Startups and scale-ups face unique challenges due to their rapid growth, need for scalability, and constant pressure to innovate while managing limited resources. Three organizational theories—Conway’s Law, Parkinson’s Law, and Price’s Law—provide valuable frameworks to help young companies navigate these challenges effectively. I imagine them as three wise men—whose wisdom should be guiding even today’s rapid-scaling AI startups. It’s worthwhile realizing that these ARE laws and that they have proven out empirically over decades. It’s not someone’s intuition or the latest management fad.

Conway’s Law states that organizations are constrained to produce systems that mirror their own communication structures. Introduced by Melvin Conway in 1967, this law suggests that the design of systems developed by an organization will reflect the organization’s structure and communication pathways. Conway’s Law explains some of the disjointed experiences you have with companies and software. My favorite example is the onboarding process for new hires at most companies. The lack of communication between HR, Finance, and IT functions is always evident in your new hire experience—your laptop arrives late, it takes days to get access to all the systems you need, and you spend hours filling in the same information into dozens of different HR and payroll forms. (Don’t get me started on dealing with a mortgage application in the US!).

Parkinson’s Law was first articulated by Cyril Northcote Parkinson in 1955. It posits that “work expands so as to fill the time available for its completion.” This observation highlights inefficiencies in time management, suggesting that tasks can become more complex and time-consuming if excessive time is allowed for them. Yes, any of us who left a school paper till the last minute can attest to this one. And any teacher will know that if you leave more time, folks will still rush to hand in on time at the last minute.

Price’s Law, attributed to Derek J. de Solla Price, suggests that a small number of people in any group will produce a significant portion of the work. Specifically, it posits that roughly the square root of the number of people in a domain does half of the productive work, indicating a disproportionate contribution from a few individuals. If you’re the type of person who reads blog posts like this one, you probably fall into the latter category and have likely experienced firsthand that all successful endeavors and companies are carried on the backs of a desperate few. We talk of “10X engineers” in software engineering since you can get folks who are 10X more productive than the guy next to them. You see the same outsized performance with some of the rainmakers you can find in sales.

So, how do you apply these in Startups and Scale-ups?

Implementing Conway’s Law:

Applying Conway’s Law to startups can significantly enhance their operational efficiency and product development processes. The ideas of “two-pizza” size cross-functional teams, Agile, etc., are all efforts to address this. However, I still find that companies will implement “agile teams” without much regard for Conway’s fundamental thesis. If your teams don’t look like your ideal offering structure, you’ll still have a problem—agile or not. If your backend team doesn’t talk to your front-end team, it will show how your clients experience your product. In an excellent article on Conway’s Law, software engineering thought leader Martin Fowler points out that this doesn’t become a problem until companies start to scale beyond a dozen or two dozen people. Below that number, you’re small enough to have deep and informal communications across functions and roles. Once you grow well beyond that size, and once humans need organizing, Conway starts to exhibit itself. So when you recognize that Conway holds true, one strategy is to do a “Reverse Conway Maneuver” and design your teams to match the desired system architecture. This is essential to do when you consider implementing substantial platform additions or changes.

But this goes well beyond software. How do you solve the onboarding problem I described? Simply form a temporary cross-functional team of HR, Finance, and IT folks, get feedback from all the recent hires in your company, and compile a written process and checklist to design the smoothest process ever. You’ll fix it in a single week-long design sprint.

Leveraging Parkinson’s Law:

Again, much of Agile development and the practices of two-week sprints and “ship early and often” is an effort to address the problems described by Parkinson’s Law. The agile approach helps to prevent procrastination, ensures focused work, and promotes a sense of urgency, pushing team members to complete tasks faster than they might otherwise. Agile methodologies also encourage prioritization, forcing teams to identify and concentrate on the most critical tasks, enhancing overall productivity and accelerating growth.

However, again, I’ve seen “Agile” frequently abused. When you start going through the motions of agile without actually applying the principles and delivering regular and distinct increments of value, you’re back to where work starts spreading over multiple weeks and months. These days, engineering management tools like Jellyfish, of which I’m a big fan, will tell you if your teams are producing on par with what is typical with other startups (Jellyfish keeps benchmark statistics with which you can compare). So, assuming that’s the case, agile fails because stakeholders and clients aren’t actively involved in confirming regular increments of value get delivered. With all this said, engineering teams are generally not the culprits of Parkinson’s Law. How long did it take to upgrade the finance system or roll out the new HR system? Our colleagues in these functions would do well by adopting agile methodologies like Scrum and Kanban. Even the most progressive large companies are doing exactly that.

Utilizing Price’s Law:

Price’s Law is largely a question of talent management. Recognizing that a few individuals often produce most of the results, startups can focus on identifying and supporting these high performers. You’re still getting an absolute bargain if you’re paying a 10X engineer double market rates. Yet very few companies have the foresight to do this, and it’s based on the misguided socialist idea that “everyone is equal.” It’s simply not the case—in startups, impact, like IQ, sits on a Gaussian distribution. Price’s Law holds up in software engineering and sales specifically (and of course, in sports and entertainment), where that distribution is particularly wide. Minimally ensure that your best talent is provided with additional resources, or they are placed in roles where their impact can be maximized. Understanding Price’s Law can guide more strategic hiring and team development processes as startups grow. Ensuring that new hires and team structures are designed to amplify the contributions of these key performers can lead to more effective scaling. And whereas in startup land, you may not be able to pay them double, you SHOULD reward them with stock options proportionate to their contribution.

Conclusion

In conclusion, understanding and applying the principles of Conway’s Law, Parkinson’s Law, and Price’s Law can significantly enhance the efficiency and scalability of startups and scale-ups. By aligning team structures with desired system architectures, fostering a culture of focused and timely task completion, and strategically managing talent to leverage the contributions of high performers, young companies can navigate the complexities of rapid growth and innovation. These empirical laws provide a robust framework for addressing common challenges in organizational dynamics, ensuring that startups not only survive but thrive in competitive and fast-paced environments.

Author

Write A Comment