Why your AI prompts feel generic — and how solo consultants fix it

Ask Claude to draft a section of a strategy memo and what comes back is often on topic and instantly forgettable. The sentences are grammatical and the paragraphs are organized, but nothing in it could only have been written by someone who sat in the meetings, met the client, and has a view on the industry. That's the part the client pays for, so the draft gets rewritten from scratch and AI gets the blame.

The usual cause is the brief, not the model. A generic request gets a generic draft, because that's what was asked for, and the fix is a better-shaped prompt rather than a better model. Generic drafts come about in roughly four ways, and each has a specific fix.

This article is for solo consultants and small agency owners who use Claude or ChatGPT regularly but keep cutting the output from every deliverable. If you're new to AI and want the broader frame for which tasks to give it in the first place, start here instead. The rest of this assumes you've already decided this is a Category 2 task and you're trying to get a usable draft.

The four reasons your AI output is generic

Reason 1: You didn't describe the output in 100 words

Most prompts that produce generic text are vague briefs. The prompt asks for "a section about onboarding" or "an executive summary of this engagement," and AI does what any junior writer would do with that brief: produces something competent, on topic, and unspecific.

The fix is a short brief, written in your head or on paper before you open Claude:

• What is this section's job in the document? Not the topic, the job. Is it the part that builds the case for the recommendation, the part that heads off the obvious objection, or the part that gives the reader the one number they'll quote later?
• What's the one claim? A well-written section advances one claim. A vague one advances three loosely related ones.
• What evidence will I point at? List the facts you actually have. Don't ask AI to make any up; that's a different failure (see Reason 4).
• What's the rhythm? Two paragraphs that open and build, a five-paragraph argument, or one deliberate closing sentence?

It usually takes about ninety seconds. Paste the brief in front of the actual request, and the draft gets sharper for very little extra effort.

Reason 2: You're asking AI to generate instead of expand

This is the failure that ruins consulting deliverables. You ask AI to write a 600-word section from a topic. It doesn't have your data, your interview notes or your judgment, so it invents, and the result is fluent and hollow.

The fix is to reverse the workflow: outline, then bullets, then AI as the prose engine, rather than AI as the author.

You write the bullets. They're your thinking in shorthand: the specific data, the quote from the interview, the recommendation, the evidence. Then AI turns the bullets into paragraphs. The result contains your thinking in fluent prose, in a fraction of the time it would take to write from scratch.

The prompt looks like this:

Turn the following bullets into [section] of [document type]. Output is the
section text only — no headers I didn't include, no commentary, no marketing
voice.

Rules:
- Match the voice of the example below
- Keep all facts and specifics from my bullets. Do not generalize them.
- Do not add facts I didn't include. If a transition needs context I haven't
  given you, leave a [TK note] and let me fill it in.
- Sentences should average 15–20 words. Vary length.
- End on a substantive sentence, not a transition.

Voice example (1–2 paragraphs of my actual writing):
[PASTE]

My bullets:
[PASTE]

Two parts of this prompt do most of the work. The voice example pulls the prose toward yours, and the "do not add facts I didn't include; leave a [TK note]" line stops AI from inventing what it doesn't have.

This is the single most useful prompt for consulting writing, and it's covered in detail in Chapter 6 of the Playbook.

Reason 3: You didn't give it a voice example

Reasons 1 and 2 are about what AI writes. This one is about how. AI defaults to a particular dialect: slightly hedged, fond of three-part lists, announcing its transitions, and ending sections with a summary paragraph. Anyone who reads consulting writing for a living can spot it in the first sentence.

The fix takes ten seconds: paste a sample of your own writing into the prompt and ask AI to match its rhythm. Two paragraphs is enough.

In prompt terms, "match the rhythm" means:

• Sentence length and how much it varies
• Vocabulary (do you say "leverage" as a verb? Probably not, and you want AI to avoid it too)
• Punctuation habits (are you an em-dash person? AI is by default; if you're not, say so)
• How you open and close sections (do you lead with the claim, or let it emerge?)

The sample doesn't have to come from the same document. A LinkedIn post, a paragraph from an old memo, an excerpt from your About page: anything that sounds like you. Keep it in a Claude Project so you don't have to paste it every time.

If you're not sure whether AI matched it, run a separate check on the output: "Tell me which sentences in this section read as AI-written. Don't edit; just identify them." Claude is very good at flagging its own tells when you ask it to look.

Reason 4: You're trusting AI in your blind spot

Sometimes the output isn't generic; it's confidently wrong. Ask AI a question outside your expertise (legal language, technical specifications, regulatory framing) and it produces a fluent, plausible answer that doesn't survive scrutiny. It slips past you because nothing in the language sets off your "this seems off" instinct.

No prompt fixes that, because the task was the wrong one to hand over. AI is fine for work where you'd spot a problem in the draft, and risky where you wouldn't.

The cure is the second question of the three-question screen: can I tell at a glance whether the output is good? If not, do the task yourself, or pay a specialist.

This comes up most in proposals, where it's tempting to have AI draft legal- or financial-sounding boilerplate. Don't. Reuse a clause from a contract that's already been vetted, or ask your lawyer.

Putting the four together: a worked example

Say you're writing the recommendations section of a strategy memo for a B2B SaaS client whose customer-success numbers dropped after a pricing overhaul. You've done the analysis, you have three recommendations, and you need about 600 words.

Bad prompt:

Write the recommendations section of a strategy memo for a SaaS client whose
customer success metrics dropped after a pricing change. Make three
recommendations.

This produces three plausible, generic recommendations that don't match your data and could apply to any SaaS company in the world.

Good prompt:

Turn the following bullets into the "Recommendations" section of a strategy
memo for [Client] whose customer-success NPS dropped from 41 to 28 after the
March pricing overhaul.

Rules:
- Match the voice of the example below.
- Keep all facts and specifics. Do not generalize them.
- Do not add facts I didn't include. Use [TK note] for missing transitions.
- Average 15–20 word sentences.
- End each recommendation with the one concrete first action, not a summary.

Voice example: [paste 2 paragraphs of your last memo]

Bullets:

Recommendation 1: Roll back the auto-upgrade to annual on tier 2.
- Discovered in interviews: 7 of 12 tier-2 churners cited "surprise annual
  charge" as the trigger
- Cost of rollback: roughly $180k ARR exposed (per Finance, April 12 model)
- Time to implement: 2 weeks if Eng prioritizes; otherwise 6
- First action: brief Sarah (CRO) by April 30 with the churn-interview data

Recommendation 2: ...
[and so on]

This produces the section in your voice, tied to your data, with a concrete first action for each recommendation, in minutes rather than an hour and a half.

The bullets take about fifteen minutes. The prompt is reusable, and the output needs a light edit rather than a rewrite.

What this looks like as a habit

"The output is too generic for my work" is usually right about the output and wrong about the cause. The fix isn't a better model or more prompts; it's a sharper brief and writing the bullets first.

It takes a few engagements to become a habit. After that it's the default, and the time savings are easy to see: a section that took an hour and a half becomes a twenty-minute draft and edit.

What to do this week

Pick the next deliverable section you have to write. Before opening Claude, spend five minutes on the brief and the bullets. Then paste the prompt structure from this article, with your voice example, get the draft, and time the edit.

If the result still feels generic, the problem is one of these four, and it's almost always Reason 1 (the brief was vague) or Reason 2 (you asked AI to generate when you should have asked it to expand).

Going deeper

This article distills the workflow from Chapter 6 of The Solo Consultant's AI Playbook, which adds the outline-review prompt, the voice-checker prompt, and a worked example of writing a documented ops playbook section from interview notes. If deliverable drafting is where you're losing the most hours, the chapter is the longer version.

Generic output usually means the context never reached Claude. The Claude Workspace Kit keeps that context as files instead of retyping it: your ICP, offers, rate floor and voice, sitting in a folder Claude reads without being asked, so the specifics show up in the first draft instead of the third.

The frame for which tasks to give AI in the first place (Category 1, 2 or 3) is in the previous article. If you're asking AI to do Category 1 work, the stuff that needs your judgment, no prompt structure will fix the output. That's the filter to apply first.

— Digital Kreative