The bottleneck changes with stage.

There is no universal growth playbook, and the brands that believe there is tend to apply last year's solution to this year's constraint. The single most useful question in eCommerce growth is not what should we do. It is what is actually holding us back right now.

Because the answer changes. The lever that doubled the business at one stage is often irrelevant at the next, and occasionally it is the thing now getting in the way. Growth is a sequence of different constraints, each needing a different fix. Diagnose the stage wrong and you can spend a year solving a problem you no longer have.

The constraint moves.

A simplified version of the arc looks like this. It is not a law, but it is a useful map.

• Roughly $1M to $10M, the constraint is usually acquisition cost. The product works and the margins are workable. The job is buying customers efficiently. At this stage paid media efficiency is the whole game, and small movements in CAC compound hard into cash a couple of months out.
• Roughly $10M to $50M, it often shifts to margin and pricing. Revenue is large enough that the operational levers start to dominate. A fractional improvement in maintained markup can free more cash than another grind on CAC, and most marketing teams never look there because it does not feel like a marketing problem.
• $50M and up, it tends to move again, to retention and inventory velocity. At scale, another few points of returning-customer revenue or sell-through can add more than a CAC improvement ever will. The growth lever stops being acquisition and becomes the efficiency of the machine behind it.

The exact thresholds matter less than the principle. The constraint migrates. Generic advice is almost always wrong by the time you apply it, because it is answering the average brand's question, not yours.

Time spent on the wrong problem is the most expensive line item in the business, and it never shows up on the P&L as such.

Why teams keep solving the last problem.

Because the last problem is the one they got good at. The skill that broke the previous constraint becomes the reflex, and the reflex keeps firing after it has stopped working. A team that scaled from $5M to $20M on creative volume and aggressive acquisition will reach for more creative volume and more acquisition when growth stalls at $40M, even when the actual constraint has quietly moved to margin or retention.

This is also why a brand can do everything that worked last year and watch the number flatten. Nothing got worse. The constraint moved, and the effort kept pointing at where it used to be. By the time anyone says the platform is broken, the real story is usually a diagnosis that never got refreshed.

How to actually find the current constraint.

The honest version is unglamorous. You look at the whole system, not just the ad account, and you find the point where the next unit of effort produces the least return. Three places it usually hides:

• Spend allocation versus profit contribution. If a large share of budget is going to existing customers with no real exclusions, acquisition is not the lever you think it is. You are buying repeats you would have got anyway.
• Creative supply versus spend. If creative volume is not keeping pace with the budget, you fatigue winners faster than you replace them, and no amount of bid tuning fixes a supply problem.
• Whether your numbers agree. If acquisition MER, platform ROAS and 30-day LTGP : CAC point in different directions, the constraint is data integrity, and every other decision is being made on a signal you cannot trust.

So if you take one thing away from this

Stop asking what the playbook says and start asking what is holding this business back at this size, right now. The constraint that defined the last stage is rarely the one defining the next. The brands that keep compounding are not the ones with the best single tactic. They are the ones that re-diagnose honestly every time the growth changes shape, and move their effort to wherever the limit has gone.