A term commitment has eighteen unused months left and the team has already chosen to leave — when do you actually cut over?
answer
- is it owed either way
- a cost in both options cancels
- sunk cost dressed as prudence
- carrying a rejected platform is not free
- leverage exists only before signing
basics
~20 sStart from whether the remaining months are owed regardless. If they are, they are not a reason to wait — the real comparison is the cost of carrying a rejected platform for eighteen months against the value of consolidating now, with the stranded amount as a fixed line in both options.
solid answer
~50 sThe remaining commitment is months multiplied by the spend it obliges, and on most terms it is owed whether or not the usage appears. If that is true, it lands in both options and cancels, which turns the argument from *we cannot afford to walk away from it* into a straight comparison: what does it cost to carry a platform you have rejected for eighteen more months? That cost is a split estate — two rotas, two toolchains, two sets of operational knowledge — plus whatever the consolidation was worth per month. The honest middle answer is usually partial: move what the commitment does not cover, keep behind whatever consumes it, and schedule the final cutover against the copy window rather than against the contract date. What would shrink this number is a wind-down or transfer term agreed before signing, which is only available once.
go deeper
Remember that a commitment buys a lower rate by promising spend for a period, so leaving early can leave months you still owe with nothing running behind them.
Compute the stranded amount as remaining months times committed spend, and be able to say whether that obligation survives if usage drops.
Show that a cost appearing in both options cancels, and that the live comparison is the carrying cost of a split estate against the value of consolidating sooner.
Own the framing: name the assumption behind every number, refuse a cutover date reverse-engineered from a contract, and say what the exit terms should have been before signature.
## What a stranded commitment actually is A **term commitment** is a promise to spend at some level for a period, in exchange for a lower rate. It is cheaper because *you* absorbed the risk of not needing the usage. When the decision to leave arrives mid-term, the remaining months become a **stranded commitment**: months still to run, multiplied by the spend the term obliges, with declining usage behind it. The first question is factual, not strategic: **is the remainder owed whether or not you use it?** Terms vary — some are consumption commitments that simply bill the shortfall, some allow the obligation to be applied to other usage on the same account family, a few are only reducible by agreement. Find out which you have, because the answer decides whether the rest of the discussion is a real trade-off or a sunk-cost argument wearing a spreadsheet. ## If the money is owed either way Then it is in both options and it cancels. It does not make staying cheaper; it makes leaving no more expensive than it already was. The remaining decision is the one that was hiding behind it: | Option | What you pay | What you get | |---|---|---| | Cut over as soon as the move is ready | exit lines, plus the stranded months with little usage behind them | one estate sooner; the rewrite team released | | Land the cutover at term end | exit lines, plus eighteen months of carrying a rejected platform | the commitment consumed; a deadline nobody chose | | Partial: move what the commitment does not cover | exit lines spread over longer; a smaller stranded amount | some consolidation now, at the cost of a longer split | Carrying a rejected platform is not free and is routinely priced at zero. It is two on-call rotas, two sets of tooling and access reviews, two places where every new service must be told it does not belong, and a team that has to stay fluent in a platform it is leaving. It is also delay on whatever the merge was supposed to deliver, which in an acquisition is usually the reason the move was funded at all. ## The two traps 1. **The sunk-cost argument.** *We have already paid for eighteen months, so we should use them.* If the money is owed regardless, using it is only worth something when the usage is genuinely useful — running a platform you have decided against, so that a line item is consumed, buys nothing. 2. **The date reverse-engineered from the contract.** Once a term-end date is in the plan, the copy window and the rewrite estimate get squeezed to fit it. That is exactly backwards: the copy window is a measurement, the rewrite is an estimate with a range, and the contract date is arbitrary with respect to both. Let the measured constraint set the date, then decide what the commitment costs against that date. ## Framing the decision for the people who own the budget 1. State the stranded amount as a fixed number with its assumption: months remaining, committed monthly spend, and whether the obligation survives reduced usage. 2. State the carrying cost of the split estate per month, including on-call and the delayed consolidation, as a range. 3. Show the cutover date the copy window and the rewrite actually support, independent of the contract. 4. Compare only the differences between options, so the commitment cancels where it genuinely cancels. 5. Say what evidence would change the recommendation, and when it arrives — usually after the first component port and the first measured copy pass. ## What would have made this smaller The leverage over a commitment exists once, before it is signed. The terms worth asking for are the ones that map onto the exit lines: an ability to wind the commitment down or apply it to the successor estate; a defined exit window in which outbound data is capped or waived; a period of read access after termination so the archive can be copied without a race; and delivery of your data in a documented, widely readable format rather than only through the platform's own interfaces. None of these is available at the moment you want them, which is the point: the cheapest time to price the exit is before signing the commitment, when the number still changes what you sign.
- What terms would you ask for before signing, to make a future exit cheaper?The ones that map onto the exit lines: the ability to wind the commitment down or apply it to a successor estate, a defined exit window with outbound data capped or waived, read access for a period after termination so the archive can be copied without a race, and export of your data in a documented, widely readable form. All of them are negotiable only before signature.
- When is waiting for the term to expire genuinely the right call?When the obligation really is reducible by leaving early and the saving is material, or when the copy window and the rewrite land near the term end anyway so waiting costs almost nothing. It is also right when the carrying cost is small — one team, no separate rota — and the consolidation has no time value. Say which of those holds rather than asserting the delay is prudent.
saying these in an interview costs you the question
- Argues for staying because the commitment is already paid for.
- Assumes unused commitment months are refunded once workloads leave.
- Prices the carrying cost of a split estate at zero.
- Sets the cutover date from the contract rather than the measured copy window.
- Never checks whether the obligation survives reduced usage.