The resignation letter had been sitting on the HR director’s desk for three weeks. And it was not just any resignation: it came from a developer who single-handedly maintained a system that handled roughly eighty percent of the company’s order processing. He had resigned because he had been waiting for over a year for a salary adjustment that had long been promised to him verbally. For three weeks the letter sat there. Everyone knew about it. Nobody acted.
When I asked the HR director why nobody had responded, she looked at me with that mixture of exhaustion and quiet anger and said: “Because we have to complete the job evaluation before we are even allowed to make a counter-offer. And that has been stuck at board level for four months.”
The developer left. What followed — external replacement, lost knowledge, delayed projects — ended up costing roughly twenty times the raise he had asked for. But the job evaluation had followed the correct process.
The pattern: every step is reasonable, the sum is fatal
We see this pattern in mid-sized companies again and again, and it rarely stems from incompetence. Most HR departments we work with are committed and diligent. The problem is structural: every single step in the chain — job evaluation, budget approval, board sign-off — is individually sensible and formally required. It is the sum of these steps that arrives too late.
A process designed to protect the company from a bad decision ends up producing a far more expensive outcome than the risk it was built to prevent. The safeguard becomes the hazard.
Why no KPI catches it
The cost of this slowness appears in no HR metric. Process compliance? One hundred percent. Time-to-hire, training budget, attrition rate? All within target — until the quarter after a key person walks out. The one number that would have told the real story — time-to-decision for retention-critical cases — is measured almost nowhere.
This is the same blind spot we describe in our work on KPIs vs. metrics in R&D: organizations measure what their processes produce, not what their processes cost.
What to do instead
- Keep a key-person risk register. Which systems, products or customer relationships depend on a single head? If the answer to “who else can do this?” is “nobody”, that person belongs on the list — before a resignation letter appears.
- Create a fast lane for retention-critical decisions. When someone on that list signals dissatisfaction or resigns, the standard process is suspended. A predefined group — typically the department head, HR and one board member — decides within days, not months.
- Put an SLA on internal decisions. A job evaluation that takes four months is not thoroughness, it is an unpriced risk. Measure time-to-decision the way you measure delivery times to customers.
- Price the alternative. Before rejecting or delaying a counter-offer, calculate the replacement scenario honestly: recruiting, onboarding, lost knowledge, delayed projects. In our experience the factor is rarely below ten — in the case above it was twenty.
The question to ask your own organization
Where does a “correctly executed” process end up costing you more than it ever protected? If you cannot answer that within a day, that is usually the finding itself.
This case is taken from our consulting practice; details have been altered to protect the company involved. If you want to find out where process latency is quietly costing you key people, get in touch — the first two hours of consultation are free.
