Stop treating Claude like a search box
A search engine hands you a link and leaves you to do the work. Claude does the work, when you actually hand it over. That's the difference — and it's why most small business owners are only using a fraction of what it can do for them.
This isn't about replacing your team. It's about removing the bottlenecks that only exist because you haven't handed them off yet.
What to hand Claude every week
These are five real jobs to start with. Each one comes with a prompt you can copy right now and adapt to your business.
Paste in your call notes and Claude writes a full proposal in your voice, ready to send.
Here are my raw notes from a sales call: [paste notes]. Write a client proposal based on this. Include the problem they described, the outcome they want, what we'll deliver, and a clear next step. Use a [tone] voice that sounds like me, not like a template. Do not invent pricing, timelines, or claims that aren't in my notes.
Describe what you need to say and Claude writes it. You review, tweak, and send — two minutes instead of twenty.
I need to email [who] about [what's going on]. Here's what I actually want to say, in my own words: [paste your rough notes or voice-to-text]. Turn this into a clear, professional email that still sounds like me. Keep it short. End with [what I need from them].
Give it your topic, your audience, and your tone. Claude maps out a week of social media posts.
My business: [what you do]. My audience: [who you serve, what they care about]. My tone: [3 words that describe how I sound]. Plan a week of social posts around this topic: [topic]. For each day, give me the platform, the hook, the core point, and a natural call to action. Don't repeat the same angle twice.
Instead of spending an hour reading ten articles, ask Claude to summarize what you need to know and point you to the gaps.
I need to understand [topic] well enough to make a decision about [decision]. Summarize the key things I actually need to know, in plain language. Then tell me what's still unclear or where you're not confident, so I know what to check myself before I act on this.
Before you commit to anything big, run it through a pre-mortem. It tells you what could go wrong before you find out the hard way.
I'm about to decide: [describe the decision]. Imagine it's six months from now and this decision turned out badly. Working backward, give me the most likely reasons why it failed. Then tell me which of those risks I could actually reduce right now, before I commit.
The simple rule: if the output sounds generic, the input was too. Add the context Claude can't guess — your notes, your voice, your actual numbers — before asking for another draft.
Your first week
Pick one job from the five above
Start with whichever one is eating the most of your time this week.
Copy the prompt and fill in your own details
The more real context you give it, the less generic the result.
Review before you send it
Check facts, numbers, and tone. Claude drafts it, you're still the one accountable for it.
Add a second job next week
Once one is routine, hand off the next bottleneck.
- Is every fact and number in the output actually correct?
- Does it sound like you, not like a template?
- Is there a clear next step for the reader?