The Peter principle
David Peterson

Whether you know it by name or not, you’ve probably heard of the Peter principle. First appearing in Laurence J. Peter’s book The Peter Principle, it goes as follows: “in a hierarchy, every employee tends to rise to his level of incompetence.”

If you’ve spent any time in any organization, big or small, this aphorism likely rings true. Or just think back…isn’t it true that almost every bureaucracy you’ve dealt with lately is filled with people ill-suited to their tasks? As Peter puts it: “the cream rises until it sours.”

The irony is that, if you were to read The Peter Principle, you’d notice that Peter was not offering management wisdom…it was satire! This was a send up of stuffy management books (e.g. according to Peter, one of the many hazards of ladder climbing was so-called “Tabulatory Gigantism” — an all-consuming obsession with having a bigger desk than one’s colleagues).

And yet, it turns out that the Peter principle may indeed be provably true.

Authors Benson, Li and Shue found in their paper “Promotions and the Peter Principle” that firms promote based on past performance, not future potential, and are suffering as a result. In the paper, they were able to show that firms promote high-performing salespeople ahead of lower-performing salespeople with greater managerial potential, and they were able to estimate these suboptimal managers “decreased subordinate performance” by 30%.

To be clear, they’re quick to caveat this estimate. They do not mean to suggest “that firms would actually achieve a 30% gain in sales” by changing their promotion policy, as such a change could alter incentives in other ways that aren’t easy to model.

But still…that’s a pretty big result! If you could somehow boost sales by 30%, even 10%, just by changing your promotion policy, wouldn’t you do it?

One of the glorious things about working at an early stage startup is that the Peter principle is irrelevant. Everybody is either building or selling. And the team is small enough that no hierarchy is needed. You’re all on the same page simply by existing in the same room together.

But one thing I’ve noticed is that once a company raises a series A, founders suddenly want to grow up. And the most obvious way to grow up is to start adding hierarchy. And the most obvious way to add hierarchy is to promote your top performers.

The Peter principle strikes again.

This is the hierarchy trap that I’ve written about before. It’s hard to blame a massive organization for being slow to evolve, but startups should be able to plot a more optimal course given the data at hand. So if you’re starting to grow up, remember the Peter principle. Maybe even give that paper a read. And think carefully about how you promote and who you promote, because it turns out, there’s quite a bit of alpha in doing it well.

David

FROM THE BLOG

Kafka Survivors of the World, Unite!
Why we backed Memphis.dev.

The Problem with Startup Advice
The best founders and investors know the rules, but also know when to break them.

Looking Back to Move Forward
How to survive this extraordinarily exciting and wildly disconcerting age of generative AI.

LLMs and the Future of Customer-built Software Design
How will LLMs change software development and design?

EUROPE & ISRAEL FUNDING NEWS

Israel/Generative AI. AI21 raised $155M for its developer platform for building custom text-based business apps.

UK/E-Commerce Tooling. Shop Circle raised $120M to continue to build out its suite of tools for e-commerce operators.

Hungary/Autonomous Vehicles. Commsignia raised $15M to scale its “vehicle-to-everything” monitoring platform.

Germany/Developer Tooling. AskUI raised $4.66M to build its LLM-powered UI test automation platform.

UK/Compliance Tech. Worldover raised $3.8M to build a compliance platform for launching new products in multiple markets.

UK/LLM Tooling. Context.ai raised $3.5M to bring product analytics to LLM-powered applications.

WORTH READING

ENTERPRISE/TECH NEWS

IPOs on the horizon. While funding for startups has remained low — deal counts in August 2023 were almost half of those in August 2022 — there is a glimmer of hope with the IPO market finally starting to open back up. “Last month, two well-funded venture-backed unicorn companies — grocery delivery platform Instacart and targeted marketing platform Klaviyo — filed plans to go public in September. When they do, their offerings and Arm Holdings’ planned Nasdaq debut — slated to be the largest tech IPO in almost two years — could buoy the entire startup funding landscape.”

Meta’s mega AI drama. After releasing the free and open-source Llama and Llama 2 to much acclaim, the teams at Meta responsible have been plagued with drama. In fact, over half of the 14 authors of Meta’s February 2023 Llama research paper have since quit — largely due to internal battles over compute resources.

Future Fund’s deadline looming. In the midst of the pandemic, the UK government created the Future Fund as a lifeboat for British startups. However, this well-intentioned initiative may now have dire consequences. The convertible loans offered by the Future Fund convert to equity in the next funding round, which should happen within three years. “As private funding has become scarcer for startups, the three-year conversion deadline is now looming for hundreds of Future Fund-backed businesses. The terms of the Future Fund agreement say that if a startup fails to raise a priced round before that date, it must repay the full amount, plus a 100% “redemption premium”. Struggling companies either pay a bill double that which they borrowed — or face bankruptcy.”

HOW TO STARTUP

No VC, no problem. The Information recently shared a great profile on David Holz, Midjourney’s founder. Midjourney was able to grow to $200M in revenue all while not raising any money from VC firms.

YCs’s AI takeover. This year’s YC batch is over 65% comprised of AI startups, a massive increase from previous batches and there’s a lot we can learn from these YC AI startups. While YC’s AI frenzy has been noteworthy, some investors are still choosing to sit out their Demo Day as YC’s signal has faded a bit in recent years. “Some investors don’t like that the cohorts have gotten larger and less selective, while others think the base valuations and round sizes have become too inflated. Demo Day doesn’t seem to come with the same pomp and circumstance as it used to.”

HOW TO VENTURE

Seed investor biases. Jonathan Lehr from Work-Bench thoughtfully penned his take on seed investors’ biases and incentives. While the entire article is worth reading, his section on advantages for founders to work with a concentrated seed fund, compared to accelerator programs like YC is key: “while faced with more net dilution, the trade-off is that founders receive significantly more hands-on support throughout their scaling process given there is a reputational and financial impact hanging on each companies’ success. Concentrated seed firms offer support ranging from recruiting services, go-to-market counseling, technical support to aid product development, and more. For first-time founders, this added support can be a game changer in the survival of their company. Even for multi-time founders, the added boost of early support can help them achieve more milestones in a shorter amount of time. Additionally, concentrated seed firms help garner additional capital beyond the seed round. This can take the shape of providing financial support to struggling companies that need extended runway, supporting the raise of a seed extension round, or taking a hands-on approach in helping secure a Series A raise — between assisting to craft the narrative, creating and reviewing collateral, and making targeted VC intros.”

PORTFOLIO NEWS

LightSolver is building a laser computer which works at the speed of light. “LightSolver was created by Dr. Ruti Ben Shlomi and Dr. Chene Tradonsky, physicists of the Weizmann Institute of Science in Rehovot in central Israel. and it is “faster than any kind of alternative computer, whether it’s a quantum computer or classical traditional computer,””

Aquant’s CEO, Shahar Chen, shared insights in a recent Forbes article, debunking common misconceptions about AI.

Steadybit’s CEO, Benjamin Wilms, spoke on an episode of Category Visionaries about why resilience is everything in the modern software economy, and why so many companies struggle to build it into their more complex systems.

JFrog appointed Shanti Ariker as their new Chief Legal Officer.