Apple named its headwinds before guiding to ‘8 to 9%’ services growth. The question that asks what you saw in week one.
On November 3, 2026, TechCrunch reported on Apple’s Monday earnings call, where chief financial officer Kevan Parekh guided services growth to “between 8 to 9%” for the December quarter and, in the same answer, named what would weigh on it. “This outlook contemplates our best estimates on sequential headwinds, primarily from third-party licensing,” he said, alongside a foreign-exchange effect he put at an “additional half percentage point.” He was similarly plain about the App Store: September-quarter performance “was similar to the prior year,” there was “a record number of paid subscriptions,” and “we did see some headwinds in mobile gaming that we’ve talked about before, and expect to see those continue into the summer quarter.” The vague part of the call was the upside. Asked how AI might “contribute to revenue growth over time,” Parekh said “there’s opportunities across third-party apps and partnerships as well as first-party opportunities,” and, per TechCrunch, did not elaborate. The outlet called the plan “not entirely clear,” noted that Reuters had reported Monday that investors are wondering whether Apple could take a cut of third-party AI agents distributed through the App Store, and said it had asked Apple for more.
Read the call as an exercise in forecasting rather than as news about AI. The figure that will be quoted is the guidance. The part worth studying is the order in which Parekh built it: the headwinds first, each one named, the ones already discussed flagged as already discussed, and only then the range. The gaming slowdown was not new, and he said so. The licensing drag was anticipated, and he said that too. A forecast earns trust not by being right but by showing, in advance, which signals it was built from and which parts are still a guess, which is why the AI answer landed so differently from the guidance. That same discipline is what an interviewer is probing with a question that sounds like an invitation to tell a war story and is really a test of perception: “Walk me through a program that slipped. What did you see in week one that predicted it?”
Why they ask it
Every senior manager has a program that slipped, so the interviewer is not checking whether you have one. They are checking whether you were watching while it happened. Programs rarely slip in the week they are declared late. They slip in week one, when a dependency answers its kickoff email late, when the staffing plan carries a name that has not been confirmed, when the estimate for the hardest component is the roundest number on the page. The second sentence of the question is the real one. It asks for a leading indicator, one you had before the outcome, and it asks you to show that you had it then rather than reconstructing it now. Interviewers at this level also listen for how a candidate treats early signals: as information to be acted on, or as noise to be managed until it becomes undeniable. Parekh naming a gaming headwind before it has finished showing up in the numbers is the posture they want to hear about a program.
The trap
The common failure is hindsight narrated as foresight. “Looking back, the problem was obviously the vendor” tells the interviewer you did not see it at the time, however fluently you describe it now. The second failure is the slip with no signal: a true story in which something broke late and nobody could have known. It answers a different question, and it quietly suggests either that you choose programs where you are not watching or that you prefer not to look. The third is the weather report, in which every item on the risk register is presented as having predicted the slip. A long register predicts nothing. A strong answer names a signal or two, describes what each looked like at the time, before anyone knew it mattered, and says what was done about it, including when the honest answer is nothing. It also resists the tidy ending. The question is about perception, not rescue, and a fully recovered program is not required.
Applying STAR-T
Situation. Set the stakes and the date in two sentences. “I ran a platform migration with a committed date written into a customer renewal. Several internal teams and one external data vendor were on the critical path.”
Task. Own the date and the decision about the date. “My job was to hold the date or move it early enough that the account team could renegotiate before the customer’s planning cycle closed.”
Action. This is where week one lives, and it should take most of your time. “In the first week, the vendor’s integration lead was on the kickoff invite and did not attend, and the sample data they sent did not match the format in the signed spec. Separately, the estimate for the schema conversion was the only line on the plan that had not changed since the proposal. I put the vendor gap in writing to the sponsor that Friday and asked the conversion lead to re-estimate against the real sample. I did not move the date yet, because a missed meeting is a signal, not a slip.” Then say what the signals became. “The re-estimate came back materially higher, and the vendor’s format problem turned out to be a versioning issue on their side that took weeks to resolve.”
Result. “I moved the date once, early, and the account team renegotiated the renewal before the customer finalised their own plan. The program still slipped. It slipped on our schedule rather than the vendor’s, and the customer heard it from us first.”
Trade-off. Name the cost. Moving the date that early cost credibility with a sponsor who had just presented it upward, and the re-estimate delayed the start of build. The strong answer says that cost was spent deliberately, because a slip discovered late would have cost the renewal.
The follow-up that breaks weak answers
“Which signal did you see and dismiss?” A candidate who claims to have acted on everything is not believed, because nobody does. The strong answer names the one they waved off, the unchanged estimate, say, explains the reasoning at the time, and says what they would weigh differently now. A second version, “what do you watch for in week one of your next program,” tests whether the lesson became a practice or stayed a story.
Score your answer against the director’s bar
Q: Walk me through a program that slipped. What did you see in week one that predicted it?
Bank this story with the week-one signal written down before the outcome, and rehearse it until the signal comes first. Try it free →