The 51st Feature

There's a reflex I've watched play out in more meetings than I can count. The competitor ships something with fifty features. So the plan becomes: we need fifty-one. Maybe fifty-two, to be safe.

And there's always another competitor, always another fifty-first thing. You end up in an arms race where the prize is a product so crowded that nobody — including the people who built it — can quite remember what it was for.

What's strange is that the market keeps rewarding the opposite. Over and over, you find companies that ship less than the competition and win anyway. Fewer features. Smaller menus. Narrower scope.

The pattern is real

Google. In the late '90s the winning move in search was to become a "portal" — Yahoo, Excite, AltaVista, all stuffing their homepages with news, mail, chat, horoscopes, weather, stock tickers. Google shipped a logo and a text box. It looked almost rude. And it took over the category so completely that by the end of the 2000s it was handling something like nine out of ten searches worldwide.^[1]

The iPod. When Apple launched it in October 2001 — $399, five gigabytes — the MP3 player aisle was a thicket of devices bragging about radio tuners, voice recorders, and spec sheets. Apple's pitch was four words: 1,000 songs in your pocket. No feature list. Just the outcome. The iPod went on to sell roughly 450 million units over its lifetime.^[2]

In-N-Out. The menu is about sixteen items. A typical McDonald's carries something closer to 145. And yet a single In-N-Out location pulls in roughly $4.5 million a year against around $2.6 million for a typical McDonald's, at an estimated 20% margin — better than Shake Shack or Chipotle.^[3] Fewer things, made better, sold more.

Trader Joe's. A conventional supermarket stocks 30,000 to 50,000 products. Trader Joe's carries around 4,000, most of them private label. It generates roughly $1,750 in sales per square foot — more than double Whole Foods.^[4]

ING Direct. This one is my favorite, because it's a bank that got famous for turning customers away. Want a credit card? No. An online brokerage? No. They offered a handful of high-yield savings accounts, some CDs, a few mutual funds, and that was the menu. Every extra product drags in staffing, infrastructure, and compliance overhead — costs that quietly tax the rates you can offer everyone else. Capital One bought them for roughly $9 billion.^[5]

WhatsApp. When Facebook acquired it in 2014, the investor Jim Goetz of Sequoia described a note taped to co-founder Jan Koum's desk: No Ads! No Games! No Gimmicks! At the time WhatsApp had over 450 million monthly users and a grand total of 32 engineers. The whole thing sold for roughly $19 billion.^[6]

I could keep going — Basecamp ("build half a product, not a half-assed product"^[7]), Dropbox launching on the strength of a three-minute demo video,^[8] the Nintendo Wii outselling the far more powerful PlayStation 3 and Xbox 360 with motion controls and a $250 price tag.^[9] The pattern is everywhere once you start looking.

The seductive (and dangerous) conclusion

Here's the trap. You line up all those examples, you squint, and a tidy law of nature seems to emerge: less wins. Cut features, beat competitors, collect winnings.

But notice what just happened. We surveyed a row of winners, observed that they were simple, and quietly upgraded "simple and successful" into "successful because simple." That's a causal claim smuggled in on the back of a correlation. And it doesn't survive contact with the details.

Take Google again. Was the spare homepage really why it won? Or was the homepage just the visible tip of something far more important happening underneath — PageRank, an approach to ranking results that was dramatically better and much harder to spam than keyword-stuffing competitors? The clean page was a consequence of being a search-first company. The thing that actually won the market was a core engine that was an order of magnitude better at the one job.

The iPod is the same story. The device was simple, yes. But its durable advantage was the system around it — iTunes, then the iTunes Store in 2003, then Windows support that opened it to the vast majority of computers on earth. Getting music onto the thing was effortless in a way no competitor matched. The simplicity was real, but it rode on top of integration, design, and a marketing budget most rivals couldn't dream of.

ING Direct's "no" wasn't magic either. It was inseparable from a low-cost structure: refusing complexity is how they kept overhead down, which is how they funded better rates, which is what customers actually showed up for. The simplicity enabled the moat. It wasn't the moat.

In almost every celebrated case, when you ask "what actually caused the win," the answer underneath the simplicity is something else — a ten-times-better core, an ecosystem, a cost structure, superior timing, or just a small team executing with unusual focus.

What the research actually supports

The solid part is feature fatigue. In a 2005 paper in the Journal of Marketing Research, Thompson, Hamilton, and Rust documented a clean and slightly uncomfortable mechanism. Before people buy, they weight capability: more features look like more value. After they buy and start living with the thing, they weight usability, and the loaded-up product starts losing. They surveyed buyers and users separately and got completely different answers about which product was better — even though both groups were evaluating the exact same thing. Which means loading up a product with features might be winning customers and losing users at the same time.^[10]

The shakier part is choice overload — the idea, beloved by a thousand design talks, that more options paralyze people. Its origin is the famous jam experiment by Iyengar and Lepper: shoppers who saw 6 jams were ten times more likely to buy than those who saw 24.^[11] Irresistible. It launched Barry Schwartz's The Paradox of Choice and became a design conference staple.

The problem is that it doesn't reliably replicate. In 2010, Scheibehenne and colleagues published a meta-analysis pooling around fifty experiments, and found the average effect of more choices on satisfaction or sales was essentially zero — with wild variation between studies.^[12] A 2015 re-analysis rescued the effect, but only under specific conditions: when the choice set is genuinely hard to compare, the decision is high-stakes, and the person has clear preferences going in.^[13] Take away those conditions and the effect shrinks toward nothing.

The honest summary of the research is therefore narrower than the slogan. Cramming in features does measurably hurt long-run satisfaction. But "fewer options always wins" is not something the data supports.

So when does doing less actually cause winning?

Doing less doesn't win on its own. It wins when it's a vehicle for something else:

Notice that every item on that list is the actual cause, and "fewer features" is the thing that makes it possible or signals it to the customer. Doing less is an enabler and a signal. It is almost never the cause by itself. When teams cut features without a better core underneath, they don't get Google. They get a forgettable product that does too little.

When "more" wins — and what "more" usually means

Plenty of enormous, beloved products win in markets that reward breadth, depth, and integration. The Bloomberg Terminal is a sprawling, expensive, gloriously maximalist tool that dominates finance — and its users are experts, which happens to be one of the conditions under which choice overload doesn't bite. Salesforce wins partly through an ecosystem of thousands of add-on apps; SAP and Oracle make breathtaking amounts of money on the back of legendary complexity.

So even the maximalist winners mostly aren't winning on feature count. They're winning on depth, composability, ecosystems, and network effects — a different axis. Which lands me on the only conclusion I really trust here: be exactly as skeptical about why something loses as about why it wins. Sometimes less is more. Sometimes less is just less. And sometimes "more" is really "deeper" — the feature count was never the thing, in either direction.

So before reaching for "add" or "cut," the questions worth asking aren't about quantity at all: Expert users or newcomers? A job that genuinely needs breadth, or one occasional task? Strong network and ecosystem effects, or none? High switching costs, or low? "More or less" is the wrong question. What job, for whom, under what conditions is the right one.

A word on the graveyard

For every Instagram that won by ruthlessly cutting itself down to a single feature — its founders famously stripped a cluttered check-in app down to just photo sharing once the data showed that was all anyone used^[16] — there's an unknown number of minimalist products that died quietly of being too thin, and that nobody remembers to cite.

This is survivorship bias, and it's the occupational hazard of every "secrets of successful companies" story. We study the simple products that made it and reverse-engineer principles from them, never seeing the identically simple products that didn't. Best practices built only from winners are fragile, because they were never tested against the full field. So hold the examples in this essay — mine included — a little loosely. They illustrate a mechanism. They don't prove a guarantee. (I explored the same blind spot from a different angle in The Luck We Edit Out.)

What I actually do with this

I've stopped thinking of a product as a feature warehouse and started thinking of it as a museum. A warehouse is great because it holds everything. A museum is great because of everything that isn't on the walls — because somewhere there's a curator whose entire job is saying no. When you walk through a museum that moves you, you're experiencing the editing as much as the art.

So on the teams I'm on, I try to spend as much energy on what to remove as on what to add. Not because less is automatically more. But because subtraction is the cheapest way to buy focus, and focus is what's usually doing the real work — including the work of actually getting to done.

If you want a practical lens, it's three questions, in order:

The companies that win by doing less aren't winning because their feature list is short. They're winning because the short list let them be great at the thing that mattered. The empty space on the wall isn't the point. It's what makes you look at the painting.