A few years ago, I sat in on a last time buy decision at a mid-sized networking equipment maker. The procurement team had done solid work: they’d pulled historical usage data, built a demand forecast, added a margin for error, and arrived at a number. It was a good number. Nobody in the room questioned the math.
What nobody in the room mentioned was treasury. And that, to me, was the actual story.
A last time buy happens when a component reaches end-of-life and a supplier tells a customer this is the final opportunity to order it, ever. From that point forward, whatever gets ordered has to last for the remaining life of the product it supports. Sometimes that’s two years. And it can be ten. I’ve seen buys sized to cover programs that were still shipping fifteen years later.
Procurement teams are good at the forecasting side of this problem. They’re trained for it. What they’re generally not equipped to evaluate is what happens to the capital after the order is placed. Once that purchase order goes out, the money is gone. Not spent-and-recovered-through-sales gone. Committed gone, for as long as the inventory sits in a warehouse waiting to be consumed. If the forecast is generous, that could be a decade.
I don’t think this is a procurement failure. It’s a structural gap. The last time buy sits at the intersection of demand planning, product lifecycle management, and working capital, and in most organisations, only one of those three disciplines is actually in the room when the decision gets made.
Why this looks like procurement but behaves like treasury
Consider what a last time buy actually commits a company to. It’s a single, irreversible capital allocation decision, sized against a demand estimate that will not be tested for years. There’s no reorder point. There’s no supplier relationship to renegotiate quantities with next quarter. Get it wrong in one direction, and you’re paying spot-market premiums or redesigning a product under time pressure. Get it wrong in the other direction, and you’ve tied up capital in inventory that depreciates in relevance every year it isn’t consumed, with no natural mechanism to unwind the position.
That second scenario is the one I think gets underpriced. Excess LTB inventory doesn’t show up as a dramatic write-off most of the time. It shows up as a quiet drag: cash that could have funded something else, sitting in a warehouse, for years, because a demand curve was estimated a little too generously back when nobody was thinking about the balance sheet implications of the number.
If a treasury team saw a proposal to lock up a comparable amount of working capital in, say, a long-dated financial instrument with no early exit, they would ask hard questions about duration risk, opportunity cost, and what happens if the assumptions underlying it are wrong. Those are the right questions. They’re just not being asked, because the request never reaches treasury framed as what it is. It arrives, if it arrives at all, as a routine purchase order.
What changes when financing structure enters the conversation
Here’s where I think there’s real opportunity, and it’s the reason I’d argue this belongs in a treasury and trade finance conversation. The size of the last time buy is usually treated as fixed once the forecast is set. But how that inventory is financed doesn’t have to be.
Structures exist, and more are being built, that let a company secure the components it needs without fully absorbing the capital commitment on day one. Inventory can be held by a financing partner and drawn down over the life of the program, with payment tied to consumption rather than to the original order date. Done well, this converts a single large, irreversible capital outlay into something that behaves more like a managed exposure: sized appropriately for the risk, released gradually, and structured so the cost of being wrong in either direction is smaller.
That’s not a procurement innovation. It’s a financing structure applied to a procurement problem, and it requires someone in the room who thinks natively in terms of capital cost, duration, and risk transfer. In most organisations I’ve encountered, that person isn’t invited to the last time buy conversation, because nobody has framed it as a conversation that needs them.
A modest proposal for treasury teams
I’d suggest something simple: ask to see the last time buy pipeline. Most companies with legacy or long-lifecycle product lines have one, whether it’s formally tracked or scattered across procurement spreadsheets. Ask what assumptions the sizing is based on, what the carrying cost looks like if the forecast runs long, and whether the financing structure was ever considered as a variable.
In my experience, the answer is usually that it wasn’t considered at all, not because anyone made a bad decision, but because nobody framed the decision as one that involved treasury in the first place. Fixing that doesn’t require new tools or new capital. It requires recognising that a last time buy is a liquidity decision that happens to be wearing a procurement costume, and inviting the people who understand liquidity to help take it off.
