What's landing in Power Apps on May 4?

Hi ,

Microsoft just made every business app dramatically smarter.

Copilot is now embedded directly inside Power Apps. Not beside it. Inside it.

And on May 4 something even bigger drops for enterprise teams.

Today's prompt steals the exact words your customers use to sell more. Tips and Tricks Thursday covers how AI gets you booked on more podcasts. Then the full breakdown of what Microsoft just shipped.

🔥 Prompt of the Day 🔥

Voice-of-Customer Mining System: Use ChatGPT or Claude

Create one language-stealing research framework.

"Act as a customer research specialist. Create one system for extracting exact customer language from [REVIEW SOURCES] to use in sales copy.

Essential Details:

• Review Sources: [AMAZON/G2/REDDIT/YOUTUBE COMMENTS]

• Product Category: [WHAT YOU SELL]

• Competitor Reviews: [WHOSE TO MINE]

• Volume Needed: [REVIEW COUNT]

• Output Format: [SWIPE FILE STRUCTURE]

• Use Case: [ADS/EMAILS/LANDING PAGES]

Create one mining system including:

• Review scraping workflow

• AI pattern extraction prompts

• Pain point language library

• Desire phrase catalog

• Objection word bank

• Dream outcome quote collection

• Copy swipe file organization

Steal the words customers already use."

Variables:

REVIEW SOURCES: Where your customers leave feedback

WHAT YOU SELL: Your product or service category

WHOSE TO MINE: Which competitors have the most reviews

REVIEW COUNT: How many reviews to analyze

OUTPUT FORMAT: How you want the swipe file organized

USE CASE: Where you'll use the language — ads, emails, or landing pages

Why This Works:

Your customers already wrote your best copy. They just posted it in reviews. AI scrapes the patterns. Pulls the pain points. Catalogs the desire phrases. Organizes the objections. Hands you a swipe file of words that already convert. Stop guessing what resonates. Use what they already said.

✅ Tips and Tricks Thursday ✅

AI Podcast Outreach System

Getting booked on podcasts is one of the best free marketing channels available.

Most people send generic pitches that hosts delete immediately.

AI changes that.

The Problem

You find a podcast. You write a pitch. It sounds like every other pitch the host received that week.

No hook. No relevance. No reason to say yes.

Mass outreach with a template doesn't work. Personalized outreach at scale does. AI makes that possible.

Why Generic Pitches Fail

Podcast hosts know a template when they see one.

If your pitch could have been sent to any show, it reads like it was. Hosts want guests who understand their audience. Generic pitches prove you don't.

The Solution

Use AI to research each podcast before you pitch.

Feed it the show description, recent episode titles, guest history, and audience demographics.

Ask AI to identify the angles that fit that specific show. What topics haven't they covered. Where your expertise fills a gap in their content.

What AI Researches For You

Audience profile — who listens and what they care about.

Topic patterns — what the host covers most and what's missing.

Guest patterns — who they book and what those guests have in common.

Episode hooks — what titles perform best and why.

How To Build The Pitch

Feed the research into AI. Ask it to generate three pitch angles specific to that show.

Pick the strongest one. Ask AI to draft a personalized outreach email that references a specific recent episode.

The host reads it and knows immediately you actually listened.

The Follow-Up

Most bookings happen on the second or third touchpoint.

Use AI to write follow-up messages timed at optimal intervals. Seven days. Fourteen days. Each one adding a new angle or piece of value.

Never just "following up." Always bringing something new.

Track What Works

Log every pitch. Note the angle, the show category, the response rate.

After twenty pitches AI can identify which angles generate the most replies.

Double down on what books. Drop what doesn't.

What To Do

Pick five podcasts in your niche this week.

Feed each one into Claude or ChatGPT with recent episode titles and guest history.

Ask for three custom pitch angles per show.

Draft personalized emails referencing specific episodes.

Send. Follow up. Track results.

Strategic outreach beats mass emailing by a wide margin.

Did You Know?

Demis Hassabis and John Jumper won the 2024 Nobel Prize in Chemistry for AlphaFold — marking one of the first Nobel Prizes ever awarded primarily for an AI-driven scientific discovery and signaling that machine learning has become a first-class tool of fundamental science.

🗞️ Breaking AI News 🗞️

Microsoft Puts Copilot Inside Power Apps — And Agents Are Next

Most enterprise AI lives outside the tools where work gets done.

Microsoft just changed that.

Copilot now sits inside Power Apps. Same app. Same permissions. Same business rules. AI with full context built in.

Agent feed drops May 4. Mark the date.

What Changed Today

Fill out a form and AI converts incoming emails into structured fields automatically. Review before anything saves.

Ask a question in plain English and the app reshapes the view instantly. No filters. No menus. Just type what you need.

Long activity histories summarized on demand. Seconds instead of scrolling.

Microsoft 365 Copilot answers questions grounded not just in the records on screen but across your full business and productivity data through Microsoft Work IQ.

Same security. Same permissions. Same business logic the app already enforces.

App Skills Going Broad

Data entry, exploration, visualization, and summarization — all generally available now.

These skills are also being exposed through the app's MCP server as reusable tools for agents and Copilot running outside the app.

A recruiting app with years of hiring policy can now power an agent that operates under the exact same rules and accesses the exact same records. App intelligence flowing outward. Not just AI flowing in.

What's Coming May 4

The agent feed gives business users a dedicated space to see, review, and guide agent activity as it happens — inside the app, not in a separate monitoring tool.

Makers set the approval threshold. Routine actions run quietly. Higher-stakes actions surface for human sign-off before anything executes.

Insurance claims team example — agent extracts data from incoming emails and populates case forms. Adjusters approve in the feed before anything enters the system. Clean. Controlled. Efficient.

The Bigger Picture

This is Microsoft's answer to the question every enterprise is asking right now.

How do we bring AI into real business processes without rebuilding everything from scratch?

Embed it in the apps people already use. Ground it in the data and rules already in place. Give humans visibility and control where it matters.

That's the answer they just shipped.

For teams on Power Apps — Copilot is live now. One workflow is worth testing this week just to see what changes.

For IT leaders — tenant-level enablement. Configured in clicks. No new infrastructure required.

For anyone watching enterprise AI — the race to own the workflow layer just got more competitive.

Over to You...

Microsoft just made enterprise software dramatically smarter overnight. Are you taking advantage of it?

Reply - I read them all.

To AI where the work actually happens,

Jeff J. Hunter 
Founder, AI Persona Method | TheTip.ai

P.S. Want to turn AI Agents into a consulting offer? Book your AI Certified Consultant strategy 👉 here.

NEW: “AI Money Group” to Learn to Make Money with AI