To succeed in a startup, you must intentionally adopt 'odd' advantages—like superior technology—that your competitors are too habituated to notice or understand.
Startup success requires abandoning "average" behavior because average performance leads to business failure; instead, you must leverage technical advantages that your competitors lack the mental models to recognize.
This is for founders and technical leaders who feel stuck in standard technology stacks or are struggling to differentiate their product from a crowded market. It solves the problem of "commodity" competition by teaching you how to turn a niche, high-power tool into a competitive moat through rapid development and strategic surprise. After reading this, you will be able to identify your own "median" habits and choose technologies that provide a genuine speed advantage over the crowd.
The math of startups is brutal: if you do everything the way the average startup does it, you should expect average performance. However, for a startup, "average" means going out of business because the survival rate is significantly lower than fifty percent. To survive, you must be doing something "odd."
This "oddness" can be found in your choice of technology. Most programmers are stuck in "habits of mind"—they use certain languages not because they are the most powerful, but because they are what everyone else uses. Because these habits change so slowly, a massive opening exists for those who can break away from the median language and adopt a more powerful tool (like Lisp).
By using a superior technology that your competitors don't understand, you gain a "secret weapon." You can develop features faster than they thought possible, creating a cycle where you can duplicate or improve on a competitor’s announcement before the news even reaches the public. In business, as in war, surprise is worth as much as force.
What it is — Doing what everyone else does leads to average results. For startups, average results are synonymous with failure. Why it matters — It shifts your goal from "being standard" to "being odd." If you aren't doing something different, you are in trouble. In the book — Graham notes that big companies grow at 10% a year; if a startup mimics this average behavior, it will fail because its survival rate is so low.
What it is — Programming languages are not just tools; they are "half technology and half religion." They dictate how a programmer thinks about programs. Why it matters — This explains why competitors won't switch to better tools: they are satisfied with whatever language they happen to use because it dictates their mental model. In the book — Graham argues that programming practice is often ten to twenty years behind the processor because habits of mind change much slower than hardware.
What it is — Languages fall along a continuum of abstractness/power, from machine language up to the most powerful tools (like Lisp). Why it matters — Using anything other than the most powerful (reasonably efficient) language for application software is a mistake of the same kind as programming in machine language. In the book — He argues that while Cobol is "high-level," it is closer to machine language than Python is, and therefore less "powerful" for general applications.
What it is — A programmer using a median language (like "Blum") cannot see the power of higher languages because they lack the features required to think in those languages. Why it matters — You cannot trust the opinions of others about what tools are "good enough." Only those who understand the most powerful tool can see the difference. In the book — He uses a hypothetical programmer who thinks in "Blum" and therefore views Lisp as merely "weird" rather than more powerful.
What it is — A technical advantage that your competitors don't understand provides a massive strategic edge. Why it matters — It allows you to move faster without being perceived as having a "secret weapon." You can hide your advantage in plain sight. In the book — He compares his secret weapon (Lisp) to an assassin dressed as an old man on crutches; the police don't suspect him because they don't see him as a threat.
What it is — Using a high-power language allows you to get features done faster than competitors, especially when those features can be released instantly. Why it matters — It creates a "technological barrier." You can duplicate a competitor's new feature within a day or two of their announcement. In the book — Viaweb used Lisp to build a WYSIWYG editor that felt like a desktop app, while competitors were stuck with slower CGI scripts.
What it is — You can gauge a competitor's technical depth by looking at their job descriptions. Why it matters — It tells you which companies are "dangerous" (real hackers) and which are safe (IT flavor). In the book — Graham looked for Lisp hackers; if he saw them, he knew the company was a serious threat. If they wanted "Oracle experience," he knew they were safe.
Use the momentum of the median language against your opponents.
Use this checklist to determine how much to worry about a new competitor:
The Old Man on Crutches — An assassin hides in plain sight by dressing as an old man; the police don't suspect him because they don't see his true nature. Lesson: A technical advantage your competitors don't understand is a powerful weapon.
Viaweb vs. CGI Scripts — While competitors were building slow CGI scripts, Viaweb used Lisp to build a WYSIWYG editor that felt like a desktop application. Lesson: Choosing the right tool creates a technological barrier.
The 48-Hour Duplication — Because of their Lisp development cycle, Viaweb could sometimes duplicate a new feature within a day or two of a competitor announcing it in a press release. Lesson: Speed is a defensive moat.
"If you do everything the way the average startup does it, you should expect average performance." Why it matters: It establishes that "oddness" is a requirement for survival, not a luxury.
"Programming languages are not just technologies, but habits of mind as well, and nothing changes slower." Why it matters: This explains why the status quo persists and why there is an opening for those who break away.
"In business, as in war, surprise is worth as much as force." Why it matters: It highlights the value of a technical advantage that remains misunderstood by the competition.
This week
Ongoing
Reflection questions
Flashcards Q: Why is doing what the average startup does a problem? A: Because average performance leads to business failure; startups need to be "odd."
Q: Why do programmers stay with median languages even when better ones exist? A: Because programming languages are habits of mind, and nothing changes slower than those habits.
Q: What is the "Blub Paradox"? A: A programmer stuck in a median language cannot see the power of higher-level languages because they lack the mental model for them.
Q: Why is Lisp considered a "secret weapon" in this context? A: Because it allows for rapid development and features that are hard to do in other languages, while competitors remain stuck in their habits.
Q: How can you identify if a competitor is technically dangerous? A: By looking at their job listings; "IT flavor" (like Oracle experience) is safe, but "real hackers" (Lisp/Perl/Python) are dangerous.
Q: What is the value of surprise in business? A: Surprise is worth as much as force; a technical advantage that competitors don't understand provides a massive edge.
Confidence: High. This summary is strictly based on the provided text from Paul Graham’s "Beating the Averages." I am certain about his arguments regarding the "Blub Paradox," the necessity of being "odd" to survive, and the specific use of Lisp as a case study for rapid development. The reader should verify the specific technical details of Lisp (macros, parse trees) against the original text if they wish to understand the deep mechanics of why Lisp is more powerful.