At some point in the life of a struggling HR system, someone finally says it out loud: this is not working. The value never showed up. The workarounds have become the process. The thing costs more to keep alive each year than anyone will admit. And then, almost immediately, someone else says the sentence that ends the conversation before it starts: we have invested too much to walk away now.
It is one of the most reasonable-sounding sentences in business, and one of the most expensive. It feels prudent, even responsible — a defence of a serious commitment against a rash reversal. It is, in fact, exactly the logic that turns a bad investment into a far worse one. Because the money already spent is not a reason to keep going. It is gone either way, and the only thing that sentence actually guarantees is that more will follow it down the same hole.
This is the sunk-cost trap, and HR systems are unusually good at springing it. They are expensive, they are visible, and their purchase was championed by people who are still in the building. All of that makes the amount already spent the loudest voice in the room — and the loudest voice, in this particular decision, is the one that should not get a vote at all.
The Money You've Already Spent Is the Worst Reason to Spend More
The uncomfortable principle underneath all of this is simple, and it runs against every instinct: money already spent is irrelevant to what you should do next. It is sunk. It is gone whether you continue or stop. It cannot be recovered by pressing on, and it is not lost any more completely by walking away. It genuinely does not belong in the decision.
The only question that matters looking forward is this: given where you are today — not where you started — is the next dollar better spent continuing with this system, or somewhere else entirely? That comparison uses only future costs and future value. What you paid to get here is history, and history is not a line item in a decision about tomorrow. A system that would not be worth buying today, at today's prices, for today's remaining benefit, is not worth continuing to fund simply because you already bought it once.
The reason this is so hard is that walking away makes the loss visible and final, while pressing on keeps the hope alive that the spending will somehow be redeemed. So organizations keep paying, not because the next dollar is well spent, but because stopping would force them to admit the previous dollars were not. That is not financial logic. It is the avoidance of an uncomfortable feeling, dressed up as prudence — and it is paid for in real money, every year it continues.
Three Different Questions, Constantly Confused
Part of what keeps failing systems alive is that the decision to continue gets tangled up with two other questions it has nothing to do with, and untangling them is most of the work.
The first question is "should we have bought it?" This is a question about the past, and it is often genuinely painful, because the honest answer is sometimes no. But it is also irrelevant to what happens next, and worse, it is the question everyone is secretly arguing about when they think they are debating the future. Cutting the system feels like answering "no" to this question in public — like admitting the original decision, and the people who made it, were wrong. That is why the debate gets so charged, and why it so rarely reaches a clear-eyed conclusion.
The second question is "is it working?" — an operational question about whether the system runs and gets used. The third is the only one that should drive the decision: "is it worth continuing to fund from here?"
These come apart in ways that surprise people. A system can be genuinely poor and still worth keeping one more year, because the cost and disruption of replacing it right now outweigh the remaining pain. And a perfectly decent system can be worth cutting, because something dramatically better has appeared and the switch would pay for itself. The decision is not a verdict on the system's quality or on the original purchase. It is a cold comparison of futures — and it can only be made well once it is separated from the two questions that make it emotional.
Someone Has to Be Able to Call It
Even when the analysis is clear, the call rarely gets made, because cutting a system requires someone with both the standing and the willingness to say the expensive thing out loud. Left to a group, it almost never happens: the people who championed the purchase are in the room, the admission feels like blame, and the safest individual move is always to let it limp along for another year on someone else's watch. Inertia is nobody's fault, which is exactly why it wins.
This is where it matters that the system has a genuine owner — one person accountable for its value rather than its uptime, close enough to the business to compare futures honestly and senior enough to be heard when the answer is unwelcome. That person's job includes the part nobody wants: naming the moment the numbers stop justifying the spend, making the forward-looking case free of what was sunk, and separating "was this a mistake?" from "what should we do now?" so the second question can actually be answered. An owner who can only defend the system is not really an owner; the willingness to recommend cutting it is part of the role, and usually the most valuable part.
A Final Thought
Knowing whether a system delivers value, which is where this began, is only half of the discipline. The other half is being willing to act when the answer is no — and that is far harder, because acting means making a loss visible and telling the truth about a decision that has champions and history behind it. Measurement without the willingness to act on a bad result is just expensive record-keeping. The two only work together.
Getting it right does not mean being quick to abandon systems; most are worth keeping and improving, and constant reversal is its own kind of waste. It means making the continue-or-cut decision on the only terms that are valid — future costs against future value, with what you have already spent left firmly out of it. Ask what you would do if you were arriving today, with this system in front of you and this bill ahead of you, owing nothing to the past. Whatever the answer to that question is, is the right answer.
The money already spent has no vote, however loudly it argues for one.
You cannot get back what a failing system has already cost you. You can only decide whether to keep paying it to fail. Those are different decisions, and only one of them is still yours to make.
The money already spent is gone whether you continue or stop. It is not a reason to keep going — it is only the reason it feels impossible to stop.