From the library
Work the System
Sam Carpenter, 2008
Your business is not one big blob of problems. It is a set of separate machines, and you cannot repair any of them while you are inside it, running.
In one paragraph
Carpenter's account of stepping outside a business that was destroying him and seeing it as separate machines rather than one continuous emergency. The contribution is the sequence: strategic objective, then operating principles, then procedures, in that order. Most useful once you already run the operation and are the bottleneck in it. Least useful for creative or novel work, and the conversion-narrative framing pads a fairly ordinary operational insight.
Where this sits in the operating system
This is the execution end of Systems Over Goals. Where other books argue that systems matter, Carpenter is almost entirely concerned with the mechanics of writing one down: what a procedure should contain, who owns it, how it gets reviewed. It is the most literal treatment of Process Documentation in the library, and the reason it sits under four sub-principles at once is that documenting a system inevitably pulls in the memory, automation and operations questions alongside it.
Systems Over Goals
Systems Over Goals
Systems Over Goals
Systems Over Goals
What it does not cover
It says almost nothing about whether the systems are worth running. Carpenter assumes the work is already the right work and optimises the doing of it, so a reader can come away efficient at the wrong things. For deciding what should exist at all, the framework routes to Focus and Essentialism, particularly Elimination Over Addition. For whether the people running those systems are the right people, Right People First.
Read it if
- You own or run an operation where the same work recurs weekly
- You are the bottleneck and most answers still route through you
- Things work when you are there and quietly degrade when you are not
Skip it if
- Your work is mostly novel or creative, where written procedure fossilises fast
- You want strategy. This is an operations book and does not pretend otherwise
- You already internalised The E-Myth. The overlap is substantial
Before you read on
Your most capable operator hands in their notice tomorrow morning. How much of what they know exists anywhere other than inside their head?
Key insights
Your business is not one blob of problems. It is separate machines, and none can be repaired while you are inside it running.
Strategic objective, then operating principles, then procedures. Almost everyone starts at procedures because procedures feel like progress.
Documentation is not a one-time build. Maintenance is the actual cost and the book is thin on it.
Recognise it early
The failure and the correction sit side by side deliberately. Read the left column asking whether any of it is already true of you.
Overcomplicated SOPs
A procedure longer than one screen does not get read under pressure, and pressure is the only condition where it matters.
Docs not used
If the document does not live where the work happens, consulting it costs more than improvising, so people improvise.
Single-point expertise
That knowledge walks out at five o'clock and might not come back. Every undocumented day is a bet you did not know you placed.
No metrics
With no number attached, whether a process is working is a matter of opinion, and nobody can separate real improvement from noise.
One-time documentation
Reality drifts from the document within weeks. A stale SOP is worse than none, because it teaches the team that the documents lie.
How it actually works
Carpenter spent fifteen years inside a telephone answering service that was slowly killing him: eighty to a hundred hour weeks, near bankruptcy, health going. The shift, when it came, was not a new tactic. It was a change of position. He stopped experiencing the business as one continuous emergency happening to him, and started seeing it from outside as a set of separate machines, each of which could be stopped, opened up and worked on without touching the others. His claim is that almost none of those machines were ever designed. They accreted. Which is why they run badly, and why that is not anybody's fault.
Strategic Objective
One page naming what this business is, who it is for, and what it will refuse to do. Every later decision gets filtered through it.
Without it you optimise processes that should not exist, and get very efficient at serving people you should have turned away.
General Operating Principles
Roughly thirty standing guidelines describing how decisions get made here. Not tasks. Judgment rules.
Without them every unscripted situation routes back to you, because nobody else has been told how to think about it.
Working Procedures
The actual step by step documents, one per system, written by whoever does the work rather than by management.
Without them the knowledge lives in people's heads and leaves when they do.
The order is the part most people get wrong. Almost everyone starts at procedures, because procedures feel like progress. Carpenter's sequence is direction, then judgment, then execution. Procedures written without principles produce people who follow steps and cannot adapt. Principles written without a strategic objective produce people making confident decisions in no particular direction.
Worked through
Onboarding a new client takes three weeks, every one runs differently, and you are personally pulled into all of them.
From inside the machine
You work the weekend, apologise to the client, and consider hiring another coordinator. The load moves. The three weeks do not.
Step outside it
Onboarding is one system. It is not sales and it is not delivery, even though it touches both. It can be examined on its own.
Filter it through the objective
Are these clients the business actually wants? Some of the three weeks may be the cost of onboarding people you should have declined at the point of sale.
Write the judgment rule
"We do not begin work before the scope document is signed." Now nobody has to come and ask you, and the answer no longer depends on who is having a bad week.
Then, and only then, the procedure
Seven steps, owned by the person who runs them, with two numbers attached so you can tell next month whether it actually got faster.
The same idea, argued differently
All four books arrive at the same diagnosis: the owner is the constraint. Where they split is what they want you to do about it.
Ryan Daniel Moran
Bill Bonney, Gary Hayslip, Matt Stamper
David Allen
John Warrillow
Reframes the payoff. Warrillow systematises for transferability, because a business that needs you is one nobody can buy. Systems as valuation, not relief.
Where it overreaches
Taking the book seriously means knowing where it is weakest.
The systems mindset is presented as a near-spiritual awakening.
The underlying operational insight is sound and fairly ordinary. The conversion-narrative framing, repeated at length, oversells it and pads the book considerably.
Evidence is one company: Carpenter's own telephone answering service.
That is close to a best case for proceduralisation. High-volume, low-variance, repeatable work. The book does not seriously engage with knowledge or creative work, where writing the procedure down is often what kills the quality.
Documentation is treated as mostly a one-time build.
Maintenance is the actual cost and the book is thin on it. His own pitfall list flags SOP decay, then offers a monthly review as the fix, which is optimistic for a library of hundreds of procedures.
It assumes you have the authority to change the system.
Almost every prescription presumes owner-level control. If you are running a team inside somebody else's org, half the book is unactionable as written.
What the book still has that this page does not
Everything above is the transferable part. Here is what does not survive compression, so you can decide whether the full read is worth it to you.
Fifteen years of grind before the insight arrives
The systems idea is simple enough to state in a sentence, which is exactly why it slides off. What makes it land in the book is watching what it cost him first. Desperation does not compress.
Dozens of worked examples from one real company
You get the same idea applied to billing, to hiring, to answering the phone at 2am. Volume is how pattern recognition actually forms, and one worked example on a page cannot do it.
The full text of his operating principles and document templates
You can read all thirty of his and see which ones are load-bearing versus idiosyncratic to a call centre. That judgment is hard to make from a description.
Repetition
The book says the same thing many different ways, which reviewers call padding and which is arguably the mechanism. A page can tell you the mindset shift. A book can wear you into it.
The full breakdown
Before you close this
Naming the specific moment you will act roughly doubles the odds you do.
You will know it is time to use this when you catch yourself answering the same question for the third time in a month.
Reference
- Author
- Sam Carpenter
- Published
- 2008
- Category
- Business
We value your privacy
We use cookies to enhance your experience, analyze site traffic, and personalize content. You can manage your preferences or learn more in our Privacy Policy.