The Growth System Diagnostic: The Worksheet

I scored six companies I have worked inside against the same rubric, dimension by dimension, without looking at what happened to them.

Three of them survived or exited well. They land between 66 and 72 out of 100.

One of them shut down. It lands at 51.

That company was not lazy. In its final years, it cut cost per acquisition by 2.5 times and gained fifteen points on measurement. It got better at something it was already good at, and died anyway, because the dimension holding it down never moved.

I have stopped being surprised by that pattern. Companies improve their strengths and call it progress.

So here is the worksheet I use instead. Six dimensions, four observable levels, and one piece of arithmetic that most maturity models get wrong.

It will not give you a grade for a board deck. It gives you a floor, and this article argues that your floor is the only number in your growth system worth acting on.

1. Score six dimensions, not five layers

Last month I published a five-layer version of this diagnostic. It was right about the sequence and wrong about the surface area.

Those five layers described what a growth plan does. They left out what decides whether the plan survives.

The worksheet scores six dimensions:

• Positioning and market clarity. Is it written, differentiated against what buyers actually use instead, and consistent across the site, the deck and the ads?
• Product and activation. Do you know what an activated customer is, and where people drop off before they get value?
• Retention and growth loops. Do you read retention by cohort and source, and does anything bring the next customer without paid spend?
• Unit economics and commercial model. Do you know acquisition cost, margin and payback by channel and segment, not blended?
• Measurement and data infrastructure. Are marketing, product and revenue data joined to one customer record you trust?
• Growth leadership and alignment. Does one person own the commercial number across product, marketing and data?

The last two are the additions, and they are the ones that break companies quietly.

Measurement is almost never funded as growth. It shows. Only 8% of companies have a full view of their marketing performance across channels, and only 18% are very confident they can attribute accurately.

Zilingo is the version that ends badly. It raised $308 million at roughly a $970 million valuation, then posted cumulative losses over $430 million while filing no Singapore financials after FY2019. Every other dimension looked fundable until nobody could verify anything.

💡 Key Takeaway: A dimension nobody funds still caps everything you do fund.

2. Score observable behaviour, not intent

Most self-assessments ask you to rate yourself one to five. That measures confidence, not capability.

Kruger and Dunning put a number on the gap. Poor performers overestimated their own percentile rank by roughly 50 points. The people furthest from competent were the most certain.

So the worksheet never asks how good you are. It asks which description matches what someone would see sitting with your team for a week.

• 1. Implicit. It exists in someone’s head. Not written, not measured. Nobody could hand it to a new hire.
• 2. Stated. It exists as a document, a dashboard or a number. It does not drive decisions.
• 3. Managed. It is defined, instrumented and owned. Someone is accountable for it moving.
• 4. Engineered. It is tested, targeted and forecast forward. It sets limits on other decisions.

Score each dimension one to four against those descriptions.

One rule keeps the exercise honest. If your answer is “I am not sure,” score it a one. Not knowing is not a gap in your data. At the bottom level, not knowing is the observable state, and it is the finding.

💡 Key Takeaway: “I’m not sure” is not a blank. It is a score.

3. Your score is a ceiling, not an average

Multiply each dimension score by 25. That puts all six on a 100-point scale: 25, 50, 75, 100.

Now average them. That number is not your result. It is your best case.

Your result is your weakest dimension plus 20, or your average, whichever is lower.

Take a shape I see constantly. Positioning 100, activation 100, loops 100, economics 25, measurement 75, leadership 75. The average is 79, and most founders would report the 79.

The weakest dimension is 25. The score is 45.

None of the 100s is fake. That company is genuinely excellent at three things. It cannot convert any of it into money, because nobody has worked out what a customer costs or returns.

The obvious objection is that the cap punishes you arbitrarily. It does not. In the theory of constraints, total throughput only improves when the constraint improves, and work spent anywhere else returns close to nothing. The cap is that idea as arithmetic.

It is also generous. It grants twenty points of headroom rather than dragging you down to your floor, and it only bites on lopsided profiles. Score evenly and your average stands, because you have no single bottleneck.

For a current client, the customer clustering model I built for lifecycle communications lifted conversion rate by 38%. That work sat in retention and growth loops: the dimension companies score generously and fund last.

💡 Key Takeaway: Your average is what you are capable of. Your floor is what you will get.

4. The worksheet tells you where, never why

Two limits, and they are why this is usable rather than decorative.

The first is causation. This ranks your dimensions. It cannot tell you which one caused the others. One company in my set scored positioning as its weakest dimension while the real cause sat a row higher: a chief executive who would not delegate. Better messaging would not have saved it.

The second is precision. Treat any gap under ten points as noise. I once scored the same dimension at the same company 45, and then 65 two days later, against the same unchanged rubric. Same rater, same evidence, twenty points apart.

That is the honest resolution of any self-scored instrument. A worksheet claiming better than that is selling you a horoscope.

What survives both limits is still worth an hour. You get a hypothesis about where your system is weakest, specific enough to argue with. Establishing why it is stuck is a different job.

💡 Key Takeaway: A hypothesis you can test beats a grade you can frame.

Final Thoughts: fix the floor, not the average

Go back to the company that scored 51 and shut down. It spent its final years improving measurement and cutting acquisition cost. Both were real gains. Neither was its floor.

That is what makes this worksheet uncomfortable and useful. It gives you no credit for what you already run well.

So score all six this week. Multiply by 25. Find the lowest. That number is your next quarter, whether you choose it or not.

If your floor is obvious and you know why it will not move, you do not need me. If you have scored it and cannot work out what holds it there, that is the conversation worth having. Book a discovery call, or connect with me on LinkedIn.

---

A note before you close this tab. The fact that you read this far tells me something. You already sense that the way you’ve been thinking about growth might be incomplete. That instinct is worth following.

Mervyn Chua is a growth-transformation consultant helping founders and CEOs build the strategic clarity and systems to grow in an AI-first world. If this raises questions worth exploring for your brand, let’s talk.