Small business owners looking for product lessons usually study the same handful of tech giants. There is a better case study hiding in plain sight, and it is not glamorous: the slot machine. The core loop, stake something, trigger a random outcome, see the result, has not changed since the 1890s. Everything around that loop has been rebuilt at least four times. Few products anywhere have survived that many platform shifts with the fundamentals intact, and how it happened is worth ten minutes of any product owner’s attention.
One core loop, a century of rebuilds
Charles Fey’s Liberty Bell in the 1890s was the first three-reel machine that paid out on its own. The first video slot did not arrive until 1976, a gap of eight decades where the mechanics barely moved. Then the pace changed: online versions appeared in the mid-90s, and the whole category re-platformed to the browser and then the phone. New Zealand shows the compressed local version, with the machines arriving in pubs in 1987 and the country’s first casino opening its doors in 1994. The strategic lesson is the discipline of knowing your invariant. The industry never touched the loop. It rebuilt literally everything else.
Features are revenue experiments with a rollback plan
Each era added one mechanic on top of the familiar base rather than replacing it. Bonus rounds arrived with video slots. The first online progressive jackpot launched in 1998 and turned a single game into a networked prize pool, and by the late 2010s individual jackpot wins were exceeding NZ$20 million. Megaways went further and randomised the grid itself, so a single game can offer anywhere from 243 to 117,649 ways to win on a given spin. Notice the pattern: every addition was legible to an existing customer within one session. That is the iteration discipline most small software businesses get wrong, shipping features that force relearning instead of features that extend what the customer already understands, a failure Forbes innovation coverage returns to in case study after case study.
From manufacturer to marketplace
The other structural shift is who makes the product. A modern operator does not build games, it curates them. The slots catalogue on the Christchurch Casino official site runs past 2,300 titles drawn from more than 80 studios, with new releases added weekly. That is marketplace economics, not manufacturing: the operator’s real assets are the storefront, the account relationship, and the curation, while a long tail of specialist suppliers competes to fill the shelves. Plenty of SMBs face the same fork eventually, keep building everything yourself or become the trusted front end for other people’s supply, and this industry picked the marketplace side decades before the word was fashionable.
Publishing your odds, on purpose
Here is the counterintuitive bit. This industry voluntarily publishes the exact profitability of every product. Each game lists its return-to-player figure, the theoretical share of stakes paid back over time, and some titles advertise numbers as high as 98 percent. Volatility, whether returns come as frequent small wins or rare large ones, is disclosed too. A business built on chance decided that publishing its margins builds more trust than hiding them, and independent labs verify the figures so the disclosure means something. Most small businesses hide their pricing logic reflexively. It is worth asking what the equivalent of a published RTP would look like for your own product.
Mobile did not shrink the product, it re-specified it
The latest rebuild was the phone, and the industry treated it as a new specification rather than a smaller screen. Games are engineered for portrait mode, touch targets, and short sessions, with features like configurable autoplay built for how people actually use a handset. The account carries the state, so a session started on a laptop continues on a phone without friction. That is the version of mobile-first worth copying: not squeezing the desktop experience down, but re-deciding what the product is when it lives in a pocket. A category that began with a cast-iron box and a lever made that call cleanly. Most 10-person software companies still haven’t.
Last Updated on August 21, 2026 by Frederique Bros
