Amazon kept its $220 billion plan as free cash flow turned negative. The question that tests what you give up.
TechCrunch reported on October 29, 2026 that Amazon's third-quarter earnings had missed analysts' expectations for both total revenue and earnings per share, even as Amazon Web Services beat revenue expectations and grew 39% compared to the year-ago quarter. The detail the outlet called "the biggest ouch of the report" came next: free cash flow fell to a negative $42.4 billion for the trailing 12 months, against a positive $14.8 billion for the trailing 12 months ended September 30, 2025. The big spending was driven largely by investments in AI, the company said, and with one quarter left in its fiscal year, Amazon reiterated that it "still expects to end its 2026 with cash capital expenditures of about $220 billion."
Nothing in the report says Amazon cut anything, and it would be a stretch to read a scope decision into it. What the story does show is the plain mechanic underneath every such decision. A commitment was left standing with a deadline in view, and its cost appeared on another line, where anyone could read it. TechCrunch's own comparison gives a sense of the size: Amazon expects to spend almost as much on capital expenditures over the full year as it expects to bring in from sales in the fourth quarter, which it forecast at between $240 billion and $247 billion. Whatever is held fixed pushes its cost onto whatever isn't. Project teams live with the same arithmetic at a smaller scale, usually without an earnings report to force the disclosure, and one interview question exists to find out whether a manager has noticed.
Why they ask it
Describe a time you had to cut scope to hit a date, and how you decided what to cut. The sentence has two halves, and the second carries nearly all the weight. Almost every experienced manager has cut scope. Far fewer can reconstruct why one item went and another stayed.
The interviewer is testing three things. The first is whether a criterion existed before the decision, or whether one was assembled afterwards to make the outcome look reasoned. The second is whether the candidate knew what the date was protecting, because a contractual deadline and a date somebody once put on a slide deserve different treatment. The third is ownership. A cut that someone chose, explained and defended says something different about a manager than a cut that simply happened to the project.
The trap
The most common failure is the rescue story. The candidate describes a slipping project, a hard push and a full delivery on time. It may be true, and it is an answer to a different question. Nothing was cut, so nothing was decided, and the interviewer has learned that this manager's instinct under pressure is to spend the team's evenings.
The second failure is the painless cut. The candidate says the team dropped the nice-to-haves. If nobody wanted the items that went, no judgment was required, and the story has no information in it. A real cut removes something a real person wanted.
The third is the borrowed criterion. Saying the team prioritized by impact names a category without naming a test. Impact on whom, measured how, compared with what? A strong answer states the rule in a form that could have produced a different result, then shows the rule costing something.
Applying STAR-T
Situation. Our billing migration had to finish before the old vendor's contract ended, and missing the date meant paying for another year of a system we were leaving. Midway through, the data team reorganized and we lost the engineers who understood the legacy reconciliation jobs. Keep this short, and include only the facts that made the date real.
Task. I owned delivery, so deciding what would make the cutover was my call, and I needed to make it before the team spent another sprint on work we might abandon. The test I set was accuracy for existing customers: anything that couldn't produce a wrong invoice on cutover day was negotiable. Put the criterion here, before the options appear, so the listener can watch it work.
Action. I checked the date first. I asked procurement to price a contract extension, and the finance lead and I rejected it on cost. Then I ran every remaining item through the test. Reconciliation stayed even though it was the slowest work left, and the self-service invoice portal went, even though sales had already shown it to a prospect. The sales lead objected, so I took both options and their costs to our VP, and then the sales lead and I called the prospect together. The escalation matters as much as the sorting, because it shows the decision was made in the open.
Result. We cut over on the date with no invoicing errors in the first billing cycle, and the portal shipped the following quarter. A real answer would attach its own measured figures here, and report the date of the cut's return if it had one.
Trade-off. The prospect signed later than planned, and support handled invoice requests by hand until the portal arrived. I'd make the same call, though I would have raised the risk to sales earlier. Naming this unprompted is what separates a decision from a tidy recollection of one.
The follow-up that breaks weak answers
A good interviewer rarely stops at the story. The next question is usually some version of this: What did the cut actually cost, and did the work ever come back?
Weak answers fold here, because the deferred work was filed for a later phase and never looked at again. The candidate either hasn't checked or has to admit that the cost was never measured. A strong answer knows what happened next. It can say who absorbed the loss, whether the item was rebuilt, and whether it turned out to matter at all, which is sometimes the most instructive finding. It can also say what would have changed the decision, such as a larger customer waiting on the portal, because a criterion that could never have gone the other way wasn't doing any work.
Amazon's report is a useful standard here, whatever one makes of the spending itself. The commitment and its bill appeared in the same document. An answer to this question should do the same.
Score your answer against the director’s bar
Q: Describe a time you had to cut scope to hit a date, and how you decided what to cut.
Bank the scope cut you can defend line by line, and rehearse it until the cost comes out as plainly as the decision. Try it free →