Every purchasing department I know has one. The list. An export of open order lines, sorted by "overdue", with a column full of notes: chased, no response, promises next week, escalate.
Buyers spend hours a week on it. And the list only looks one way: at the supplier. Who is late? The supplier. Who do we call? The supplier.
After more than thirty clients and years in manufacturing, I see a different pattern. Trace a late order line back to the moment the need arose, and the delay starts in-house more often than we like to admit. Not in one department, but in the handovers between engineering, planning, work preparation and purchasing. Purchasing very much included.
This is not a plea to let suppliers off the hook. Suppliers are late too, through capacity shortages, quality problems or scarcity in the market. That is exactly why you want to know which part of your list really sits with them.
Where the delay often starts
The order goes out late. Sometimes the system works with a lead time that is no longer right: the item master says eight weeks, while in practice the supplier needs ten. MRP then plans every order for that item two weeks late. And that lead time is often maintained by purchasing itself.
Sometimes the order is waiting for a revised bill of materials or a final drawing. In an engineer-to-order environment that comes with the territory, because the customer changes things too. But if the process doesn't allow for it, a large part of the available time is gone before the supplier can even start. Harder still: the customer order is already running, and new items keep being added to the bill of materials throughout the lead time. Purchasing can only order those with less time than the supplier needs. They are on the list as late from day one.
When an order goes out late, the supplier often confirms a later date straight away. If your list measures against the requested date, the line still shows as late. So you end up chasing the supplier about a date he never agreed to.
The requested date is not (or no longer) the real need. Many orders carry a date that was right once but no longer matches actual demand. Urgency is often judged by looking at which production order a purchase line is linked to. But that link shows one order, not the full picture. Look at all demand, stock and open orders per item over time, and a "critical" line sometimes turns out to be covered already by stock or another order. Meanwhile the real shortage sits on a line that isn't even on the list, because according to the order date nothing is wrong.
Orders move, but not everything moves with them. A production or sales order is brought forward or pushed back. The system flags what needs to happen to the orders underneath: expedite, postpone, cancel. But those signals are far from always actioned. The purchase order keeps its old date, or a planned order is only released when it is already too late. The result: material arrives for an order that doesn't start for months, while the order that was brought forward waits for its material. And the supplier only gets the changes that happened to be passed on, without knowing why.
Material is moved from one project to another. Project B is in a hurry, and project A's material is already on the shelf. So it goes to B. That solves today's problem, but the shortage on A only shows up when the material is needed. By then there is hardly any time left to correct course, and that order goes to the supplier as urgent too. One rush job creates the next.
With subcontracting, the supplier depends on you. You subcontract an operation, such as coating, hardening or machining, and supply the material yourself. If that material leaves late, the supplier can no longer meet the requested date, however well he works. If nobody monitors that delivery, the line still shows as late in his name. And on the scorecard too.
Everything becomes urgent. That is often the sum of the points above. But when every escalation is urgent, no escalation is urgent.
You are training your supplier not to take your dates seriously
This is the point we least like to hear. A supplier who gets new dates every week, and sees a "firm" date slip three weeks just as easily, draws a logical conclusion: your dates are negotiable.
There is nothing malicious about that; it is prioritisation. A supplier with limited capacity has to choose, and is more likely to choose customers for whom a missed date has real consequences. If you want priority when supply is tight, you need to be a preferred customer, as the purchasing literature calls it. How much you buy from a supplier and what he earns on it play a part, but so does how predictable you are as a customer. You can't always change your spend. You can change your predictability.
Delivery reliability works both ways. You can demand 95 per cent on time from a supplier. Expecting it is harder when your own dates aren't stable.
What does work
Expediting is still needed, but the answer is not to expedite harder. The answer is to need less expediting. Five things that, in my experience, make the difference:
-
Make sure orders can go out on time. Keep the lead times in your system up to date; MRP is only as good as the data it calculates with. And with long lead times, don't wait for the final drawing if you don't have to: reserve capacity or order the raw material in advance, where the risk is manageable.
-
Link every date to a demonstrable need. Judge urgency on the full picture per item: all demand, stock and open orders over time. Not on the single production order a line happens to be linked to. A date you can back up with the actual shortage is a date you can defend. For anonymous stock material upstream of your customer order decoupling point it works differently: there the need is a stock shortage, and you solve that structurally with inventory policy and framework agreements, not by chasing.
-
Only send the supplier changes that matter. Review the rescheduling messages from MRP (expedite, postpone, cancel) at a fixed moment, for example weekly, with purchasing and subcontracting together. Not every message has to lead to a new date for the supplier. Agree within which window you only change dates by exception, and from what deviation a new date goes out. And give the reason with every change.
-
Escalate on impact, with data. A line isn't critical because it's late, but because something stops if it doesn't arrive. That can be an engineered part for one customer order, but just as easily a two-euro C-item. So don't look at individual lines, but at all lines per customer order together: is the delivery date to the customer still achievable? If one line already makes that date unachievable, expediting the rest makes little sense. Focus on that one line, or discuss the delivery date with the customer in good time. "You're late" gets a defensive response. "These three lines are needed for a customer delivery in week 46; the other twelve can move to week 49, and that date will then really stand" gives the supplier something to work with.
-
Measure your own reliability as a customer. How often do we change a date after the order has been placed? How many orders ask for delivery faster than the supplier's normal lead time? How long was the need already in the system before the order went out? And do you measure the supplier against the date he confirmed, or against a date we changed ourselves afterwards? Do you measure a subcontractor from the moment he had your material, or from the order date? Without these figures, your supplier scorecard only tells half the story.
From chaser to conductor
Selecting suppliers, negotiating and ordering remain the foundation of purchasing. But more and more, the added value also lies in an overview of the whole chain: from engineering through planning and work preparation to the supplier and back again.
None of the points above is new; they are in the planning handbooks. Yet I keep running into them, because they fall between departments. That is where purchasing has a role. It takes knowledge of the ERP system beyond the purchasing module, an understanding of planning concepts, and the willingness to have the internal conversation about dates that aren't right. That last part is often the hardest. Not because colleagues do their jobs badly, but because it is easier to call a supplier than to challenge an internal process.
This isn't about shifting the blame from the supplier to planning or engineering. It is about knowing which part of the problem sits where, so you solve it where it starts. And if you have your own part in order, you stand stronger with the supplier, precisely on the lines that really are his.
Try it yourself
Take the ten lines that have been on your list the longest and find out, for each one:
- when the need arose, and when the order went out;
- what lead time was in the system, and how long the supplier actually needs;
- whether the item was added to the bill of materials late;
- whether the need moved without the purchase order moving with it;
- whether the material was given to another project in the meantime;
- whether it was subcontracted work, and when your material reached the supplier.
For how many of those ten did the delay start with the supplier, and for how many in-house? I'm curious to hear your results.