What does this article cover?
- Which book does your situation point to?
- When does Collins apply?
- When does Christensen apply?
- When does Goldratt apply?
- What each book loses in transmission
- Where has each aged?
- Who should this article send away?
- Where does this break down?
- People also ask
- About the Stagnation Assassin
Everybody recommends these three and almost nobody says which situation each one is for, which is the only useful thing to say about them. They answer different questions. Collins is about people and capital allocation. Christensen is about a specific kind of competitive threat. Goldratt is about why effort is not reaching output. Reading the right one at the right moment is worth considerably more than having read all three, and reading the wrong one is worse than reading none, because you will find the pattern you went looking for.
Watch the companion video on The Goal by Eliyahu Goldratt: Theory of Constraints.
Which book does your situation point to?
Three questions. Is the problem who runs this and where the money goes, which is Collins. Is something cheaper and worse eating your low end, which is Christensen. Is your output limited at one specific point, which is Goldratt. Most situations answer only one.
If more than one of those describes you, you have a sequencing problem rather than a reading problem, and the constraint question usually comes first because it is the cheapest to answer and it changes what the other two are worth.
If none of them describes you, none of these books is your answer, and the next section of this article will not change that.
When does Collins apply?
When the question is people and capital. Good to Great is about who you put in which seat and what you concentrate resources on, and its most durable contribution is the discipline of deciding both before deciding direction. It is a governance book more than an operating one.
The specific situations it fits are narrower than its reputation suggests. A board choosing a chief executive. An owner deciding whether the current leadership team is the team for the next phase. A company with capital to deploy and several plausible places to deploy it. In all three the question is allocation, of people or money, and Collins is genuinely useful there.
Where it does not help is execution. Having decided who and where, the book has little to say about how the work gets done, and reaching for it when the problem is operational produces a leadership discussion in a business that needed a process change.
When does Christensen apply?
When the threat comes from below. The Innovator’s Dilemma explains why competent firms lose to inferior products: doing the right things, listening to your best customers and pursuing better margins, vacates the bottom of the market for somebody who will improve.
The mechanism is what makes the book valuable and it is frequently misremembered. Christensen’s argument, set out in 1997, is not that incumbents are lazy or blind. It is that good managers face a dilemma because the behaviours that make them successful, serving their best customers and investing where returns are highest, are the same behaviours that cause them to cede the low end to a rival whose product is currently worse.
The diagnostic signature is specific and you should require all of it before concluding you are in this situation. A competitor exists whose product is genuinely inferior on the dimensions your best customers care about. It is cheaper, and it is good enough for a segment you have been happy to lose. Its trajectory of improvement is faster than your market’s requirement is rising. If any of those three is missing, you are facing ordinary competition, which is a different problem with different answers.
That last point matters more than it sounds, and the section on ageing explains why.
Christensen’s argument is not that incumbents are lazy or blind. It is that the behaviours which make a company successful are the same ones that cause it to cede the low end to a rival whose product is currently worse.
When does Goldratt apply?
When effort is not reaching output. The Goal argues that one thing limits a system and improvements anywhere else are close to worthless, which is the only one of the three that tells you where to work before telling you how. It is also a novel.
The format is not incidental. Goldratt wrote it as fiction, which means the method arrives through a plant manager working out a problem rather than through a framework diagram, and that is why people finish it. I have recommended it to operators who do not read business books and they read this one.
Its diagnostic signature: a functioning improvement effort, people genuinely working, local metrics improving, and the number at the end of the business flat. That combination points at a constraint somewhere nobody is looking, and no amount of additional effort elsewhere will move the outcome.
I read it early in a major transformation and it changed how I worked, which I do not say about many books. What it changed was not technique but sequence: I stopped asking what could be improved and started asking what was limiting, and those produce different lists. I now reread it at the start of every new transformation, which is a habit with a cost that the last section of this article covers.
What each book loses in transmission
All three have been degraded by their own popularity into a single phrase that no longer carries the method. If your exposure is the phrase rather than the book, you are holding a slogan, and the slogan is in each case the part that survives being wrong.
Collins becomes a hiring slogan. Getting the right people on the bus is remembered as advice about recruitment, which is roughly the least interesting reading available. The argument underneath it is about sequence: decide who before deciding where, because the where will change and the who has to be able to change with it. Used as a recruitment maxim it justifies hiring well, which nobody was arguing against.
Christensen becomes a synonym for competition. Disruption now describes any new entrant, any technology, and any uncomfortable market development, which strips out everything that made the idea useful. The original claim was narrow and testable: inferior products, serving a segment the incumbent is content to lose, improving faster than requirements rise. A word that applies to everything predicts nothing.
Goldratt becomes the word bottleneck. People use it to mean anything slow or annoying, which is the opposite of the method’s point. The claim is not that bottlenecks are bad. It is that exactly one thing limits the system at any moment, that everything else has slack by definition, and that improving anything with slack produces nothing. Somebody describing three bottlenecks has not found three constraints. They have described three things they dislike.
The common failure is the same in each case. A book makes a specific, falsifiable claim; the claim gets compressed into a memorable phrase; the phrase spreads far beyond the conditions the claim required; and the phrase then cannot be wrong, because it no longer says anything precise enough to be tested. That is what happens to good ideas that become popular, and it is a reason to read the source rather than the summary.
Where has each aged?
Unevenly, and only one of the three has been tested properly. Christensen’s theory was examined across seventy-seven of its own cases and many did not fit. Collins and Goldratt have their own critiques, which I have written about separately rather than compressing them here.
The Christensen test is worth setting out because it is unusually direct. Andrew King and Baljir Baatartogtokh, writing in MIT Sloan Management Review in 2015, surveyed and interviewed 79 industry experts on each of the 77 cases Christensen and Michael Raynor used in their two books. They found many of those exemplary cases did not correspond closely with the theory’s own four elements. Only a handful matched all of them. They also noted that the theory’s validity had rarely been tested, and that the studies which had been published did not provide confirmatory evidence.
Two things stop that being a demolition, and both belong here.
The authors themselves declined to draw the strong conclusion, writing that they do not advocate discarding the theory and arguing instead that managers should evaluate difficult problems from several perspectives rather than one.
And the journal published substantial rebuttals. The sharpest objection is that King and Baatartogtokh extracted four concepts from a large body of work and then treated the percentage of cases matching all four as a verdict, when Christensen never claimed to explain every business decline. On that reading the theory is a test for identifying which currently inferior products are likely to take share, not a general account of corporate failure, and measuring it against the wrong job produces a misleading score.
My own reading is that the critique is right about overuse and wrong about worthlessness. Disruption is invoked constantly for situations that are simply competition, and the three-part signature above exists to stop that. Where the signature genuinely holds, the book remains the best available explanation of why capable people lose.
On the other two, briefly. Collins has a well-known methodological problem shared with the genre, concerning how conclusions are drawn from companies already known to have succeeded, and the subsequent record of his sample is a separate and longer discussion I have written elsewhere. Goldratt’s constraint logic is strongest in a single plant with a single product flow and gets considerably more complicated across multiple products and sites, which also has its own article. Neither summary here is a substitute for those.
Who should this article send away?
Anyone looking for the one book that explains everything. These three answer different questions and none of them answers the other two, which means a reader who wants a single lens will pick whichever they read most recently and apply it to situations it was never about.
The failure is predictable and I have watched it repeatedly. Somebody reads Christensen and every competitor becomes a disruptor. Somebody reads Collins and every problem becomes a people problem. Somebody reads Goldratt, which is my own vulnerability, and everything becomes a constraint. In each case the book is fine and the application is the error, and the error is invisible from inside because the framework will always find something.
The practical guard: before applying any of the three, write down what you would expect to see if this framework did not apply. If you cannot describe that state, you are not testing a diagnosis, you are confirming one.
Where to go instead, and all of it beats a comparison article. Read the primary texts rather than summaries, mine included, because all three are well written and two of them are enjoyable. If you want the critical apparatus for the whole genre, Phil Rosenzweig’s The Halo Effect will make you a harder reader of every business book including these. And if your situation is none of the three questions above, the honest answer is that your problem is probably commercial or financial rather than strategic, and neither a business classic nor I will be much use to you.
Where does this break down?
I reread The Goal before every new transformation, which is precisely the single-framework attachment I have warned about elsewhere. Priming yourself to look for constraints means you will find one, whether or not the constraint is what is actually binding.
That is a real defect in my own practice rather than a rhetorical concession. I have written on this site that implementing from one book is a mistake and that every framework carries the context it was derived from, and I begin each engagement by deliberately loading a single lens. The habit has served me well, and it would serve me well even in situations where it was the wrong lens, because a constraint analysis always produces a constraint.
The partial guard I use is to run the diagnostic questions above before I reach for the book rather than after. Whether that is sufficient I genuinely do not know, and anybody who tells you they have solved their own confirmation bias by noticing it is overstating the case.
Two further limits. My sourcing here is heavy on one study, and the several outlets carrying it are not independent evidence of it. And I have deliberately kept the ageing discussion short for two of the three books because each has a dedicated article on this site, which means this piece is less balanced than it looks on that dimension.
Before applying any of the three, write down what you would expect to see if this framework did not apply. If you cannot describe that state, you are not testing a diagnosis, you are confirming one.
People also ask
Which business book should I read for my situation? Collins if the question is who runs this and where capital goes. Christensen if something cheaper and worse is taking your low end. Goldratt if local metrics are improving while the final number stays flat.
Has the theory of disruptive innovation been disproven? Not disproven. A 2015 MIT Sloan Management Review study found many of the 77 cases in Christensen’s own books did not fit the theory’s four elements, though the authors declined to advocate discarding it and the journal published substantial rebuttals.
How do I know if I am facing real disruption? Require all three signs: a genuinely inferior competitor, cheaper and good enough for a segment you are content to lose, improving faster than your market’s requirements rise. Missing any one means ordinary competition.
Why is The Goal written as a novel? Because the format works. The method arrives through a plant manager solving a problem rather than through a framework, which is why operators who do not read business books finish this one.
Can I use more than one? Yes, but sequence them. The constraint question is cheapest to answer and changes what the other two are worth, so establish whether output is limited at one point before deciding it is a people problem or a competitive one.
About the Stagnation Assassin
Todd Hagopian is a Fortune 500 transformation executive who has generated $3B+ shareholder value across five turnarounds at Berkshire Hathaway, Illinois Tool Works, and Whirlpool. He is VP of Product Strategy and Marketing. Known as The Stagnation Assassin, he is the author of two published books, The Unfair Advantage: Weaponizing the Hypomanic Toolbox and Stagnation Assassin: The Anti-Consultant Manifesto. His blog is published in 15+ languages and read by operators worldwide. Bring him to your stage via his speaking page or connect with him on LinkedIn. His newest book is 10-Minute Transformation (February 23, 2027).
The talk version of this asks the three diagnostic questions before naming any of the books, which usually reveals that half the room has been applying the one they read most recently.

