Intuit, built today

The original wedge

Intuit did not start with a market. It started with one observation: Scott Cook watched his wife struggle to balance a checkbook and bet that ordinary people would pay for software that felt easy rather than powerful. Everything that followed, the plain screens that looked like a paper check, the refusal to add features that confused people, came from that single insight. The wedge is a specific person doing a specific annoying task, not a category you want to enter.

What changed since 1983

Three things moved. Distribution went from retail boxes and mail-order ads to search, content, and app stores, so you can reach the same frustrated bill-payer for a fraction of the cost. Capital went from Cook carrying more than three hundred thousand dollars in personal debt to a working prototype you can build in a weekend for the price of a few subscriptions. And the moat inverted: in 1983 the edge was shelf space and word of mouth, today it is data, bank integrations, and the switching cost of a system that already knows your money.

How you would build it now

Start with the stack that did not exist in 1983. The single hardest thing Intuit ever built, a reliable connection to the banks, is now a line of code. Aggregation APIs like Plaid, MX, or Finicity hand you account balances and transactions for cents per user, so the plumbing that cost Intuit years is a first-day integration. Wrap it in a modern web app, Next.js or SvelteKit deployed on Vercel, add Stripe for billing and Clerk or Auth0 for login, and use an AI coding assistant to go from idea to a working product in days instead of months. You can ship a real, money-touching tool with a team of one or two.

Most of what nearly bankrupted Cook is now bought, not built. Payments, bank data, tax calculation, identity and KYC checks, even the compliance scaffolding, are all APIs you rent. The founder's job is no longer to construct the machine. It is to design the one experience that makes a specific person's money feel handled, and to own the data that keeps them from leaving.

Go-to-market flips the same way. Intuit needed retail shelves and a hundred and twenty-five thousand dollars of print ads to reach people who could not otherwise find the product. You reach them now through search: write the definitive guide to the exact problem you solve, rank for it, and let the people already looking find you. Add one narrow community play, the subreddit, Discord, or forum where your customer already complains about this task, and a product-led funnel where a free tier does one thing so well people pass it around. Charge from day one, because a tool that touches money has permission to. Your first hundred customers should be people you can name, not a market you are guessing at.

Harder now, easier now

Harder: the incumbents are the very companies Intuit became, and acquisition costs in consumer finance are brutal. You cannot out-spend them, so you have to out-specific them. Easier: the plumbing that took Intuit a decade is now an API call, and a credible product can reach its first thousand users without a sales team or a shelf.

The verdict

Still a good business to start, but not as another Quicken. The opening is a wedge the big players are too broad to serve well, owned end to end for one type of customer, then expanded outward the way Intuit went from Quicken to QuickBooks to TurboTax. Start absurdly narrow. Earn the right to be broad.