What You Need to Know
A first output, Claude's first answer to a prompt, is a draft. The prompt that set the task is the brief. Feedback is the follow-up you send about the output. Iteration is the cycle of output, specific feedback, better output. It is the normal working rhythm, not a failure state. The Help Center's getting-started advice says it directly: do not hesitate to refine your prompts or ask follow-up questions. People who get the most from Claude budget a round or two of feedback into every task. People who get the least accept a mediocre first attempt or abandon the tool after one.
At a dental practice, the manager asks Claude for a letter reminding patients that their check-up is due. The draft has the right parts but is stiff, too long and full of clinical terms.
Feedback That Names the Gap
The Help Center names feedback as an official remedy for an unhelpful answer: provide follow-up instructions, clarify, or ask for a rewrite. "Make it better and try again" reproduces the problem, because no new information enters the conversation. Suppose the manager writes "Cut it to 200 words, keep the paragraph on fees, and lose the marketing adjectives." That fixes it, because each clause names a gap.
Good feedback does one of three things:
- It adds a missing constraint, such as the audience, the format or the thing to avoid.
- It corrects a misreading.
- It narrows the scope.
All three rest on diagnosis. Work out the gap between the output and the need, then say it plainly. A useful shortcut is to check the output against the five components of the original brief. The gap usually sits where a component was thin or missing. The manager's one-line prompt never named the audience.
Asking for variants is iteration too. The manager can request three versions of the opening paragraph: formal, conversational and urgent. Claude becomes an options generator and the manager becomes the judge. This is often the fastest route to the right register when you know something is off but can't yet name it.
Refine or Re-Brief?
When the output is close, refine. Stay in the conversation and give targeted follow-ups. Sometimes the output shows the original prompt was wrong: wrong task, wrong framing, missing context. Then rewrite the brief and start clean, rather than patch a misunderstanding through six follow-up turns. A long correction trail muddies the context the conversation carries forward. Each patch also fights the original framing instead of fixing it.
The miscall runs in both directions. One is patching a brief that asked for the wrong thing, turn after turn. The other is abandoning a conversation that was one specific fix away from done. If the manager's letter needs only plainer words, one follow-up finishes it.
Three Responses to a Disappointing Draft
A draft that is structurally fine but off target has three possible responses. The vague retry, "not quite, try again", produces the same shortfall, because nothing new has entered the conversation. The targeted follow-up names the tone, the phrases to avoid and the reader the piece speaks to. It produces a usable revision, because each named gap gives the rewrite something to aim at. The manager names a warm tone, no clinical terms, and patients as the reader.
The third response is the re-brief, for when the original one-line prompt was the real problem. An effective prompt contains as much context as possible, and one line holds very little. Re-brief properly, with the style guide attached and the audience named up front. This fixes the current batch and improves every batch that follows.
Keep the Wording That Worked
When two rounds of feedback finally land the output, the feedback that fixed the draft is exactly the instruction the original prompt was missing. Fold it back in, and the reusable prompt starts the next run from the improved baseline rather than from scratch. The manager's next batch starts from a prompt naming the tone, plain words and patient audience.
In the exam
The exam tests the quality of feedback and ranks the three responses. It rewards the targeted follow-up over the vague retry every time. It expects you to recognise when the re-brief is the professional move. Questions use the miscall in both directions. When a task recurs every week, promoting its wording into standing configuration, such as the instructions of a Project, is Domain 5 territory. Domain 1 ends at refining the request in hand and keeping the wording that earned its place.
Why start clean?
Longer chats consume more of your usage limit. When a chat approaches its length limit, the documented advice is to start a new conversation or use a Project. Restarting on a wrong brief is recommended as professional working practice, because a long correction trail makes a muddled starting point. Official documentation does not claim that long conversations degrade output quality.
Exam Traps
The vague retry dressed as diligence
Regenerating, rewording slightly, or asking Claude to try again adds no new information to the conversation, so the next output repeats the shortfall. Feedback only works when it names the gap: the audience, the tone, the length, the constraint that was missing.
Patching a fundamentally wrong brief through many follow-up turns
When the output shows the original prompt asked for the wrong thing, each patch fights the framing instead of fixing it. Rewriting the brief and starting clean is faster and gives every later batch of the same work a better starting point.
Starting a fresh chat when the output was one specific fix away
This is the miscall in the other direction. A restart discards the context and approved material the conversation already holds; when the output is close, a single targeted follow-up in the same conversation is the cheaper, faster, professional move.
Practice Scenario
A retail merchandiser asks Claude for product descriptions for a seasonal homeware catalogue. The first batch is structurally fine but generic and off-voice. The same prompt in a new chat came back just as generic, and she wants the copy fixed in one more round. What is the most effective next prompt?
Build Exercise
Run Two Feedback Cycles
What you'll learn
- How to diagnose the gap between an output and the need
- How to write feedback that names the gap plainly
- How to use variants in named registers to choose a direction
- How to fold successful feedback back into a reusable prompt
- Take a recent Claude output that disappointed you, or run the five-component brief from objective 1.1 and use its first draft
Why: Feedback practice needs a real gap to diagnose; invented examples flatter the exercise.
You should see: A first draft in front of you plus the prompt that produced it.
- Diagnose the gap: name what specifically differs from what you need, such as audience, tone, length, scope or a missing constraint
Why: Diagnosis is the tested skill; feedback can only name a gap you have identified first.
You should see: One or two written sentences naming specific gaps, not a general feeling of wrongness.
- Write feedback that names each gap plainly and run the first revision
Why: Specific feedback gives the revision something to aim at; vague feedback reproduces the problem.
You should see: A revision that is measurably closer on the gaps you named.
- Diagnose again and run a second cycle, asking for variants in named registers if you are choosing between directions
Why: The second cycle builds the rhythm: output, specific feedback, better output, and variants make you the judge rather than the editor.
You should see: A second revision, or a set of labelled variants, close enough to finish by hand.
- Write down each piece of feedback you gave, mark which of it should have been in the original prompt, and fold it in as your reusable version
Why: The feedback that fixed the draft is the instruction the original prompt was missing; folding it in turns a correction into an asset.
You should see: A revised prompt containing the folded-in feedback, saved wherever you keep working notes.
Sources
- Get started with Claude — Anthropic
- My prompt isn't giving me a helpful answer — Anthropic
- How do usage and length limits work? — Anthropic