How well do your pieces fit?

How well do your pieces fit?
Photo by Omar Flores / Unsplash



I could bow out of writing today's newsletter by asking you to watch the entirety of this 1994 talk by Russ Ackoff. It's just that good. (He also praises architects [my late father's profession, Happy Father's Day, if you were able]).

Instead, I'll commit a crime against the thesis of the talk and pluck out the part that makes my heart sting:

Class is in session.
I read in the New York Times recently that 457 different automobiles are available in the United States.

Let's buy 1 of each and bring them into a large garage ... and determine which car has the best engine ... which one has the best transmission ... the best battery ... and one by one, for every part required by an automobile ...

Now we remove those parts and put them together into the best possible automobile ... because now we'll have an automobile consisting of the best parts.

What do we get?

You don't even get an automobile. For the obvious reason that the parts don't fit.

The performance of a system depends on how the parts fit.

Not how they act taken separately.

This one truth has cascading implications for your product, your team, your company, even your life:

  • If your product is failing to deliver enough value, one of the worst things you can do is add more features, because you'll only increase the difficulty of making all of the features work together, and it's the fit of those features that make the product valuable.
  • If you're failing to break into your prospect's consciousness, one of the worst things you can do is add more use-cases to your website, because you'll only make it more difficult to weave those use-cases together in a way that fits your positioning.
  • If your team is failing to produce enough decisions to get product out the door, one of the worst things you can do for its performance is hire more people, as you'll only increase the difficulty of fitting those people into the communication and collaboration streams that make everyone effective.
  • (Life application left to the reader.)

I don't know how much you want to let it wash over you, but the levee it breaks for me is this:

Your startup is a system that's creating and delivering value to a customer. Whatever part of that is failing, the solution is to focus on the fit, not the presence, of the parts.

This explains why so many x/y-fit phrases have emerged in recent years. Product/market-fit, founder/market-fit, employee/culture-fit, and on and on. Everything demands fit.

That Tolerant Market

As engineers we know it's amazing anything works at all. But good engineers also know that within each of these systems there is also something called tolerance. Errors will occur, the hole won't exactly be where it should, we need to allow for some margin, some wiggle. This is good news: fit isn't binary.

The counterbalancing bad news is: fit is multiplicative. Sloppy or imperfect fit at lower levels creates a structure that sways, leading to volatile performance:

  • You don't have much cash, but this doesn't prevent you from hiring someone.
  • Your compensation isn't quite optimal given the risk someone is taking, but this doesn't prevent you from making someone an offer anyway, and it doesn't prevent them from taking it.
  • Feeling less than fairly rewarded, your employee is slightly unhappy. But this doesn't prevent them from working.
  • They do their work, even though it's slightly less than their best, but it's good enough (this less-than-my-best also makes them even less happy, but that's another branch of the tree).
  • Your product is now composed of features that are good enough, but that doesn't mean it doesn't exist or can't be sold.
  • It's hard to demo a product that's not quite matching your vision, but you come up with a way of demoing it anyway as a loose collection.
  • This looseness means it's impossible for the messaging to be totally crisp, but that doesn't prevent people from seeing some value, so some people buy anyway.
  • Bless its heart, the market is tolerant. But your revenue charts are sort of ¯\_(ツ)_/¯ instead of ᕕ( ᐛ )ᕗ

When approached from the bottom of the stack, the behavior of the top of the stack is a lot less surprising, isn't it?

Obviously, more cash at the bottom of the stack wouldn't solve the whole system. You could still apply it incorrectly, and you can introduce other toxicities at other levels that negate the presence of that reward. Bob with golden handcuffs know this well.

A Fresh Perspective

So let's assume, for the sake of argument, that you want to find product/market-fit, but reading this makes you wonder if even have feature/product-fit, or product/messaging-fit.

Should you start by asking why you don't have that fit? Maybe not.

Five whys is great for rooting out bugs. But a lack of revenue acceleration (or however you want to define P/M-fit) can't be treated like a bug. It must be treated as a report on the overall fitness of your system. Things are, metaphorically, too loose. And this lack of strong fit at each lower level is costing you the big picture.

If your system isn't performing, take a look at how well the pieces fit, starting at the bottom, with yourself, then working up and out. At each level, question the shape of the pieces in play and how well they're serving the whole.