What the latte factor actually is
The phrase comes from author David Bach: take a small habitual purchase, price it over decades with investment growth, and be astonished. The idea earned its fame honestly, because the arithmetic is real. It also earned some fair criticism, because the famous versions tend to quote 10 to 12% returns, skip inflation entirely, and carry a whiff of blaming lattes for problems that are really about rent and wages. This page keeps the true part and fixes the rest: a modest 7% default you can change, the answer in today's buying power alongside the headline number, real market history next to the projection, and no lecture at the end.
Worked example
A $5.45 latte every day is $1,989 a year. Put into the S&P 500 for the actual last 30 years, 1996 through 2025 with dividends reinvested, it would have grown to $377,859, from $59,678 of skipped lattes. That is recorded history: it includes the dot-com bust, the 2008 crash and the 2022 slide, and it built that number anyway.
Looking forward at a deliberately modest 7% assumption instead: $193,861 in 30 years, of which $134,184 is growth, and which is about $79,868 in today's buying power at 3% inflation. The real past beat our modest projection. Nobody can tell you whether the future will, which is why the page shows you both.
The known entity: what buying the S&P 500 actually looks like
The S&P 500 is not an abstraction, it is the 500 largest US companies in one basket, and buying it takes one purchase. The original way in is the fund traders nicknamed the Spider: SPY, the SPDR S&P 500 ETF, launched in 1993 as the first US-listed exchange-traded fund and still among the largest funds on Earth. Its younger siblings (VOO, IVV) and countless index mutual funds track the same 500 companies for fees near zero. Naming them here is vocabulary, not a recommendation: the point is that "invest the latte money in the whole market" is not homework, it is a thing an ordinary person does in an app in five minutes, and it has a famous name.
The part nobody argues with
Whatever return you assume, one piece of this is bedrock: a small amount, moved regularly, for a long time, becomes a large amount, and the growth eventually outweighs the deposits themselves. The chart above draws it to scale: the flat line is the money just spent, the curves are the same money working, and the widening gap between them is compounding itself. Even in our cautious 7% projection the skipped purchases are $59,678 and the final figure is more than three times that. The habit is small. The decades are not. That is the entire lesson, and it works in both directions: it is why a small investment matters and also why a small recurring fee quietly matters, which is worth remembering next time something costs $12.99 a month forever.
The part worth arguing with, argued honestly
Three corrections to the folklore version of this math. First, inflation is not optional. A dollar figure 30 years out is quoted in dollars that will buy less; our today's-dollars line deflates it at an assumed 3%, and that line is usually 50 to 60% smaller than the headline. Second, the return assumption does the heavy lifting. At 7% the daily latte becomes $193,861 in 30 years; quote 11% instead and the same habit prints over $400,000, which is how the brochure versions reach their numbers. We default to 7 and let you set your own. Third, and most important: nobody's rent is made of lattes. Small-purchase math cannot fix a housing problem or an income problem, and pretending it can is how this idea got its bad name. What it can genuinely do is show that investing does not require being rich first, which for a lot of people is the belief actually standing in the way.
Why the whole market and not the company's stock
The tempting version of this page says: skip the Coke, buy Coca-Cola stock. We deliberately do not compute that, and the reason is a bias with a name. Survivorship bias: every buy-the-stock story you have ever heard is about a company that survived and thrived, because the ones that did not do not get stories. For each Coca-Cola there is a Sears, a Kodak, a Blockbuster, and the person who faithfully bought those with their skipped purchases has the same discipline and a very different ending. The S&P 500's history includes its failures by construction, its returns are published in a primary dataset we verify against, and betting on the whole scoreboard rather than one team is the version of this idea that does not depend on hindsight. The one-company version is a lottery ticket wearing the costume of a plan.
The honest fine print, above the fold
The projection assumes the return arrives smoothly every year, which real markets never do; the backtest line shows actual history precisely so you can see the difference. Investing costs are assumed to be near zero, which broad index funds have genuinely made possible. Taxes are ignored, which flatters the result unless the money sits in a tax-advantaged account. And a habit's price rises with inflation too, so the skipped-purchase side is understated in the same direction. Every one of these choices is visible, and every input on the page is yours to change, which is the difference between arithmetic and advertising.