If a store generates revenue, keeping it open can seem sensible. The same logic is often applied to a channel, product category or customer segment that still brings in income. But revenue does not tell us what a unit contributes to the company. It only tells us how much money passes through it.
The better question is: if this unit disappeared tomorrow, what would we genuinely lose—and which costs, capacity and resources would we release?
That separates growth from volume. Volume rises as a company adds stores, products, channels or customers. Scalable growth requires every unit to keep earning its place. Closing one is not always a retreat; it can be a capital-allocation decision that enables more efficient growth.
Revenue does not reveal a unit's value
U.S. store openings in the first half of 2026 fell to their lowest level since 2020, but the divide between expansion and contraction was category-specific. Dollar and discount chains with clear value propositions continued opening locations, while apparel, department-store and electronics operators contracted. The signal is not that physical retail is disappearing. Capital is moving toward formats that can economically justify their existence.
Cato's recent decision makes that distinction tangible. The company said it plans to close roughly 70 additional underperforming stores in the third and fourth quarters of 2026, bringing planned closures for the year to approximately 120. Management explained that marginal stores had previously been given more time to improve, but meaningful recovery was no longer expected under current conditions. Exiting is not free: Cato estimates $1.0 million to $1.3 million in costs for the additional closures. Even so, it expects the decision to benefit operating results from 2027 onward. This is a decision based not merely on sales, but on forward-looking marginal value.
The Marginal Unit Equation
We frame the decision through a three-part equation:
Net marginal value of a unit = Contribution − Burden − Displaced Opportunity
Contribution is the revenue and margin uniquely created by the unit—and genuinely lost if it disappears. If some customers move to a nearby store or the online channel after a closure, those transferred sales are not lost contribution. Conversely, a store may support brand visibility or sales elsewhere. That indirect contribution belongs in the equation only to the extent that it can be evidenced.
Burden extends beyond rent, payroll and logistics. It includes working capital tied up in inventory, operational exceptions, head-office time and management attention. The last is frequently omitted because it does not appear as a separate line in the accounts. Yet a small unit demanding constant intervention can cost the organisation far more than its reported loss.
Displaced Opportunity is the value the same capital, people and time could create elsewhere. A unit can report a profit and still be economically weak if it occupies a scarce resource that has a substantially better use. This produces the framework's most important counterintuitive insight: not every loss-making unit should close, and not every profitable unit should stay.
One equation, four decisions
Consider a hypothetical store that covers direct expenses but requires constant central support and carries excessive inventory. If much of its sales would migrate to nearby stores or e-commerce, closure could release more value than it destroys.
A hypothetical sales channel may generate little immediate volume but serve as the first point of contact for customers who later purchase directly. If that effect is demonstrated, the channel's visible profitability understates its contribution. If it is not, “brand visibility” can become an expensive assumption.
A hypothetical product category may earn a low margin on its own yet trigger purchases of more profitable products, making its system contribution positive. By contrast, a high-revenue category with frequent returns, slow inventory turns and heavy operational demands may be less valuable than its size suggests.
A hypothetical customer segment may provide recurring revenue while requiring custom processes and long collection periods. If it prevents the team from pursuing better segments, the decisive figure is the value of the capacity released.
Closing is an investment decision too
When net marginal value is positive, the unit stays. When it is close to zero, the answer is not indefinite patience: the business sets a defined improvement period, outcome and owner. When the value is persistently negative, an exit is planned.
The equation does not end when closure becomes the preferred option. Lease penalties, severance, inventory liquidation, customer migration and possible brand effects must enter the decision as exit costs. A one-off exit cost should not be confused with negative value that recurs every year. A closure that looks expensive today may create a stronger operation over the following years—the logic underlying Cato's decision.
At Sellf, we do not read growth only through additional volume. We look at how sustainably and efficiently capital works. A marginal-unit review should therefore not be a one-off cost exercise. It should become a quarterly management rhythm, using the same definitions, decision horizon and equation across stores, channels, product groups and customer segments. If the decision depends on a manager's habit or the pressure of a particular month, there is no system yet.
If you had to open every unit you operate today again from scratch, would you still open it?
