That’s fine. But it’s also leaving a ton of value on the table.

The difference between a basic scheduled task and one that actually runs part of your business?

The prompt behind it.

A basic prompt gets you a basic result that undercuts Cowork’s real value.

Brian just built a prompt framework that fixes this. It’s a (another prompt that builds you a fully customize prompt. Specifically for your business. You answer a few questions, and Claude generates a production-grade scheduled task with tracking, error handling, historical data, and daily briefings.

He’s giving it away free below. Here’s how it works.

The 3-Mode System That Makes It Work

The secret sauce is structure. Brian’s framework breaks every scheduled task into three modes that run automatically based on where you are in the process.

Mode 1: Initialize

This runs the first time only.

The system builds your entire file structure from scratch, folders, CSVs with proper headers, tracking files. You don’t configure anything manually. The prompt handles it.

Mode 2: Daily Scan

This is the daily workhorse.

After initialization, this mode runs every day. It executes your research, logs findings to a CSV, calculates what changed since the last run, and generates a short briefing with the 3–5 things you actually need to know.

Mode 3: Baseline Backfill

On a brand new system, your CSV is empty. Mode 3 runs once to look back over the past 30 days and build a historical baseline so you’re not starting from zero. After that, Mode 2 takes over.

Pro Tip: Once the task has run successfully a couple of times, you no longer need Mode 1 or Mode 3 (nor do you need the initial instructions to process Mode 1 and Mode 3)

Brian’s Example: Daily Competitor Intelligence

To show the system in action, Brian built a competitor tracking agent that monitors 13 competitors across YouTube and LinkedIn. Every day it:

• Scans for new videos and LinkedIn posts since the last run

• Logs everything to a unified CSV with titles, descriptions, URLs, metrics, and tags

• Checks for milestone signals and topic overlaps with his own content

• Searches for new competitors entering the space and flags them for review

• Generates a daily briefing with high-activity competitors, strategic signals, and quiet ones

The output? A clean spreadsheet that grows over time and a short daily brief he can scan in 30 seconds.

It’s Not Just for Competitor Intel

The framework customizes itself to whatever you need. For example:

• generating a fresh list of sales targets each day

• writing newsletter topic drafts

• running weekly performance reports

You just answer Claude’s questions about your business and it builds the full prompt, all three modes, file structure, error handling, everything.

One Thing to Watch Out For

Scheduled tasks only run if Claude Cowork is actively open on your desktop at the scheduled time. If your laptop is closed or the app isn’t running, it skips that day. Brian recommends running it on a machine that stays on if you need consistent daily execution.

Watch the Full Walkthrough + Grab the Free Prompt

Brian walks through every mode, shows the actual output, and links the full 8-page prompt builder doc for free.

Watch it here:

Talk soon,

Brian + Andrea

P.S. Already using scheduled tasks in Cowork? Reply and tell us what you’re running, we’d love to hear.

Rather than just tell you what it does, Brian decided to test it himself. He put together a full video walkthrough if you want to see it in action.

Below we’ll cover what happened, how to set it up, and what to watch out for.

Plus — if you missed last week’s workshop on building an AI agent, we just posted a link to the recording (bottom of the email).

First, What Actually Changed

Up until now, Claude’s desktop app (Cowork) was confined to a sandbox. It could read your files and write new documents, but it couldn’t touch anything else on your computer.

That wall is gone.

Claude can now:

• see your screen

• move your mouse

• click buttons

• use your open applications

• type into fields

It takes screenshots to understand what it’s looking at, then performs actions based on what it sees.

What Brian Tested

Brian started simple. Folder organization. Research across open tabs. Light website scraping. Nothing mission-critical. On purpose.

What worked: Folder cleanup was smooth. Research tasks, where it pulled info from a few open sources, were solid.

What he avoided: Sensitive data. Platforms that flag bots (LinkedIn). Anything where a mistake would actually matter.

His setup: Before every task, he closed everything except the data he wanted Claude to reference. Because Claude screenshots whatever’s open. If your email is visible, it can see your email.

The Phone Dispatch Trick

This is the part most people are sleeping on.

Brian ran tasks from his iPhone. You pair your phone with your desktop through Claude’s Dispatch feature, send a task, and let it work while you’re away.

You’re at lunch. Your desktop is doing the work. You come back and it’s done. The beta has limits, but the bones of something powerful are here.

How to Set It Up

Desktop: Click your name → Settings → Desktop App → General → scroll to “Computer Use” → Turn On. You may need to enable accessibility and screen recording permissions on your Mac / PC.

iPhone: Tap Dispatch → check “Computer Use” → open the sidebar → hit Dispatch. It pairs your phone with your desktop.

Browser control: Settings → General → Browser Use → Allow All Browser Actions. Read the warnings first.

Brian’s Honest Take

Would he recommend going all-in right now? No. The beta is too new, the safety controls aren’t there yet, and the screenshotting is no joke.

But if you’re going to try it, here’s how to stay safe:

• Use a separate device if you can. Don’t give AI free rein on the machine you use every day.

• Close everything you don’t want seen. Treat it like someone’s looking over your shoulder.

• Keep it low-stakes. Folder cleanup, research, basic scraping. Nothing you’d panic about if it went wrong.

• Watch the scraping. Platforms like LinkedIn could flag your account.

Once the safety controls catch up, this changes how people work. We’re watching it closely and we’ll keep testing so you don’t have to.

Watch the Full Walkthrough

Try it on a spare machine this week and see how it feels.

Watch Brian’s full walkthrough here: Claude Cowork’s New “Computer Use” Feature is Insane

Missed last week’s workshop?

We ran a live workshop last week on how to build your own AI agent in Claude. If you missed it, the full recording is up now.

Stay tuned for upcoming workshops!

Talk soon,

Brian + Andrea

The world’s leading experts exclusively focussed on your: Fitness, Finances, Nutrition, Relationship, Parenting, Business….

Except they’re highly coordinated and fully aware of each other’s advice and direction.

Billionaires don’t even get this type of service.

Brian just built it using Claude’s Cowork feature. He created a full system of AI-powered coaches that know his schedule, his goals, his family situation, and his fitness level.

They reference each other.

They check his calendar.

They propose his ideal week.

And the whole thing runs locally on his machine. No data sent to the cloud.

He walked through the entire build on video and we’re sharing the key steps below so you can start building your own.

How It Works

Step 1: Set up your core context

First, you create a set of simple markdown files that describe who you are — your background, family, weekly schedule, current season of life, and goals.

You don’t have to write these yourself. Claude interviews you one question at a time and builds the files based on your answers. Put in as much or as little as you’re comfortable with. You can always update later.

Step 2: Build your coaches

Each coach is a plugin with its own context files and commands. Brian built a fitness coach, business coach, life coach, health coordinator, relationship coach, and more.

The cool part: once a coach is installed, you just hit / and select a command like “workout today” or “fitness check.” No re-explaining. No long prompts. It already knows your situation.

Brian created the prompts for every coach. It’s a 23-page doc you can grab and customize.

Step 3: Connect your apps

This is where it gets powerful. Under Cowork’s “Customize” menu, you can connect Gmail, Google Calendar, and other apps. That means your coaches can check your schedule for conflicts, send you reminders, and slot activities into your real calendar.

Step 4: Schedule your agents

Want your fitness coach to build tomorrow’s workout every morning at 7am? Just hit /schedule, drop in the prompt, set the frequency, and you’re done. It runs automatically.

Step 5: Activate the head coach

This is the payoff. Brian built a “head coach” that coordinates across all the other coaches. He asked it to propose his ideal week and it built a full local website with weekly targets, day-by-day scheduling, and activity breakdowns pulling from every coach’s expertise.

If he likes the plan, one command pushes it all to Google Calendar.

Why This Matters

This isn’t just a cool demo.

Think about what this looks like for your business.

Replace “life coaches” with the world’s best business advisors including a: CFO, CMO, CRO, CTO, COO…

Whatever advisors you need on your board. Same architecture. Same approach. A coordinated team of AI agents built around your specific business, running locally, checking your real data.

That’s where this is headed.

Watch the Full Build

Brian walks through every step and includes all the prompts you need to build this yourself.

Watch it here →

Reminder: AI Weekly Workshop Tuesday - Come Build an AI Agent with Us!

Speaking of agents, next Tuesday, on our community call, we’re building one together. Step by step, in real time. No slides. No lectures. Just building.

Whether you’re brand new to agents or already tinkering, this is a session you can follow along with.

Not a member yet? Start your free trial here. Look for the call link in the welcome email upon signing up.

Already a free or paying member? Keep an eye out for a link to the Weekly Workshop in your email.

Talk soon,

Brian + Andrea

P.S. If you give this a try, drop us a note and tell us how your board of advisors is working for you. We’d love to hear your feedback.

And you’re probably doing it the slow way.

In March, Google quietly rolled Gemini AI directly into its Workspace apps.

You can now search Drive using plain English, draft documents in your own writing style, generate data in Sheets, and build full slide decks with visuals, all without leaving the apps you already use.

The catch?

Google didn’t exactly make it obvious. Most people have no idea these features exist, let alone which ones are actually worth using.

Brian just dropped a full walkthrough breaking it all down. Here are the highlights.

The 4 Big Wins (and a Few Honest Misses)

1. Google Drive search is finally smart

The new AI search doesn’t just match file names, it reads into your documents. Brian searched “pull any documents related to prompting” and Gemini returned relevant files even when their names had nothing to do with prompting. Each result came with a summary.

Apple and Microsoft have been trying to nail this for years. Google got there first.

Verdict: Use it. AI search is a legit upgrade.

2. Google Docs: decent, not dominant

Gemini can draft content, shorten it, add tables, and reference your writing style from other docs. It overlays changes instead of editing the original, which is a nice safety net.

Verdict: Fine for light editing. Claude is still the best for writing.

3. Google Sheets: promising but limited

Gemini handles dummy data, basic tables, and simple charts. But it can’t work across multiple tabs, which limits anything beyond simple tasks. Brian compared it to Claude in Excel, and Claude pulled ahead with pivot tables, dashboards, and P&L statements.

Verdict: Basic tasks only. Claude in Excel is stronger.

4. Google Slides: the surprise winner

Using “Help Me Visualize” (powered by Nano Banana / Gemini’s image generation), Brian created 10 fully designed slides with real visuals. Not just text and bullets.

Verdict: Great visuals via Nano Banana. You have to use the “Help Me Visualize” option specifically. Regular prompts won’t generate slides.

Tradeoff: visuals aren’t directly editable. Claude in PowerPoint gives you editable slides but without the visuals. Pick based on what matters more.

Watch the Full Walkthrough

Brian demos every feature in real time, the wins, the fails, and the workarounds.

Watch it here →

Don’t Miss This: Build an AI Agent (Live)

In next week’s live community call we’re going hands-on. We’ll walk through how to build an AI agent from scratch step by step, in real time. No slides. No theory dumps. Just building. For FREE.

Not a member yet? Start your free trial here. Look for the call link in the welcome email upon signing up.

Already a free or paying member? Keep an eye out for a link to the call in your email.

Talk soon,

Brian + Andrea

It works. Really well.

We’re talking pivot tables, P&L statements, chart generation, error fixing, and even pushing charts into PowerPoint. All from a chat window inside Excel.

If you’re using Excel for anything in your business, this tutorial is for you.

Here are some highlights:

The Setup (Takes 2 Minutes)

Head to marketplace.microsoft.com and search “Claude by Anthropic.” Install the add-in, log in with your Claude credentials, and you’re in.

You’ll need a Pro or Max subscription.

Once installed, Claude lives in the top-right corner of Excel.

5 Use Cases That Actually Matter

1. Generate dummy data instantly

Need sample data to test a model or build a template? Just tell Claude what columns and headers you want. It creates the data, names the tab, and auto-formats everything so the columns are properly spaced. Done in seconds.

2. Build pivot tables quickly

Pivot tables are powerful — but setting them up has always been annoying. Claude builds them from your data and even asks which layout you prefer (category by business unit, monthly breakdowns, etc.). If you’ve avoided pivot tables until now, this is your way in.

3. Create charts that surface real insights

Ask Claude to build charts that highlight anomalies in your data, and it will. In the walkthrough, it generated a scatter plot with outlier thresholds (mean + 2 standard deviations), monthly spend comparisons, and category breakdowns. Not just pretty visuals, but actual analysis.

4. Generate a full P&L statement

This one surprised us. Claude pulled from two separate tabs (expenses and revenue), built a monthly profit and loss statement, added gross profit margins, operating expenses, net income, and percentage calculations. Then it created board-ready charts from that data.

And here’s the key: it pushed everything into a 10-slide PowerPoint presentation complete with executive summary, key takeaways, and strategic outlook. All from one prompt.

5. Find and fix formula errors

Misspelled a function? Created a circular reference? Claude hunts them down and fixes them, even if you’re on a completely different tab. For anyone managing complex workbooks, this alone is worth the subscription.

Bonus: Formatting in One Prompt

You can also ask Claude to reformat an entire sheet. In the video, a plain P&L got transformed with alternating row colors, better text styling, and cleaner structure. Before and after — big difference.

Watch the Full Walkthrough

👉 Watch Brian’s walk-through here: Claude in Excel — Install & Use It in 12 Minutes.

If you’re running a business and touching spreadsheets even once a week, this is 12 minutes well spent.

Cheers,

Brian + Andrea

Explore Lawton Learns AI courses for founders.

Watch Brian’s free AI tutorials on YouTube.

without posting content
without running ads
without hiring a bigger team

That’s exactly what a growing number of solo founders are doing.

They’re building lead magnets that run 24/7.

AI tools that solve one specific problem.

• a curated database
• a niche calculator
• a custom GPT trained on useful data

Once they’re live, they keep bringing people in day and night.

And thanks to vibe coding (our current obsession), most of these tools can be built in under two hours.

Here’s the playbook on how to build one.

Step 1: Build one useful micro-tool

Pick something your audience already searches for.

Examples:
• a viral hooks generator
• a LinkedIn post tool
• a niche calculator

One founder built a custom GPT loaded with viral TikTok hooks.

Thousands of creators now use it, and her core product gets mentioned inside the GPT (hello free advertising).

The pattern is simple: find something people want. Build the simplest useful version.

Step 2: Vibe code it

You don’t need to be a developer (we aren’t).

You just need to describe what you want.

AI builds it.

Here are some tools that make this ridiculously easy:

• Cursor or Claude Code

• Antigravity (Google)

• Lovable

• Custom GPT

Pro Tip: Before you start vibe coding, create a Product Requirements Document (PRD). This is as simple as opening Gemini, Claude, ChatGPT and prompting: “Write me a comprehensive PRD for an application with the following requirements. Include all front end and back end components. This will be coded in [Cursor or Claude or Loveable…]

• Function #1

• Function #2…”

Then drop the PRD into your vibe coding tool of choice.

Your first version can take 60–90 minutes.

Not weeks.

Step 3: Let the tool lead to your product

The best lead magnets don’t feel like marketing. They’re genuinely useful tools. Your paid product just becomes the next logical step.

Example:

The free tool solves problem A

Your paid product solves A + B + C

When people like your free tool, they’re way more likely to upgrade to a paid product.

No hard sell required.

Step 4: Put it where your audience already is

Building the tool is half the work. People still need to discover it.

A simple distribution stack of where to share it:

• Product Hunt or directories

• Social media, especially LinkedIn

• Niche communities (Reddit, Skool, Facebook groups)

• Your newsletter

• Your website

• GPT store

You don’t need millions of views. You need the right people seeing it.

Here’s the magic part: once your micro-tool is live, it keeps working.

You build it once. It compounds over time.

Compare that to paid ads (which stop when you stop paying)
or daily content (which never ends).

A simple challenge

Pick one tool idea this week. Spend 60–90 minutes building it. Then share it in three places where your audience hangs out.

That’s it.

You don’t need a huge audience.

You just need one useful tool that solves one real problem.

Let it run.

Then build the next one.

Talk soon,

Brian + Andrea

Explore Lawton Learns AI courses for founders.

Watch Brian’s free AI tutorials on YouTube.

That’s still important. But it’s not the whole game anymore.

The increasing use of LLMs and AI overviews means your site also needs to make sense to:

• ChatGPT

• Claude

• Gemini

• Copilot

Because more people are getting answers from AI tools before they ever click a website.

Which means this question matters a lot now:

When AI gives an answer in your category, does it mention you?

That’s exactly what Brian breaks down in his new video.

And if you want the short version, here are the 5 big things he says matter most right now:

1. Restructure your content so AI can pull from it
AI tools love clear question-and-answer formatting. So instead of vague headings like “Our Services,” use headings your customers would actually ask, like “What plumbing services do we offer in Des Moines?” Then answer that question directly in the first couple of sentences.

2. Add FAQ sections to your key pages
Your homepage, service pages, and about page should all have FAQs with short, direct answers. Not just the basic stuff either. Include the kinds of questions that show your expertise and prove you know your industry.

3. Add structured data to your site
Yes, this is the technical one. But it matters. Structured data helps search engines and AI tools understand what your content actually means. Brian talks through the main types to focus on, like local business schema, FAQ schema, and service schema. (Sounds nerdy. Is useful.)

4. Make your business info match everywhere
Your name, address, phone number, and hours should be consistent across your website, Google Business Profile, Yelp, and anywhere else your business shows up. If the details don’t match, it creates trust issues, and AI tools notice that.

5. Create content that proves real expertise
Generic “we’re the best” copy won’t cut it. AI tools are more likely to pull from content that actually teaches something. Think specific, helpful, experience-based content that shows you know your stuff.

That’s the core playbook.

Watch the full video:

Talk soon,

Brian + Andrea

Explore Lawton Learns AI courses for founders.

Watch Brian’s free AI YouTube tutorials.

How long does it usually take you to build a slide deck?

• An hour?

• Half your afternoon?

• Literal DAYS?

For us, the answer is days, which is why this particular AI use case really caught our attention.

Claude can now build PowerPoints.

And it’s actually pretty great.

Brian walks you through how it works in his latest YouTube video:

You can literally prompt it like this:

“Create a 10-slide presentation on AI workflow automation for founders.”

Claude will:

• Write the slide content
• Structure the presentation
• Pull in charts and visuals
• Even reference your existing slide templates

The really cool part?

It can also connect with Excel, meaning it can pull real data and automatically turn it into charts for your slides.

It’s still in beta, so it’s not perfect.

But it’s already a huge time saver.

If you make presentations for clients, prospects, workshops, or courses, this is worth seeing.

Talk soon,
Brian + Andrea

Explore Lawton Learns AI courses for founders

Watch Brian’s Free AI Youtube tutorials

And most founders don’t even know it.

It’s called AI Mode.

And it turns a normal search into something closer to a mini business analyst. Instead of a pile of links, you can get:

• structured answers

• better research

• clearer next steps

All without leaving Google.

Brian just dropped a short YouTube video showing how it works (and how to use it properly).

Watch it here:

What you’ll learn

AI Mode can help with a lot of founder tasks—strategy, planning, content, research, you name it.

In this video, Brian walks through one super practical use case: competitor research.

You’ll see how to use AI Mode to:

• research competitors without opening 20 tabs

• turn a basic question into a simple analysis (pricing, weaknesses, positioning angles)

• pull insights you can use immediately for marketing + product decisions

The big shift (and the founder mistake)

Traditional Google search gives you links. AI Overviews give you a summary.

AI Mode goes further. It reasons through your question and organizes the answer so you can act faster (and still shows sources so you can verify).

The mistake most people make is using AI Mode like regular Google.

The better move is to treat it like a business analyst and ask for structured outputs (matrices, comparisons, patterns, pros/cons, weak points).

Brian shows exactly what that looks like in the video.

Bonus: two features most people miss

Brian also demos a couple bonus features:

• Uploads (in certain settings): add your own context so results are less generic

• Create images: quick-and-dirty creative direction right inside the AI Mode flow

Watch the video on YouTube to see the full walkthrough (plus exact prompts).

Talk soon,
Brian + Andrea

P.S. If you’re already in Google all day, AI Mode is the easiest upgrade you can make this week.

Whether you’re onboarding for an in-house role or contractors/freelancers, the problem with onboarding is that YOU’RE the delivery system for it.

Every “quick question” turns you into a live help desk, and the day disappears in tiny interruptions. Meanwhile your hire is trying to be proactive, but the only way to learn is to dig through docs that may or may not be current.

The solution?

A ChatGPT Project lets you hand off the constant explaining and give them a self-serve hub that actually reflects how your company works.

Here’s the 4-step process for setting up your onboarding hub in ChatGPT.

Step 1: Create a new Project.

In the sidebar of ChatGPT under “Projects,” select “New project.” Name it something clear like “New Hire Onboarding” or “[Company Name] Knowledge Center.”

*Note: During this step, select the settings icon and choose “Project-only” so people using the project can’t access memories from your other chats.

Step 2: Upload your company materials.

Drop in:

• SOPs and process docs

• Job descriptions

• Company background and mission

• Org charts

• Any onboarding packages you have

All of this becomes the Project’s knowledge base. Your new hire asks questions, the project pulls from these docs to answer.

Step 3: Write a custom system prompt.

In the Project settings, add instructions like:

You are a knowledge center for [Company Name]. You help new [role] hires understand the company, their responsibilities, and our processes. You have access to our SOPs, org charts, job descriptions, and onboarding materials. Answer questions clearly and specifically using the documents provided. If something isn’t covered, say so — don’t guess.

Step 4: Share the Project link.

From inside the project, click “Share” in the upper right corner. Then you can enter your new hire’s email address or copy a link to share the project with them.

They’ll get their own thread inside the Project. Every question they ask pulls from your actual company docs.

Quick tip: start with the essentials. Upload 5–7 key documents first rather than your entire Google Drive. A focused knowledge base gives sharper answers.

Final thoughts

The beauty of this system is that your new hire now has a go-to source for any onboarding questions, freeing up your time for actual work.

If you try this, start small and iterate. Upload the essentials, see what questions your new hire asks most, and then add documents to fill those gaps. You’ll soon have a project that feels like a real onboarding hub and not just a mish-mash of random docs.

Talk soon,

Brian + Andrea

Google dropped Gemini 3.1 Pro last week, and if you haven’t tried it yet, you’re missing out on one of the biggest reasoning upgrades AI has seen in months.

We’re not talking about slightly better answers. We’re talking about Deep Research that actually investigates like a consultant, Gems that clone your brain for repeatable tasks, and Canvas that turns messy notes into shippable assets.

Most founders are still using AI like a Google search. Meanwhile, the tools have evolved into actual business systems.

So today, we’re walking you through the 5 Gemini workflows we use constantly as founders (the ones that save us hours each week).

Let’s dive in:

Use Case 1: Cloning yourself for repeatable tasks (Gems)

As a founder, you shouldn’t have to explain your brand voice or goals every time you start a new chat. Gems allow you to build specialized assistants that already know the context so you can use them for tasks you do again and again. Simply add sample documents reflecting your writing style into the Knowledge section of the Gem. Provide relevant instructions and off you go!

Founder use cases:

• My Marketing Planner Gem (weekly campaign ideas + content calendar)

• My Sales Follow-Up Gem (turn call notes into follow-ups + next steps)

• My COO Gem (turn chaos into a prioritized plan)

The Result: Stop repeating yourself and start getting work out the door faster.

Step-by-Step Instructions for Creating a Gemini Gem

1. Access the Gem Creator

• Open the Gemini web application.

• Locate the left-hand navigation sidebar and select Gems.

• On the next page, select + New Gem.

2. Set Up the Basics

• Name: Give your Gem a clear, identifiable name (e.g., “Python Code Reviewer” or “French Language Tutor”).

• Description (Optional): Add a brief sentence describing what the Gem does. This helps you remember its purpose if you end up creating multiple Gems.

• Choose your Default Tool (Optional): Tell the Gem to default to image generation, video generation, music generation, etc. Otherwise it will choose the best tool for you.

3. Write the Instructions. This is where you define your Gem’s role, goals, and formats.

• Share Your Context: What’s your domain / environment / situation (e.g., I’m a Founder looking to improve my marketing, I’m an accountant looking to update financial reports)

• Define the Role: Tell the Gem who it is. (e.g., “You are an expert copywriter with 20 years of experience in B2B marketing.”)

• Set the Action: Clearly state what you want it to do. (e.g., “Your goal is to help me rewrite technical product descriptions into engaging social media posts.”)

• Provide a Format: Give it a structure that best articulates your needs. (e.g., create a table, an image, prose, a list, code, etc.)

• Establish the Target Audience: This way the Gem knows how to structure the tone of its response. (e.g., “Business professionals, art students, a Chef”)

4. Add Knowledge Items

• Attach your documents, images, spreadsheets, PDFs here to provide comprehensive context for the Gem to reference.

5. Click Save

• You can now run your new Gem and it will appear pinned in your left-hand sidebar.

Use Case 2: Hiring a “Virtual Analyst” for market intel (Deep Research)

Most AI just browses the web. Gemini’s Deep Research (supercharged in the Gemini 3.1 Pro release) actually investigates, and its output is impressive. It breaks big questions into steps, pulls from dozens of sources, and synthesizes a real report.

We like that you can see the AI’s thinking upfront. You can tweak the plan, add specific competitors to the list, or remove irrelevant sections before it spends time digging.

Founder use cases:

• Competitive analysis (pricing, positioning, messaging angles) - Have Gemini find details about your competitors including their: website, products, customers, pricing, primary channels, etc. Based on the results, ask Gemini to create a detailed competitive analysis.

• Market scan (what’s changing, what’s working, what’s noise) - have Gemini find industry reports and competitor product sheets and perform relevant industry analysis.

• Customer research (patterns from reviews, forums, job posts) - have Gemini perform analysis into your customer base (or who you believe your customer base would be). Instruct it to conduct a scan of reddit and other forums, customer reviews etc. Then have it build you a customer persona based on the insights.

The Result: You get a consultant-level report that informs your business in minutes.

Use Case 3: Turning Notes into Shipping (Canvas)

Canvas is Gemini’s side-by-side workspace where your ideas actually become assets (not to be confused with the Canva design website). You can turn a report into an infographic, quiz, simple web page, or even a lightweight app. This is the move when you want your thinking to become something you can ship.

Founder use cases:

1. Investor update / board update
   Deep Research → Canvas → clean visual summary + key charts + “what we’re doing next.”

2. Lead magnet in 30 minutes
   Deep Research → Canvas → Founder’s Checklist web page + downloadable cheat sheet.

Use Case 4: Content machine for the Too Busy founder

If you talk better than you write, this is your secret weapon. Gemini’s massive context window (up to 2 million tokens) means it can process hours of Zoom meetings or voice notes at once.

Founder workflow we love:

Record a messy voice note → upload → get:

· a clean transcript

· 5 key takeaways

· a LinkedIn post draft

· an email draft

· a short script for a video

Basic Prompt:

Here’s an audio transcript. Pull out: 5 punchy insights, 3 hot takes, and 10 short hooks. Then write: a LinkedIn post, a newsletter section, and a 45-second video script.

Use Case 5: High-end visuals without the $$$ (Nano Banana)

Need a social ad, a blog header, or a website hero image? Gemini uses the Nano Banana model to generate state-of-the-art imagery. Unlike older models, it’s incredible at rendering text accurately.

Use it to kill the two biggest founder bottlenecks:

1. I need visuals but I’m not a designer.

2. I have visuals but they don’t look consistent.

Founder use case:

Start with one master image style you love, then keep iterating: same vibe, same colors, same layout rules. That’s how you stop getting random outputs that don’t look like your brand.

Quick Prompt to steal:

Use this image as the style reference. From now on, keep: the font vibe, spacing, color palette, and overall design system. Create 5 variations for:

1. product update, 2) testimonial, 3) tip, 4) promo, 5) social post.

A better next step (so this doesn’t become “cool ideas you never use”)

Pick one workflow from this email and run it in the next 24 hours.

Because the people who get results with AI are the ones who turn one workflow into a weekly habit.

Talk soon,

Brian + Andrea

P.S. Want to go deeper? The free trial of Lawton Learns is open: our AI course + community for entrepreneurs. If you’ve been wanting to build systems, automate tasks, or scale without working more, take a test drive.

Google has a whole set of AI experiments most founders don’t even know exist.

It’s called Google Labs, and it’s basically where Google ships “half-baked” ideas before they become real products.

If you haven’t tried Google Labs, it’s a missed opportunity, because a few of these tools can save you hours this week with almost zero setup.

Below are the ones worth trying first:

1. NotebookLM: Turn Research Overload Into Insight

The Problem: You’re drowning in PDFs, articles, and documents. Reading everything takes hours.

The Solution: Upload everything to NotebookLM. It analyzes all your documents and becomes your personal research assistant.

Key Features:

• Audio Overviews: Turns documents into podcast conversations. Listen during your commute.

• Deep Research: Scours the web and generates comprehensive reports.

• Video Overviews & Infographics: Transforms summaries into visual presentations.

• Gemini is slowly rolling out access (starting in the US) to NotebookLM within Gemini. This means you can have a “conversation” with your Notebook. Super powerful.

2. Opal: Automate Tasks Without Hiring Developers

The Problem: You have repetitive tasks eating your time, but you can’t code and can’t afford developers.

The Solution: Describe what you want in plain English. Opal builds custom AI-powered workflows and mini-apps for you.

Example:

Create a podcast outline generator that searches trending topics, generates episode ideas with talking points, and creates visual concepts—all in minutes. Or build tools to automate research, generate custom reports, or streamline weekly newsletter updates.

3. Pomelli: End Social Media Content Burnout

The Problem: Creating on-brand social content every week is exhausting and time-consuming. You wish you could just upload your brand assets and have all the social graphics instantly created.

The Solution: Pomelli analyzes your website to understand your brand—tone, fonts, colors, style. Then generates complete social campaigns that actually match your voice and branding. Keep the social graphics as-is or use the concepts as inspiration for your own designs.

Example:

Generate a month’s worth of product launch content, Instagram posts, Facebook ads, graphics, in one sitting. All matching your exact brand colors, fonts, and voice.

4. Google Vids: Create Marketing Videos in Seconds

The Problem: Video content drives engagement, but production is expensive and slow.

The Solution: Google Vids is an AI-powered video creation app built into Google Workspace. Start with a prompt or template, and it generates full video presentations with AI voiceovers, stock footage, and custom visuals. You can also generate 8-second video clips using Veo 3 to add dynamic b-roll to your projects.

Example:

Create customer support walkthroughs, employee training videos, or product launch announcements. Use AI avatars and voiceovers to scale your message without recording yourself.

5. ImageFX: Stop Paying for Stock Photos

The Problem: Stock photos are expensive, generic, and never quite right. Custom design is even more expensive.

The Solution: ImageFX (powered by Imagen 3) turns text descriptions into high-quality images in seconds. Type what you want, and it generates four variations instantly. Use ‘expressive chips’ to tweak style, mood, and details without rewriting your prompt.

Example:

Create product mockups, blog header images, social media graphics, and ad visuals—all unique to your brand. No more scrolling through stock photo sites.

Final Thoughts

Start with NotebookLM and Opal—those two alone can save you hours every week.

These are experiments. They’re not perfect, and some might disappear. But the ones that work can give you a serious competitive advantage.

Try one this week. Visit labs.google to explore all 35+ experiments.

Talk soon,

Brian + Andrea

P.S. Reply and let us know which experiment you try. We’d love to hear about the results.

It’s that they use AI consistently.

They don’t try to “do AI” all day.
They don’t automate everything at once.
They build one simple weekly habit.

Here’s an example you can try.

Once a week, block about 45–60 minutes and use AI for just a few focused things.

1. Get clarity.
   Drop in your notes, messages, or loose thoughts from the week and ask AI to summarize what actually matters. What moved things forward. What didn’t.

2. Notice what repeats and stop starting from scratch.
   Pay attention to tasks you explain, write, or think through more than once. When something keeps showing up, use AI to create a custom GPT or Gem for it. Now, when that work comes up again, you’re opening a tool instead of writing a new prompt.

3. Plan content instantly.
   Do a quick brain dump of everything in your head - ideas, opinions, questions, half-formed thoughts. You can type it or dictate it. Then paste it into ChatGPT with a simple CRAFT prompt like this:

Context: I’m a founder planning content for my business.
Role: Act as my content strategist with 25+ years of experience.
Action: Turn this brain dump into a list of content topics I could share over the next few weeks on [channel or platform e.g. LinkedIn, Facebook, newsletter, etc.]. Group them by theme and suggest a few angles for each.
Format: Bullet list, grouped by theme.
Target audience: [Insert your ICP (ideal customer profile) persona, including things like age range, location, profession, etc.]

You’d be surprised how many solid ideas are already there.

4. Offload one small task.
   Draft an email. Clean up notes. Outline a plan. Prep follow-ups. One task off your plate is enough.

That’s it.
No pressure to use AI perfectly.

This kind of weekly rhythm keeps AI useful without feeling overwhelming. And over time, patterns become obvious. Certain tasks repeat. Certain decisions come up again and again. That’s usually when founders naturally start thinking about templates, workflows, or automation.

Consistency beats intensity every time.

Talk soon,
Brian + Andrea

PS: Inside our paid plans, we help founders turn habits like this into repeatable systems so more of this work happens automatically, without you thinking about it.

And the “.2” is misleading.

Because this is not a cute update.

This is “your 3-hour workflow becomes a 1-prompt deliverable.”

Brian just did a Youtube breaking down the update, watch it here:

The 3 highlights that actually matter (for businesses)

1) It’s meaningfully more reliable

30% fewer hallucinations (from 8.8% to 6.2%). Translation: fewer weird mistakes, less babysitting, more trust.

If you’ve ever said, “This is helpful, but I can’t use it for real work”… this is the update that starts changing that.

2) It can run multi-step workflows end-to-end

This is the big shift via the API.

Instead of you doing: Step 1 → Step 2 → Step 3 → “no wait, redo Step 2”…

Now, you can assign the whole workflow and get a finished work product back.

3) The context window is massive

Adding 100,000+ words in a chat means you can include:

• every contract you’ve ever signed

• all of your competitors financial statements

• a year’s worth of customer support tickets

• all of your emails to a specific customer

The 10-minute test (do this today)

Pick one workflow that costs you 2–4 hours every week.

Not “write an email.”

Something heavier, like:

• monthly reporting

• lead qualification + follow-up

• onboarding checklist + client comms

• a customer support escalation process

Then paste this into GPT-5.2 Thinking mode:

“Act like my operations lead.

Your job is to deliver the final work product, not advice.

Ask me only for missing inputs, then produce the finished output in a format I can use today.”

If the output is even 80% right… you just found a new leverage point in your business.

Want all of the details?

Brian recorded a full breakdown on YouTube:

Watch the video here.

Talk soon,

Brian + Andrea

P.S. If you watch the video, steal one workflow, and implement it this week… you’ll feel the difference by next week.

After a tiresome battle with two other providers, we’ve settled on Substack for our ongoing newsletter platform. No action required. Your email has already been moved. All of our prior newsletters are available as well view the New

If you’re anything like us, December is a crazy month.

Client work. Last-minute planning. Holiday parties.
And gift shopping for way too many people.

So today we’ve got two time-saving workflows for busy founders:

1. A quick guide to letting ChatGPT do your holiday shopping (works for clients, coworkers, family, etc).
2. A walkthrough of the top 5 ways you can use Nano Banana to generate high-quality imagery quickly.

Let’s jump in.

1. How to use ChatGPT to do your holiday shopping (fast)

This year, let ChatGPT handle the mental load of holiday shopping ideas.

Here’s how:

Step 1: Use this prompt
“Create a list of 10 gift ideas for someone who is [age], [interests], with a budget of [$X]. Include links, pricing, and 1-sentence explanations.”

Step 2: Ask for categories
“Give me 5 gifts under $25, 5 under $75, and 5 under $150.”

Step 3: Find cheaper alternatives

Paste the link to something you want to buy.

“Find me 5 cheaper alternatives to this item with equal or better reviews. Target or Amazon preferred.”

Step 4: Let AI write the gift message
AI writes better heartfelt lines than most people do at 11 p.m.

Also, if you’re ordering from Etsy, buy directly from ChatGPT:

Ask ChatGPT specifically to conduct a product search on Etsy. Once you select a product, you can buy it directly with the “Buy” Button. Add your payment details (processed with Stripe) and off you go.

Once you make your Etsy order, ChatGPT allows you to track your order. Simply select your profile at the bottom left of the screen. Then select the Orders option. There, you can find and track your Etsy order. Other stores like Skims and Vouri will be available in ChatGPT soon.

2. The 5 ways founders should be using Nano Banana Pro image generation

Lately, we’ve been loving Nano Banana Pro (aka Gemini Advanced) to generate imagery.

It delivers:

Faster graphics.
Amazing product visuals.
Flawless diagrams.
Consistent branding.
Fast ad variations.

Here are some of the best use cases for business owners:

1. Generate no-edit social media graphics

The Pain:
Most founders (including us) use this process: generate image → download → open Canva → add text → resize → curse quietly.

The Upgrade (Gemini Advanced):
It generates the entire post, including text embedded into the image and matching the lighting, shadows, and materials.

Example Prompt:
“A minimalist workspace with a MacBook open. On the screen, clearly display the text: ‘Q4 STRATEGY’. Warm cozy lighting.”

2. Create instant virtual mockups

The Pain:
Hiring a photographer or buying mockups gets expensive fast (and you may not actually get what you’re looking for).

The Upgrade:
Gemini Advanced’s spatial reasoning makes unreal product shots look real.
No weird AI fingers or melting objects.

Perfect for:
→ Launch graphics
→ Ebook covers
→ Product packaging
→ Course thumbnails
→ App screenshots

Example Prompt:
“Professional product photography of a blank white organic skincare bottle sitting on a wet stone in a rainforest. Water droplets on the bottle. Soft, diffuse lighting.”

3. Turn a napkin sketch into a professional diagram

The Pain:
You have a brilliant business model in your head, but you cannot draw it.

The Upgrade:
Gemini Advanced turns your messy concept into a pitch-ready diagram or flowchart.

Perfect for decks, SOPs, onboarding, investor calls.

Example Prompt:
“A clean isometric 3D illustration of a factory conveyor belt. Input: ‘Data’. Output: ‘Gold Coins’. Minimalist blue/white palette.”

4. Create a consistent brand mascot

The Pain:
Old AI models changed your mascot’s face every time. Nightmare.

The Upgrade:
Gemini Advanced remembers your mascot’s core look and can place them in 50 different scenes while keeping the style consistent.

Massive win for anyone who wants cohesive branding.

Example Prompt:
“Generate a character sheet for a friendly 2D vector fox wearing glasses. Three poses: thinking, typing on a laptop, celebrating.”

5. Generate infinite A/B testing for ad creatives

The Pain:
Creating 20 ad variations used to take forever.

The Upgrade:
Tell Gemini Advanced: “Change just one thing,” and it will.

Background color. Facial expression. Lighting. The object they’re holding.

Perfect for quickly testing and iterating.

6. Create quality infographics

The Pain:
AI-generated infographics are often low-quality or the text doesn’t render correctly.

The Upgrade:
Use Gemini Advanced to quickly generate high-quality infographics for social media, reports, websites, etc.
For example:

Try this Prompt:
“Create a professional, search-grounded infographic about [TOPIC]. Use real-time data to ensure factual accuracy.

Visual Layout: [TYPE, e.g., Horizontal Timeline / 3-Column Comparison / Flowchart].

Content: Include the main title ‘[INSERT EXACT TITLE]‘ and exactly [NUMBER] key sections. Each section must have a distinct icon and concise, perfectly spelled text summarizing the point.

Style: [AESTHETIC, e.g., Minimalist Tech / Corporate Blue / Hand-drawn style].”

Pro Tip: First start by asking Gemini to generate an outline of the data points and citations so you can check for accuracy. THEN ask it to generate the infographic.

IMPORTANT NOTE: To access these features you’ll need a paid Google Gemini plan. You won’t see anything called “Nano Banana Pro.”

What you’re actually looking for is:

→ Google One AI Premium Plan
→ Product Name: Gemini Advanced (includes Gemini 3 Pro models)
→ Cost: $19.99/mo

Yes, there’s a free version, but it’s much more limited.

Final Thoughts

Try one of these workflows this week and see how much time you save.

Talk soon,
Brian + Andrea

P.S. Want to go deeper?
We just opened a free trial of Lawton Learns: our AI course + community for entrepreneurs. If you’ve been wanting to build systems, automate tasks, or scale without working more, take a test drive here.

And now we’re ready to spill the beans.

Here’s what happened:

In 2024, our focus was almost exclusively on high-level, 1:1 AI consulting.

But there was a problem.

For every one client we took on, there were 20+ interested entrepreneurs and small business owners who couldn’t afford the 5-figure price tag.

In 2025, we decided to change that.

We committed to building a resource designed specifically for the founders who don’t have huge teams or unlimited budgets.

Today, we’re thrilled to announce the launch of Lawton Learns.

A course and community designed for entrepreneurs and founders who are:

• Drowning in low-value admin work instead of driving growth.

• Sick of AI hype and unsure how to get results that move the needle.

• Trying to maximize productivity with a small team (or no team).

This isn’t just theory.

These are the same systems our consulting clients pay $45K+ to implement.

And here’s the best part: Founder Launch Pricing.

You’re here on Day One. That means you get to lock in the lowest rate we’ll ever offer (Cyber Monday won’t beat this offer).

But you need to move fast: once we hit 100 members, this price goes up permanently.

To make this a complete no-brainer, for a limited time, you can get started for just $1.

But fair warning: The counter is ticking on the 100 founder spots.

We want you to see the value for yourself. You can sample the content, join the community, and start building without committing to the full shebang upfront:

Start your $1 Trial of Lawton Learns.

We’ll see you on the inside,

Brian + Andrea

We’re talking about Sora 2: OpenAI’s video generation platform.

Think of it like joining TikTok in 2018, posting good content regularly...
and grabbing a serious following before it got noisy.

That’s what Sora 2 is right now.
A wide open lane.

The algorithm isn’t great yet, which means good content stands out.

But it will improve soon.

This is your window.

I just dropped a Youtube video breaking it all down (including how to get in): watch it here.

But here’s the TL;DR...

Prompt like a Pro

Sora isn’t about spitting out random video clips.

It rewards intentional creativity.

Here’s how to write better prompts:

• Lead with the look: “gritty street photography,” “surreal dream sequence,” “epic nature film”

• Use camera language: “drone shot, slow pan, close-up on hands”

• Keep action simple: one subject, one movement, one emotion

• Be detailed with lighting and color: “soft golden hour light,” “cool-toned, foggy cityscape”

• At the end of your prompt, add sound cues: “hiss of rain,” “soft jazz playing in the background”

• Mention gear and time of day: “35mm lens, dusk, over-the-shoulder shot”

• Stay under 1,500 characters, no JSON, just plain text.

You’re not just using a tool.

You’re directing a scene.

Want to Grow Fast on Sora? Here’s the Play:

1. Use a real profile pic. Blank profiles look spammy.

2. Comment on other videos. Barely anyone does, so it gets noticed.

3. Follow freely, there’s no penalty for following 100+ people.

4. Engage like it’s Twitter in 2009. Show up, interact, add value.

Sora is a social platform.

The algorithm loves people who act like it.

Bonus: Free Prompt Generator

I built a custom GPT that helps you write cinematic prompts fast.

Just describe the vibe, and it’ll help you script it:

Try the Sora Prompt Generator GPT

If you’re building an audience, launching a brand, or just want to stay on the cutting edge... this is your chance.

To Wrap Up

Don’t wait until it’s polished.

Start now, and build the unfair advantage while everyone else is still watching.

Talk soon,

Brian + Andrea

If you’re not familiar with him, type his name into a Youtube search.

He’s the multi-milllionaire wearing the tank top, baseball hat, and nose strip.

Hormozi sold 2.9 million books in less than 24 hours, setting a new world record.

After reading his first two books, I was first in line for his third. I’ve now read it multiple times.

What’s money model?

It’s a sequence of offers, discounts, promotions, bundles and incentives to solve a your customer’s problems in the right order, so your business earns more revenue... much faster.

Here’s the problem: Hormozi’s books don’t tell me exactly how to build money models for my very unique products and services.

So I built a Custom GPT to create money models for my business (with an 80+ page pdf of notes from his book in the Knowledge section for the Custom GPT to reference).

Guess what? I made one that you can use too!

Try it out: $100 Million Money Model Expert Custom GPT

Once you open the Custom GPT, provide it with information about your: company, products/services, pricing, target customers, geography, offers, discounts, bundles, sales, discounts, etc...

Then ask it to brainstorm money models for your business based on the book.

Select the ideas you like the most and ask it to fully extrapolate those concepts with detailed implementation guidance.

See my YouTube with all of the details.

Enjoy!

Andrea & Brian

P.S. The Custom GPT can always use improvement so please let me know what works and what doesn’t.

If you’ve ever lost a great response in a sea of untitled chats, scrolled endlessly to find one specific idea, or wondered why you can’t just star a good line for later…same.

I recently made a video about this: The ChatGPT Problems Nobody Talks About

And today, we’re sharing a few workarounds you can use right now until OpenAI catches up.

3 ChatGPT Problems (and How to Fix Them)

1. Problem: You can’t easily bookmark or revisit key info

Solution: Use GPT Projects as Smart Folders

Instead of sifting through endless chat logs, start using GPT Projects as intentional containers. Each project acts like a folder where you can group related chats (e.g., “Client A Strategy” or “Product Launch 2024”). Start new chats inside these projects so they stay organized and easy to find. Bonus: each project has memory, so GPT remembers what matters most.

2. Problem: No folders or categories for organizing chats

Solution: Use a simple tagging system in project titles

Folders don’t exist yet, but you can fake it with structured naming. When saving chats or starting new ones, use a consistent prefix like SOCIAL_, CLIENT_, or RESEARCH_ to keep things findable. Example: SOCIAL_HolidayCampaignIdeas or CLIENT_BrandVoiceGPT. This small change makes the search function way less painful and helps you visually scan your workspace faster.

3. Problem: Important chats get buried or lost

Solution: Export high-value chats regularly

If you’ve ever lost a golden thread mid-scroll, you know the pain. Make a regular habit of copying high-value conversations into a Google Doc, Notion page, or spreadsheet. The easiest way to do this is to end the chat with a prompt like:

“Create a clean exportable version of only ChatGPT’s responses in a Google Docs format.”

Or

“Create a clean exportable version of the entire chat in a Notion format.”

Then paste the output to your preferred solution (Google Docs, MS word, Notion, etc).

Pro Tip: Tag or categorize them so you can search by topic later.

Final Thoughts

We love ChatGPT. But let’s not pretend it doesn’t have UX issues.

Until OpenAI gives us folders and better tagging, we’re doing the next best thing: building our own systems around it.

Try one of these fixes this week and let us know what helps.

Talk soon,

Brian & Andrea

An ambitious goal because my prior record was 1,810 impressions. cough. gasp.

Annnnd, I post often... Four to five times a week. Now, I certainly don’t complain about 1,000+ impressions (those are exactly the people I want to engage).

But 1 million... hmmm.

So, my team and I took a couple of classes from “LinkedIn Experts,” put our heads together, and started our journey.

Weeks later, we put out a post about business leadership:

“How to keep top talent” ...with an image of a quote from a leadership book.

I thought it was solid, but nothing earth-shattering. (You know the feeling, right?)

Then BOOM—it blew up beyond my expectations.

One week later we hit half a million impressions with 200+ comments and 5,000+ reactions.

A month later it hit 950,000+ impressions (and brought a ton of followers).

I’d love to tell you it’s because I’m a LinkedIn Guru. Sadly, I’m not.

I just learned a couple of the tricks.

So what’s the truth? A lot of it is the hook (both the words and the image).

That first sentence that grabs someone by the eyeballs and forces them to keep reading ...in concert with an images that stops their scrolling.

Without a strong hook, even the best content flops. (Trust me—I’ve been creating a ton of content, and not all of it lands.)

Which is why I’ve been obsessing over hooks lately.

But writing a “1,000,000 impression” hook is no small hurdle.

So today, I’m sharing a GPT I built to help you nail hooks on Linkedin.

While the GPT can’t generate a winning image (I use Ideogram for that), it can take care of the copy.

The LinkedIn Hook Generator GPT

You know how LinkedIn can feel like a sea of sameness?

“This is how I optimized my workflow.”

“I’m honored to have been nominated...”

“This is why my team is crushing it.”

You need a hook that stops the scroll (or click away) and makes people think, “Wait, what?!”

Here’s an example:

❌ More marketers are using AI to create content.

Blah. Boring. Moving on.

Versus:

✅ “Hot take: if you’re spending more than 2 hours per week creating content you’re doing it wrong.

See the difference?

It creates a moment of curiosity.

To make your life easier, I’m sharing a GPT I built with my favorite hook templates for LinkedIn. (ChatGPT Plus required)

Note: the CONTEXT for this GPT was written for a Digital Marketer looking to grow their presence on LinkedIn. If you want to create your own version (not designed for Digital Marketers), customize the Context section and create your own custom GPT with the remainder of the prompt.

Just type in your topic and audience and these GPTs will generate click-worthy hooks instantly.

Give them a try (full GPT instructions included at the bottom of this email):

LinkedIn Hook Generator GPT

The prompt Instructions:

Here’s how I would approach creating GPT Instructions for a LinkedIn Hook Generator:

CONTEXT: [Provide ChatGPT with relevant information about you and your current situation, industry, environment, goals, etc]

For instance:

CONTEXT: I am digital marketing professional looking to grow my presence on social media, specifically LinkedIn. To grow my following, I need thought-provoking content that grabs the viewer’s attention and keeps them engaged with the entire post. The best way to get my audience’s attention is an exceptionally catchy and eye-grabbing hook.

For the remainder of the instructions, you can simply copy the following:


ROLE: You are an expert LinkedIn copywriter and marketer with more than 2 decades of success generating wildly viral posts. You are well-known for writing eye-catching, click-worthy hooks for personal Linkedin profile posts. You love coming up with wild ideas that are unique and seldom considered. Your tone is spartan, friendly, easy to read and straightforward. Wild hook ideas are strongly preferred. If you do not understand something or need more information, you are comfortable asking for them before continuing with the response.

ACTION:
1. If you are unsure of the context, topic or target audience for the post, please request it before proceeding.
2. Once you have the required inputs, please write unique and wild hooks for a Linkedin post. Your goal is get people curious about the entire post so they’ll continue reading.
3. I have provided examples of the categories of potential hooks, please provide responses for each category.
4. Wild ideas and unique perspectives are highly encouraged.
5. Take your time and think it through before starting.

FORMAT:
Please provide these hooks in list format with at least one hook per category outlined in the example section.

EXAMPLES:
These are your categories with samples for each.

#1: Common belief challenge:
Statement everyone agrees with follow by a controversial statement.

Sample response:
Content is king when it comes to attracting new business. But most people are doing it wrong.

#2: What not to do
[Audience], stop doing [X].
Option A: (Doing this cost me [Y].)
Option B: (payoff statement like: You’ll thank me later/You’ll thank me when you get $1k+ in sales, etc.).

Sample response:
Marketers, stop prioritizing lead gen over brand building.
Doing this likely cost me $10,000+ in potential sales.

#3: How to:
How to Do X

Sample response:
How to optimize your social media profile in just 30 minutes.

#4: Mistake
The number 1 mistake people make with [X] is [Y].
Here’s what to do instead.

Sample:
The number one mistake people make on social media is chasing followers.
Instead, pick a niche that you know well and consistently provide high value to your target audience.

#5: Improve
Improve [X] with one simple change, technique or tactic.

Sample:
Improve your social media content with one simple tactic.

#6: Storytelling
Statement 1 telling a personal story.
Statement 2 that builds authority.

Sample:
I quit my job after getting burned out.
Since then I’ve been tackling the root cause of burnout by automating with AI.

Sample:
95% of marketers are doubling their investment in AI in 2025.
Take the lead by learning to use the best AI Tools in your niche.

#7: The best
I found the best [X] for [Y]
This is a must-try if you [Z].

Sample:
I found the best tool for automating social media outreach.
This is a must-try if you want to grow on social media.

#8: Hot take
Hot take: Followed by a controversial or counter-intuitive statement.

Sample:
Hot take:
Your 2025 marketing hire should be an AI expert.

#9: One simple thing:
I ask all my [Audience] to do this one simple thing.
Statement about how audience feels about it.

Sample:
I ask all my social media clients to do this one simple thing.
They’re not usually happy to do it.

#10: Bold Statement
Bold statement.
Counterpoint.

Sample:
Email marketing is dead in 2024.
But I found a secret hack to revive it.

---End Examples---

Based on the post topic and target audience I provide, reference the provided examples and return at least one hook per category. Again, wild ideas are preferred.