HR & Management

Verified against ChatGPT · 2026-08-09

Prep for a hard feedback conversation about a repeated pattern, not just one bad week

Builds talking points and likely-response handling for a direct, specific feedback conversation about a recurring performance issue, so the manager stays anchored to the pattern and a concrete ask instead of drifting into vague criticism or backing off under pushback.

ChatGPT (GPT-5.1)4 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

Help me prepare for a feedback conversation with a direct report about the specific pattern below. I need talking points I can actually say out loud, not a theoretical framework.

WHO AND THE PATTERN
A senior analyst on my team who has missed three internal deadlines in the last two months, each time without flagging the slip in advance.

SPECIFIC INSTANCES (dates, what happened)
Missed the Q2 forecast deadline by 2 days (no heads up), missed the board deck data pull by 1 day (flagged same-day), missed the pricing analysis by 3 days (no heads up until asked).

WHAT I WANT TO BE TRUE AFTER THIS CONVERSATION
I want them to flag a slipping deadline at least 2 days before it's due, every time, no exceptions, starting now.

HOW THIS PERSON TYPICALLY REACTS TO FEEDBACK
Tends to get defensive fast and lists all the other things they're juggling as justification.

Open the conversation with a direct, one-sentence statement of the pattern, naming at least two of the specific instances by name or date so it's immediately clear this isn't about one bad day — vague opens like "I wanted to check in about how things have been going" give the person room to steer the conversation away from the actual issue before it's even named. Write the core of the conversation as the specific things I actually need to say, including exact phrasing options for the hardest part — naming the pattern without it turning into a character judgment ("this isn't about effort, it's about the last three deadlines specifically"). Given how this person typically reacts to feedback, prepare me for their most likely response and give me a specific way to hold the line on the pattern without escalating or backing down — if they tend to get defensive, give me language that acknowledges what's fair in their pushback while still not letting the core ask get negotiated away; if they tend to go quiet, give me a way to draw out what's actually going on rather than filling the silence myself. End with a concrete, mutually specific next step and a named follow-up date, not an open-ended "let's see how it goes."

WHAT NOT TO DO
Do not write this as a script to be read verbatim — write it as talking points and specific phrases I can adapt in my own voice. Do not soften the opening statement with a compliment sandwich; leading with unrelated praise before the real issue trains the person to brace for bad news every time I start with something positive.

OUTPUT FORMAT
1. Opening statement (1-2 sentences, exact wording).
2. Core talking points (bulleted, with at least one exact phrase for the hardest part to say).
3. Anticipated reaction and how to respond to it.
4. Closing: the specific next step and follow-up date.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

Feedback research (echoing the radical-candor and crucial-conversations literature managers are commonly trained on) consistently identifies the vague, softened opening as the single biggest reason hard conversations fail to land — an opener like "how have things been going" hands the other person the conversational initiative and lets a well-practiced deflector steer toward safer ground before the actual issue is even stated, which is why this prompt forces a named, dated, specific opening line instead. Explicitly banning the compliment sandwich targets a well-known but counterproductive habit: opening with unrelated praise before a criticism trains the recipient to associate any positive opener from this manager with incoming bad news, which corrodes the value of future genuine praise and doesn't actually soften the blow of the criticism itself, just delays it awkwardly. Preparing for the specific, named typical reaction rather than a generic "how to handle pushback" script matters because the correct response genuinely differs by reaction type — a defensive person needs their fair points acknowledged without conceding the core ask, while a person who goes quiet needs space and a direct question rather than a manager who fills silence with more talking, which is the instinctive but wrong move for that situation. Requiring a concrete, dated follow-up rather than an open-ended "let's see how it goes" close addresses the mechanism by which feedback conversations quietly fail to change anything: without a specific next checkpoint, both parties can silently agree the conversation happened and move on without either party being accountable for whether the actual behavior changed.

What you get back

Opening: 'I want to talk about a pattern — you've missed three internal deadlines in the last two months, and two of those three I only found out the day they were due.' Core: 'This isn't about effort, it's specifically about the lack of a heads-up before the deadline slips.' If defensive: 'You're right that you've had a lot on your plate — that's a real, separate conversation about workload. Separately, I need a heads-up at least two days out every time, no exceptions.' Close: 'Let's check in on this specifically next Friday.'

Verified against

ChatGPT GPT-5.1 · 2026-08-09

Changelog

  • 2026-08-09 Initial publish, verified against ChatGPT GPT-5.1.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
Write a job description that matches the actual hiring bar instead of a wish list nobody meetsTurns a hiring manager's notes into a job description built around a true must-have vs. nice-to-have split, so the posting attracts candidates who can actually clear the bar instead of scaring off good ones with an inflated requirements list.ChatGPT (GPT-5.1)2026-08-08Build a candidate scorecard that stops a 4-person panel from scoring the same interview four different waysGenerates a per-candidate scorecard tied to the specific competencies this role was actually interviewed for, so a hiring panel's individual scores are comparable instead of reflecting how generous or strict each interviewer happens to be.ChatGPT (GPT-5.1)2026-08-09Calibrate what a 1 through 4 actually means before four interviewers disagree about it in the debriefProduces the scoring rubric itself — the shared definition of what each competency and each score level means — that a hiring panel reviews together before interviews start, so scorecards filled out later are measuring the same thing.ChatGPT (GPT-5.1)2026-08-10Generate interview questions with the follow-up probes that catch a rehearsed answerWrites a set of role-specific interview questions, each paired with two or three follow-up probes designed to surface real depth or expose a memorized answer, plus a flag on any question that risks drifting into illegal or discriminatory territory.ChatGPT (GPT-5.1)2026-08-11
All HR & Management prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY