How to Know When Your Business Model Is Broken

Saim Abbasi has seen the business model problem pattern clearly enough to recognize it early now. It usually looks like this: a company is growing, the revenue numbers are impressive, the team is energized, and then somewhere around the 18-month mark, the economics start to look wrong. Customer acquisition is getting more expensive. Retention is not as strong as the early cohorts suggested. The unit economics that looked acceptable at small scale are looking problematic at larger scale.

The Unit Economics Signal

The ratio between customer lifetime value and customer acquisition cost is the single most important diagnostic in most consumer and B2B subscription businesses. When that ratio is below 3 to 1, meaning you earn less than three times what you spend to acquire a customer over the life of the relationship, the business model requires either a lower acquisition cost or a higher lifetime value before it can be sustainably profitable at scale.

The problem is that this ratio often looks acceptable in the early cohorts because the early customers are the best customers, the ones who found the product through word of mouth, needed it most, and churn the least. As paid acquisition scales, the cohort quality tends to decline, and the LTV:CAC ratio deteriorates with it. Founders who base their business model confidence on early cohort data are often surprised by this deterioration.

The Broken Model Versus the Unoptimized Model

Not every unit economics problem is a broken business model. Some are fixable with operational improvements, pricing adjustments, or segment focus. The distinction Saim makes is this: if the unit economics can reach sustainability by moving a lever that is within the company's control, that is an optimization problem. If reaching sustainable unit economics requires something outside the company's control, such as a market shift in acquisition costs or a fundamental change in customer behavior, that is a model problem.

When to Declare the Model Broken

The point at which a founder should declare the model broken rather than optimizable is when the improvement required is more than 2x better across every dimension simultaneously. Getting your acquisition cost down by 20 percent is achievable. Getting it down by 60 percent while also improving retention by 40 percent and increasing average revenue per user by 30 percent is not a realistic target within any meaningful time horizon. If that is what the math requires, the model needs a rethink, not an optimization.

"If your best customers are not profitable, more customers will not fix that. More customers will just make the problem bigger."