Which questions a manager should ask about the warehouse

A manager walking into a warehouse usually asks one question: "How much stock do we have?" The answer is a single sum, and that sum hardly helps with a decision. Which part sells, which part sits for years, whether the number in the system matches the shelf, where money is leaking: a total answers none of these. The quality of the question sets the quality of the answer. Below are four questions a manager should put to the warehouse, and what a good and a bad answer to each looks like.

"How long does it sit?"

This is a question about the age of stock. Ask not for the total value but for the distribution by age: how much has been without movement for under 30 days, 30 to 90, 90 to 180 and more than 180 days. The last group ties up the most money and gets the least attention. A good answer is ready within a working day, is split by product group, and names for each long-lying item a person responsible and a decision (discount, return, write-off). A bad answer: "it's seasonal", "we're waiting", "I don't know".

"How accurate is it?"

How closely does the balance in the system match the shelf? Ask: when was the last full count, what was the gap in value and in lines, in how many locations did cycle counts find errors, and have the causes been identified. You set the accuracy target yourself to suit the company; what matters is that the figure exists and is tracked every month. Separate two kinds of accuracy: by value and by location. Accuracy by value can look high because surpluses and shortages cancel each other out, while accuracy by location shows the true level of order. If the warehouse manager does not know the accuracy, nobody is measuring it.

"Where do we lose it?"

Shrinkage does not come from one source. Split it by stage: receiving errors, picking errors, damage, theft, expired goods, confusion in returns. One of the most useful documents is the log of manual stock corrections. How many corrections are entered a week, who enters them and for what reason? If the number is growing, or one or two people enter them, that is a clear trace of a problem. For each stage ask one more question: who will find this error, and when? If a receiving error is found not at receiving but when a customer complains, fixing it is far more expensive. You cannot stop a loss without knowing where it happens.

"What does service cost?"

Two measures are needed here: what share of orders shipped complete and on the promised date, and what it costs to process one order line or one pallet. The first is the customer's view, the second the cost view. Cutting cost alone can damage service, while raising service alone raises the price. Look at both together.

Recognising a good answer

An example, with illustrative figures. The manager asks, "How is stock?" A bad answer: "All fine, we have 9 billion so'm of goods." A good answer: "Stock is 9 billion so'm. Of that, 16 percent, or 1.44 billion so'm, has had no movement for over 180 days, and 0.3 billion so'm has less than 60 days of shelf life left. In the last cycle count 36 of 40 locations matched. I propose three decisions and owners are assigned." The second answer has a figure, a comparison, a date and an owner.

A one-page monthly report can consist of five figures:

• the share of stock with no movement for over 90 and over 180 days (in value and percent);
• record accuracy by location;
• the number and value of manual stock corrections;
• the share of orders fulfilled complete and on time;
• the cost of processing one order line or one pallet.

Signs of a good answer: an exact figure, a comparison with last month, a responsible person and a next step. Signs of a bad one: "roughly", an unchecked "that's what the software shows", needing a week to answer and taking offence when the question is repeated.

Conclusion

The question "how much is there?" keeps the warehouse on the books, while "how long does it sit, how accurate is it, where is it lost and what does it cost?" manages it. The OmborExpert specialists help prepare the indicators that answer these questions and bring them together in a one-page report for the manager.