel with.

Best for: people who walked into Unbuilt Lab with their own ideas, not catalog scanners.

The Catalog — 12+ public sources, scored on 6 dimensions

The catalog is the heart of Unbuilt Lab. We continuously scrape 12+ public data sources — Reddit, Quora, Google Trends, GitHub Issues, Stack Overflow, Product Hunt, Hacker News, Indie Hackers, Y Combinator forums, Upwork RFPs, App Store reviews and Play Store reviews. The pipeline deduplicates, clusters by topic via NLP, then runs each cluster through six independent scorers.

The six dimensions:

• Demand — how many people are asking about this problem
• Gap — how poorly existing solutions cover it
• Trend — is interest accelerating or decaying
• Competition — how crowded the solution space is
• Monetization — are people already paying for adjacent solutions
• Feasibility — can a 1-2 person team realistically build this

Free accounts see titles and one-line descriptions. Paid accounts see the full 6-dimension score breakdown and the evidence snapshot — the actual Reddit posts, Quora questions and GitHub issues that triggered the cluster.

The Idea Validation Report — your first deep document

An Idea Validation Report is roughly 12 pages of analysis on a single idea. It is generated by Claude Sonnet 4 with three Perplexity grounding queries that pull live market data, then synthesised into eight sections:

• Executive summary and recommended verdict (build / pivot / pass)
• Market sizing with TAM/SAM/SOM estimates
• Demand evidence — the specific public signals that scored the idea
• Competitive landscape with pricing and positioning
• Target audience segmentation
• Monetization model recommendations
• Risk register with mitigation tactics
• Next steps if you decide to proceed

Validation reports are unlimited on Indie, Founder and Studio. Free users get one with the trial. On-Demand buyers get three.

The Blueprint Pack — six documents, one bundle

The Blueprint Pack is what you actually need to start building. It includes six documents and arrives as a single download.

1. The Idea Validation Report (the one above)
2. Market Research — current state of the market, key trends, regulatory notes
3. MVP Feature Spec — 8-12 prioritised features with acceptance criteria
4. Tech Stack Recommendation — three options ranked by cost, speed, scalability
5. Competitor Teardown — five competitors analysed across pricing, features, gaps
6. Go-to-Market Plan — the first 90 days of distribution, with channel-specific tactics

Bundle quotas vary by tier: Indie 5/mo, Founder 10/mo, Studio 20/mo. The On-Demand bundle includes one. Each bundle costs us ~$0.89 in AI inference; that is why we cap it.

"We can build it for you" — the optional engineering service

About 10% of Blueprint Pack buyers ask the same question: can you just build this for me? The answer is yes, but as a separate engagement. We do not bundle engineering into the subscription because it changes the unit economics by two orders of magnitude.

Typical scope: a 6-10 week MVP build delivered as production-ready code (React + Flask + Postgres, or React Native + Expo for mobile-first ideas). Pricing is project-based, usually $8K-$20K depending on scope, paid 50/50 milestone-based. You own the code, the GitHub repo and the deployment.

This service is not on the public pricing page. To request a quote, email hello@unbuiltlab.com with the Blueprint Pack you want built.

What 1,000 catalog users taught us about idea selection

After the first 1,000 paying users moved through the catalog, three behavioural patterns became visible that we did not anticipate at launch.

The first pattern is that the average user marks 23 ideas as "interesting" on their first scan, then narrows to 3 within a week and 1 within two weeks. Almost nobody picks the first idea they marked. The catalog's job is to give the brain enough comparative material to recognise the right idea by contrast, not on its own merits.

The second pattern is that the strongest signal of which idea a founder ultimately picks is not the overall score — it is the cluster of ideas they bookmarked. People who consistently bookmark vertical-SaaS ideas eventually pick a vertical SaaS. People who bookmark marketplaces consistently end up building one. The catalog reveals founder preference before the founder is aware of it.

The third pattern is that buyers of the Blueprint Pack are 4-5x more likely to actually ship an MVP within 90 days than users who only generated Idea Validation Reports. The difference is the MVP feature spec — the document that converts "this is a good idea" into "this is what I build first." Without it, founders tend to spend weeks debating scope and never start.

Which path should you pick? A decision flow

Here is the short decision tree we walk new users through:

• I have no idea yet → start with the 7-day Trial ($3.99). If you find one or more ideas worth pursuing, upgrade to Indie. If not, you have lost the cost of one coffee.
• I already have three ideas to compare → On-Demand bundle ($7.49). Three reports + one Blueprint on the winner.
• I want to scan ideas continuously over months → Indie or Founder subscription. The catalog refreshes weekly.
• I have a Blueprint Pack and want it built → email us about the engineering service.

The path that almost never works: signing up for Studio and never opening the catalog. About 8% of subscribers cancel within 30 days because they did not use the product. We send a check-in email at day 5 specifically to catch this. If the email cannot find a single login in the catalog, we offer to downgrade or refund — we would rather have you not pay than have you pay for unused months.

The most efficient path for a founder who is ready to move is: trial in week one, narrow to three candidate ideas in week two, generate Idea Validation Reports in week three, buy a Blueprint Pack on the winner in week four, start building in week five. Five weeks from sign-up to a written MVP plan is realistic for almost everyone who actually uses the product the way it is designed.

Sources & further reading

• Y Combinator — essential startup advice
• First Round Review — how Airbnb validated demand before scale