What an average hides
Hotels use averages everywhere. Average occupancy. Average length of stay. Average room rate. Average laundry cost. Average consumption. They are useful because they take something complicated and turn it into a number we can understand. But averages can also hide exactly what we need to see. Suppose a hotel loses, on average, 2% of its textiles every month. That number may be useful for budgeting, but it tells us very little about what is actually happening.
Perhaps sheets are hardly being lost at all while bath towels are disappearing at three times the average rate. Perhaps most losses occur during the summer. Perhaps one building or floor behaves very differently from the rest of the property. Or perhaps the average looks stable because a serious deterioration during the last two months is being diluted by ten months of much better performance. The average is correct. The conclusion we draw from it may not be. The same problem appears when we look at consumption.
Demand varies across the operation
Imagine that a hotel uses an average of 1,000 bath towels per day. Should it therefore maintain enough stock to support 1,000 towels of daily demand? Probably not. On quieter days it may use 700. During a busy weekend it might need 1,400. A resort could move from relatively low occupancy to almost full occupancy within a few weeks. Guest profiles change. Length of stay changes. Pool and spa activity changes. Even weather can influence how certain textiles are consumed.
Managing inventory around the average can therefore create a curious situation: too many textiles during quiet periods and not enough when they are actually needed. Laundry turnaround has the same problem. An average turnaround of 24 hours sounds excellent. But it could mean that almost every load reliably returns in approximately 24 hours. Or it could mean some textiles return after 12 hours and some after 36. Mathematically, the result may be similar. Operationally, it isn’t.
Uncertainty creates a stock requirement
The second hotel requires a larger buffer because uncertainty itself creates an inventory requirement. This becomes particularly relevant when hotels rent their textiles. A contractual or historical average can gradually determine how much stock is allocated to the property. If that stock is designed around inefficient peaks, inconsistent turnaround or poorly understood consumption, the hotel may end up paying for a permanently larger textile pool simply to protect itself against occasional uncertainty.
Understand the distribution and trend
Again, the hotel pays for that inefficiency whether or not it owns the articles. None of this means averages are bad. They are enormously useful. The problem begins when an average becomes a substitute for understanding the distribution underneath it. A useful textile-management system should therefore not only tell us the average loss, consumption or turnaround. It should help us see variation, trends and exceptions. Is consumption increasing? Is turnaround becoming less predictable? Is one article behaving differently from the rest?
Was last month genuinely unusual, or is it the beginning of a new pattern? These are the questions that eventually allow hotels to move from explaining what happened to anticipating what might happen next. There is a reason revenue managers don’t simply take last year’s average occupancy and use it to price every room for the next twelve months. They understand that when something happens can matter just as much as how much happens on average. Perhaps textile management should begin thinking the same way.