Distribution & Supply Chain, as Taught by Eli Goldratt · Lesson 3 of 7
When inventory hurts sales
Official reports show two or three percent shortages; the real figure is often above twenty. And the stock meant to protect sales can itself destroy them. Goldratt's case for breaking the link between “protect sales” and “hold more inventory”, and why the three obvious ways to break it fail.
What is the true level of shortages?
Do customers always complain when they don’t find what they want? Usually they say “they might have it in a few weeks” and go to another shop. Does the salesperson record that a customer left empty-handed? Rarely. When a shop asks the warehouse about a missing product and doesn’t place an order, does the warehouse record it? Hardly ever.
So many systems report shortages of 2 to 3%, while they really suffer from more than 20%. Shortages mean lost sales, and many shortages can mean lost customers: “We should switch supplier.”
The high level of shortages comes from the judgment call in “maximum consumption”. So should we be more paranoid, and hold even more?
High inventory has its price
Holding relatively high inventory means high investment, limited cash flow and obsolescence. It significantly increases cost. And we’re back in the conflict: any level we choose is an unsatisfactory compromise.
The prevalent compromise
Most distribution systems pick a period of time, and set each product’s target to its forecast consumption for that period, regardless of its replenishment time. The period is then tuned according to the total investment in inventory. The result is statements like:
“We hold twenty million dollars of inventory: about three weeks.”
What does “about three weeks” really describe?
- Three weeks of every product, on average, across all locations
- A well-balanced system
- Zero of some products and a year’s supply of others
The compromise ignores the most important factor, replenishment time. Products that take long to replenish are short; products that replenish quickly pile up. The average looks reasonable, and hides both.
When high inventory jeopardises sales
We intuitively want to break the link between “protect sales” and “hold high inventory”, not only because inventory costs money. In some cases, high inventory actually jeopardises sales.
Short shelf life. With batching everywhere and long replenishment, goods take a long time to travel from production to the customer. Where shelf life is short, many units reach the customer close to their expiry date. Two things follow. A customer who buys a near-expiry product may find its quality poor, and tell everyone not to buy that toothpaste. And many customers simply don’t buy products near expiry: a shortage nobody notices.
Short market life. Where newer versions or competing products arrive often, holding high stock of the current product creates unpleasant dilemmas: obsolescence when the new product launches; lost income from discounting the old one, which eats into the new one’s market; or delaying the launch and losing share. “We’re doomed.”
Limited cash. High inventory of one range of products ties up the cash needed for others. So the variety offered at the point of sale shrinks, and sales are crippled. “I can’t believe they don’t have it.”
Can we break the link?
Look again at the three assumptions under “to protect sales, we must hold high inventory”:
- Replenishment time is too long. Breaking this means drastically cutting replenishment time. But suppliers charge more for priority, and changing production, ordering and transport lead times takes a lot of time and effort. Very hard.
- Vendors are not always reliable. Replacing vendors helps only if they’re exceptionally bad. Educating them takes a long time, if it works at all.
- Forecasts are inaccurate. Huge effort has gone into better forecasting. But we now know, from chaos theory, that accurate forecasting at the level of each product at each location is impossible. Only trends can be predicted.
Any solution must drastically reduce shortages and reduce inventory. That can only happen by protecting sales without holding more inventory, and breaking any of the three assumptions looks impractical. So what is the solution? Goldratt found it by asking a question nobody asks about forecasts. That is the next lesson.
Check your understanding
Answer all five. Get four right to complete the lesson.
1. Why are real shortages much higher than reported?
- Reports are faked
- Customers who leave empty-handed are rarely recorded
- Shops over-order and hide what they can’t sell
- Software errors
Often 2–3% reported, more than 20% real.
2. The prevalent compromise sets targets by…
- Replenishment time, multiplied by average demand
- Product margin
- Shelf space
- A fixed period of forecast sales, whatever the replenishment time
Hence “three weeks” that hides shortages and surpluses.
3. With short shelf life, high inventory hurts sales because…
- Customers get, or avoid, near-expiry products
- Shelves are full, so new products can’t be shown
- Prices rise
- It doesn’t
A hidden shortage.
4. With limited cash, high inventory of some products…
- Raises sales, since customers see full shelves
- Has no effect
- Shrinks the variety offered, crippling sales
- Lowers prices
Cash spent on one range is unavailable for others.
5. According to chaos theory, what can be forecast?
- Each product at each location
- Only trends
- Nothing
- Next week exactly
Not the product/location level.