The 10x Founder Myth: The 1968 Study Behind It Wasn’t What You Think

The study behind “10x” wasn’t measuring individual genius. It was comparing two computer setups, and the programmers were almost an afterthought.

Job listings still do it without irony: seeking a 10x engineer, seeking a 10x founder, must be willing to work like it’s 2011 again. The number gets treated as a settled unit of measurement, the way you’d cite a horsepower rating, backed by decades of rigorous research into what makes certain people worth ten ordinary hires. The actual research behind the number is a lot smaller, a lot older, and was measuring something else entirely when it stumbled onto the figure everyone now quotes as gospel in a job posting or a pitch meeting, decades after anyone bothered to check the source.

Here is the myth: that certain founders and engineers are a stable, identifiable order of magnitude more productive than everyone else, and that finding and backing these outliers is a coherent strategy, independent of the team and context around them.

Here is the bust: the number traces to a tiny, 1968 study that wasn’t designed to measure this at all, and the more recent, much larger research on what actually explains sustained star performance points somewhere the myth doesn’t want to look — at the team and organization surrounding the person, not the person alone, however satisfying the single-genius version is to repeat.

Where the Number Actually Comes From

In January 1968, researchers Hal Sackman, W.J. Erikson, and E.E. Grant published a study in Communications of the ACM comparing two different ways of programming a computer: working interactively at an online terminal versus submitting a program offline and waiting for it to run in a batch, at a time when computing time itself was scarce and expensive. The study’s actual subject was the computing environment, not individual talent. Almost as an aside, the researchers noticed something else in their data: the professional programmers in the study, roughly a dozen of them, varied enormously from each other — the fastest completed one task 28 times faster than the slowest, and other measurements, like program execution speed, landed closer to a 10-to-1 ratio.

That aside is the entire foundation of the modern “10x” claim. A study built to compare two work environments, run on around a dozen programmers, with no controls for how the specific problems assigned to each person varied in difficulty, and mixing results from programmers working in different levels of programming language, became the citation for an entire theory of talent that still gets repeated in venture pitch decks and job postings more than fifty years later, usually with none of these caveats attached.

How a Shaky Study Became Management Gospel

The number didn’t stay put at 10, either. Frederick Brooks cited the finding in his hugely influential 1975 book The Mythical Man-Month, using it to argue for a “surgical team” structure — one elite programmer supported by a crew of specialists, the way a lead surgeon is supported by an operating room staff. Decades later, software engineering writer Robert Glass pushed the figure even further in Facts and Fallacies of Software Engineering, citing a gap as wide as 28-to-1 between the best and worst programmers. Notice the direction of travel: each retelling made the claim more dramatic, not more precise. That’s usually a sign a number has become a legend rather than a measurement.

Even If the Variance Is Real, It Might Not Travel With the Person

Set the shaky origin study aside for a moment and grant the more defensible version of the claim: that some real variation in individual output exists. Even then, a much larger and more rigorous body of research raises a different problem — whether that outlier performance is actually a property of the person, portable from job to job, or a property of the environment they happened to be standing in.

Harvard Business School professor Boris Groysberg addressed exactly this question in his 2010 book Chasing Stars, tracking more than a thousand star equity research analysts at Wall Street investment banks over many years, backed by over two hundred candid interviews. His finding: star analysts who switched firms suffered an immediate and lasting decline in performance, one that in most cases didn’t recover for years, if ever. Their earlier brilliance depended heavily on their former firm’s resources, proprietary systems, internal networks, and colleagues — all of which stayed behind when the individual walked out the door. The main exceptions were stars who moved together with their existing team, or who moved to a firm with meaningfully better resources than the one they left. Absent that, most “stars” who switched firms became what Groysberg called meteors: bright, then quickly burned out in their new setting.

Why Startups Love This Myth Anyway

The 10x framing survives because it’s useful to exactly the people repeating it. It gives founders a tidy justification for extreme hours and unequal treatment — if someone is genuinely worth ten ordinary hires, ordinary expectations about workload and civility start to feel negotiable, and pushing back on that treatment starts to look like the smaller person’s failure of ambition. It gives investors a simple, appealing shortcut: find the outlier individual, back them, skip the much harder and less romantic work of assessing team structure, market timing, and organizational design. A single mythical genius is a much better pitch-deck slide than “we assembled a well-resourced team with strong internal systems,” even though the second sentence is closer to what the research on sustained performance actually supports, and closer to what a careful investor would actually want to verify before writing a check.

Why This Myth Stuck

None of this is really about programmers specifically. A solo genius is a better story than an org chart, the same way a moment of founding clarity beats years of unglamorous skill-building elsewhere in this category. The 10x number gives that story a satisfying quantity, which is exactly why nobody asked too many questions about where the quantity came from until it had already shaped fifty years of hiring folklore, several bestselling management books, and an untold number of job postings nobody could actually fulfill.

What To Actually Do With This

None of this means talent and skill don’t vary between people, or that some engineers and founders aren’t more effective than others. It means the specific number, and the strategy of chasing it as a portable individual trait, rests on much thinner evidence than the confidence with which it gets repeated.

• Don’t build a hiring or investment thesis around a single “10x” individual — the underlying research doesn’t support treating it as a stable, portable trait.
• When evaluating a “star” candidate, ask what part of their track record came from their team and resources, not just them.
• If you’re hiring away a “star,” consider what team or context you’d need to bring along too, not just the individual.
• Invest in the systems and team structure that let good performance compound, rather than searching for a mythical solo multiplier.
• Be skeptical of round, escalating numbers (10x, then 28x) repeated without a new measurement behind them.
• Treat “10x” language as a flag for justifying long hours or unequal treatment, not a scientific description of anyone’s actual output.

Frequently Asked Questions

Related Myths We’ve Busted

Gear Mentioned in This Article

As an Amazon Associate, Sarcastic Motivators earns from qualifying purchases.