The Grand Unified Theory of Management

Theory of Constraints

The Theory of Constraints is the closest we have to a Grand Unified Theory of Management. It cuts straight to the problems that matter, be it in business or your personal life.

It’s a simple statement:

Any system with a goal has one limit. Worrying about anything other than that limit is a waste of resources.

Eli Goldratt propounded it in his 1984 book, The Goal, as a series of observations on assembly line manufacturing. But it applies to nearly everything.

Why is it so powerful?

Three Easy Pieces

The Theory of Constraints is a set of three simple statements, in a tightly connected chain of logic:

#1: Every system has one bottleneck tighter than all the others.

A limiting factor more limiting than the others. A weakest link in the chain.

#2: The performance of the system as a whole is limited by the output of this one bottleneck.

If you increase the throughput of this one bottleneck, the throughput of the entire system increases.

#3: Therefore, the only way to improve the performance of the system is to improve the output at the bottleneck.

What does this mean? It means that any improvement not at the constraint is an illusion. For the same reason there’s no way to strengthen a chain without strengthening its weakest link.

You can do your best to increase capacity of all the non-bottleneck steps. Or, when you realize there’s actually excess capacity at all these steps (by definition – when the bottleneck is at full capacity, all other steps have excess capacity) you can do your best to fill them up.

But it will make no damn difference at all.

Stated even more simply, this is the Theory of Constraints:

Any system with a goal has one limit. Worrying about anything other than that limit is a waste of resources.


Theory of constraints - Summary

So, solving any problem is a series of three simple (but not easy) steps:

  • Step 1: Identify the current most limiting factor.
  • Step 2: Identify the most obvious way to improve this limit, and implement it.
  • Step 3: Go back to Step 1. Because there will be a new limiting factor. There is always one limiting factor.

Why do we not do this totally obvious thing?

When your business struggles to deliver its output, or you struggle to lose weight despite all your attempts – it’s usually because of one of three reasons:

Solving the wrong limit:

You’re hiring more capacity in the sales team to get more deals. But the constraint is really in the delivery team.

In the weight loss domain: you’re cutting calories, when the issue is that you’re stressed and cortisol is wrecking havoc with your body.

Solving yesterday’s limit:

This is a little like army generals who fight the last war.

Yes, sales was the constraint 6 months ago, and so you decided to hire 3 people. But it’s no longer the limiting factor after you got one person.

Once a limit is removed, doing more of it just won’t help.

As Taylor Pearson says in his article, “what got you here won’t get you there”. What gets a business off the ground is not what keeps it in the air.

Not realizing when the limiting factor has changed:

By focusing the entire team on sales proposals, guess what, delivery is suffering. So now inflow of business is no longer the limit. It’s actually servicing that business.

But you don’t realize it until 3 months from now, when the clients don’t renew.

There’s a meta-reason too.

There’s a meta-reason too, which causes us to make these errors.

Seeing something that’s hard to solve, we decide to solve something easier.

A great example of this is the classic Bike-Shed Effect, or the law of triviality.

From Wikipedia:

Parkinson provides the example of a fictional committee whose job was to approve the plans for a nuclear power plant spending the majority of its time on discussions about relatively minor but easy-to-grasp issues, such as what materials to use for the staff bike shed, while neglecting the proposed design of the plant itself, which is far more important and a far more difficult and complex task.

The example is fictional. But we’ve all seen it.

Like spending a company leadership meeting discussing how to increase profits by reducing employee per diems. Instead of the real strategic choices to navigate a recession. (I’ve seen this).

Sometimes it’s almost too banal.

We start the meeting saying, “there are 4 topics to discuss today. Let’s do the easiest 3 topics quickly, and then devote some time to the thorny one”. Well, guess what, you spend 25 min on the easy topics (Parkinson’s Principle – work expands to fill the time you give to it). And then the next 5 min scheduling a follow-up to discuss the thorny issue.

OK, I’m being a little unfair. Because this stuff is HARD. We want to make progress. It boosts our self-esteem, among other things. And ticking a box “DONE” feels like progress. Even when it isn’t.

At other times, this mistake is almost invisible. It looks like more legitimate progress.

I used to run a consumer internet startup in India. An app called Smart Saver, where people could upload receipts from their grocery purchases, to get cashbacks.

During our initial days, my team and I brainstormed and built a lot of features to keep people on the app. Including both mechanisms (do xyz tasks to earn more cashback) and content (giving cashbacks for brands we hadn’t yet partnered with), etc.

But any progress on this was an illusion.

As a small app with 20K users at that time, the main constraint wasn’t retaining those 20K users. It was a constraint, yes. But it wasn’t the critical one.

The critical constraint was going from 20K → 200K users.

Any cool features we were building would solve only the less-important constraint.

But it was actually worse than that. Onboarding funnels themselves are leaky. So, any down-funnel feature would be seen only by a proportion of the users who installed the app. The rest would have dropped out of the funnel at the top! So, our cool and nifty features would be seen only by a small proportion of the miniscule 20K users we had!

I wrote about this in Drunkards and Streetlights:

We brainstorm and build cool new features for our apps, when the onboarding funnel itself is leaky. Andrew Chen calls this the next feature fallacy….

for most of a startup’s life (after Product/market fit, but before it becomes a Facebook), the top of the funnel (i.e., how you get new users) is the biggest constraint.

And any improvement not at the biggest constraint is an illusion. For the same reason there’s no way to strengthen a chain without strengthening its weakest link.

We realized this 6 months in. We started focusing our efforts solely on that critical constraint – building our user base and onboarding flow.

We did get to 300K users. And then we hit a different critical constraint, one that we couldn’t solve. Another story for another day.

Any progress not at the constraint is a trick you’re playing on yourself.

I’ve been angel investing in startups for the last 3+ years, with middling success. It was only recently (after a 10x exit, as it happens), that I realized the critical constraint. The structure of angel investing itself.

Until I solve for that (work in progress), I won’t be more successful.

I wrote about this as well in Drunkards and Streetlights:

…the hardest part is finding the best companies.

Even if you’re almost psychic at picking winners, you can only pick winners among the startups you see (i.e., under your streetlight).

And in a power law world (which the world of startups certainly is), one startup makes all the difference. What if it’s one you just missed investing in? That one networking dinner you missed. That one week you were on holiday. The unicorn is just 5m away from your streetlight, but completely in the dark.

And it doesn’t end there. Even if you do find tomorrow’s billion dollar company, so will other, bigger institutional investors. And they will have no remorse muscling you out, as Paige Craig found out with Airbnb….

But let’s say you clear this hurdle as well. …What happens then?

Congratulations. You get muscled out in the next big round. This happened to me and OperatorVC.

OK then. So what do we do?

A cliche (or two) is a good place to start.

First, take a step back.

It’s important to see the larger picture. And you can’t do that when you’re busy solving a problem in some tiny nook of your system.

Next, make a list.

Step #1: Define your objective function: What is the definition of “victory” here?

This may sound trivial, but it isn’t. Far too often, you realize that you (and your team) might not be optimizing for the truly important stuff.

Step #2: Make a list of all the factors that will help you achieve this victory. Be as detailed (or not) as you want.

Important to think broad enough.

If your objective function is “Improving profitability”, just listing all the cost heads isn’t enough.

Far too often, the biggest factor driving (or constraining) profitability is revenue. There’s a limit to how much you can cut, but no limit to how much you can earn.

Similarly, the environment often plays a disproportionate role, but we ignore it. For example, the biggest constraint on your productivity is often in the environment, rather than in your process.

Step #3: Identify the most critical constraint.

This is deceptive. Sometimes the most critical constraint is obvious. At other times, it isn’t.

It helps to ask, for each of the factors listed: “If I make a 20% improvement in this element, will it drive a 20% improvement in the overall outcome?”

That’ll get you thinking. “Yes, improving the efficiency of the finishing machine will help! Oh, but wait a sec – actually then it’ll just bunch up at quality control.”

Remember – at any time, there’s always ONE constraint that’s more critical than others.

Step #4: Fix the critical constraint, and then go back to Step 3 (now there’ll be a new most critical constraint).

A couple of examples.

Taylor Pearson has a great example from business, in his article: Business Strategy: A Framework for Growing Any Business.

Start with the four big heads, and see which is a bottleneck. e.g., Let’s say it’s Personal operations. Then you dive in, and figure out which sub-head is causing the problem. And then you solve for that. Repeat.

This is also something I faced when I was trying to increase my writing throughput.

I used to feel that I’m not reading enough, and that was hampering my writing (due to inadequate source material).

But when I thought about it (and looked at my notes), I realized that I was actually reading enough. It’s just that I wasn’t processing the notes fast enough. i.e., that’s the bottleneck I need to solve for.

Therefore, now, whether or not I read any new articles / books, I reflect and write everyday.

Soon, the bottleneck might shift back to the top of the funnel. I’ll be ready.


Hope you liked the article. If you’d like to receive more such articles directly in your inbox, don’t forget to subscribe to Sunday Reads!

2 Responses

  1. john May 29, 2021 / 3:26 am

    randomly found your blog by searching ‘x of y’ – ive enjoyed your writing. thanks

Leave a Reply

Your email address will not be published. Required fields are marked *