Built a bilingual business valuation MVP from market research to live deployment in one session

Project Description:

- Achievements: Took a specific market insight — a real gap in software serving sub-$5M-revenue, Latino-owned businesses — all the way to a functional MVP: a bilingual, multi-page web application (homepage, assessment flow, scoring engine, valuation display, transactional email) built through direct conversation with Claude Code — positioning strategy, bilingual copywriting, source-verified claims, and the actual code changes across a real Next.js codebase, build-tested and deployed to Vercel in the same session. It's currently in beta, covering the construction and trades sector, with additional industries planned for future versions. Working name Cimiento cleared a federal USPTO trademark check.

- Tools: Claude (chat — market positioning, bilingual copywriting, research verification), Claude Code (implementation — editing the actual Next.js codebase, running builds, committing, deploying), Vercel (hosting/deployment), VS Code.

- Methodology: The idea started with a search for an underserved niche, informed by business health and valuation work I do occasionally on the side for friends and associates, drawing on my MBA. Working through that with Claude, the gap that stood out was businesses under $5M in annual revenue — particularly Latino-owned businesses — a segment with real need and no dedicated software serving it, which is why the product is bilingual (English/Spanish, one-tap toggle) rather than English-only with translation bolted on later. Early positioning framed the product around "exit readiness," but that self-selects a narrow slice of owners actively planning to sell — maybe 10–15% of the market at any given time. Working through the reframe with Claude, we tied the broader positioning to the brand name itself: Cimiento is Spanish for "foundation," so the product now leads with business foundation and health — a permanent concern for every owner — with exit readiness as one of several things a strong foundation unlocks, rather than the entry requirement for caring about your score. That's not just positioning language — the free tier itself delivers on it: every user gets a breakdown of business health across eight cross-industry categories plus the top three areas most urgent to fix, which is genuinely useful on its own to the very common case of an owner who's a technical expert in their trade but was never trained to run the business side of it. The same free assessment also returns an approximate valuation range, based on national multipliers for the construction and trades sector — the only sector those multipliers currently cover, with additional industries planned for future versions — which is where the exit conversation becomes available rather than required. Before adding a value-anchoring claim (that a professional valuation typically costs thousands of dollars), Claude verified the actual market pricing through live research instead of accepting the number as a given, and specifically flagged that a broader claim I'd considered — implying the whole assessment replaces "thousands of dollars of consulting" — wasn't defensible, so we anchored on the valuation piece specifically, which is well-sourced. From there the same conversation moved directly into implementation: editing the actual homepage, assessment intro screen, results page, and email template in both languages, running the build to catch errors, and committing and deploying to Vercel — from idea to a live update on the actual deployed site, in one continuous session. That same discipline about not losing the customer runs through the product's core design too: financial terms like SDE and owner-dependency are presented in plain, conversational language first, with the technical term offered only as a secondary note, and every financial input clarifies monthly vs. annual, since owners think in monthly terms but the underlying valuation math needs annual figures.

- Learnings: Two different kinds of lessons came out of this, and both matter. On the positioning side: the instinct is to reach for the most impressive-sounding claim, and the discipline is checking whether it's actually true before it goes on the page — a claim you have to walk back later costs more credibility than a narrower one you never have to defend. Tying positioning language to the brand name itself (foundation, in two senses) also does more work than a generic tagline, because it makes the metaphor and the product architecture the same idea. On the technical side: a clean build — compiles, typechecks, generates every route with no errors — tells you the code is structurally sound, not that every integration actually works end to end in the real world. That gap is exactly where a known bug lives right now: email delivery isn't yet working correctly in production, something a passing build doesn't catch and only real usage surfaces. Fixing it is next.

- Relevant URLs: https://cimiento-three.vercel.app/en — functional MVP, bilingual (toggle to /es for Spanish); currently in beta