The Cost of a Vacant Role Measured in Delivery Days

A role has been open for three months. On the finance sheet that shows up as money not spent, which looks like good news. On the delivery side it shows up as dates moving and two people quietly working late, which shows up nowhere.

The fix is to express a vacancy in the same unit as the plan: delivery days lost. If a role sitting empty pushes a committed date, say how many days it pushed it and which commitment it was. A budget conversation with a number on both sides is a different conversation from one with a number on only one side.

Why the argument is currently one sided

Finance has a clean number. Salary, benefits and overhead per month, all of it real and all of it verifiable.

Delivery has a feeling. "We are stretched." "It is slowing us down." "People are covering." None of that survives contact with a spreadsheet, so it loses, and the role stays open another month.

You do not win this by arguing harder. You win it by producing a number of the same quality. It is the same number missing when headcount plans fall apart in the second quarter.

Counting the days

The method is boring on purpose. Do it per vacancy, not for the department.

Step one. Write down what the role was going to do. Not the job description. The actual named work on this quarter's plan that was assigned to that seat, or would have been.

Step two. Split that work into three buckets.

• Work that stopped completely
• Work someone else picked up
• Work nobody picked up and nobody noticed yet

Step three. For the stopped work, count the days the date moved. This is the cleanest number you will get. A committed date was one day and is now another. The difference is your first figure.

Step four. For the picked up work, count the days it delayed the picker's own work. When a senior person absorbs a vacancy, their own commitments slide. Those slides are also delivery days, they are just attributed to the wrong person.

Step five. Note the third bucket without pricing it. Work nobody picked up is usually the maintenance, the documentation and the small improvements. It has a real cost but you cannot count it honestly yet, so do not pretend to. Just list it so it exists on paper.

Add buckets one and two. That is your delivery days figure for the vacancy so far. Update it monthly, because the number grows. Where the vacancy follows a resignation, start from the notice period maths, which sets the earliest date anyone could be useful again.

Two effects that make it worse than it looks

The first is that the cost is not flat. In the first weeks a team absorbs a vacancy fairly well. Later the absorbed work starts colliding with the absorber's own commitments and the rate of loss increases. A vacancy that looked cheap in month one is not the same vacancy in month four.

The second is the overload tax. People covering two jobs make more mistakes, review less carefully and take on less risk. That surfaces later as rework, and rework arrives after the vacancy is filled, so it never gets attributed to the vacancy at all.

Neither of these needs a fabricated multiplier. Just say them out loud in the meeting, in that order, once.

Presenting it so it lands

Take the days figure and attach it to something the finance side already cares about.

• Name the specific commitment that moved, ideally one with a customer or a revenue date attached
• State the days it moved
• State how many of those days are recoverable if the role is filled this month, and how many are already spent
• Give the same figure for next month if nothing changes

That last line is what changes decisions. A cost that has already happened invites blame. A cost that will happen next month invites action. Filling the seat is only half of it, since a probation review that produces a decision is what stops the same role reopening in month four.

Keep the record, do not rebuild it

The reason nobody does this is that reconstructing it afterwards takes a day of digging through two systems. The staffing side sits in HR and the date side sits in the project tool, and joining them is manual work that only happens when someone is already annoyed.

If the plan and the people records live together, the join is already there. That is the case for a single system like SicherOne, which puts project management and HR on the same records so an open seat and the work it was carrying are visible in the same place. Sold per seat, modules separable.

Whatever you use, start the count on the day the role opens, not on the day someone finally asks what it cost. Retroactive numbers are always weaker, and by then the argument has usually already been lost.