- Product Tapas
- Posts
- Why 77% of PMs don't know what they're doing, AI Job Search Strategies, Impact-First Product Teams
Why 77% of PMs don't know what they're doing, AI Job Search Strategies, Impact-First Product Teams
Plus: The Age of Everything Clones, Loneliness as a Growth Market, Apple's Secrets Spill

We track Product so you don't have to. Top Podcasts summarised, the latest AI tools, plus research and news in a 5 min digest.
Hey Product Fans!
Welcome to this week’s 🌮 Product Tapas.
New here? We're your shortcut to staying sharp. Essential stories, practical tools, real insights.
For the best reading experience, check our web or app version and sign up for future editions here.
This week’s fresh picks 🥗
Everything can be cloned in minutes (competitive moats are dead), 20 million people are paying for AI friendship (loneliness economy is booming), Apple leaks its entire roadmap by accident (secrecy is impossible), and 77% of PMs admit they have no idea what they're doing with AI (finally, some honesty).
📰 Not Boring → The clone economy arrives, AI loneliness hits 20M users, Apple's accidental roadmap reveal
⌚️ Productivity Tapas → AI video creation, analytics automation, usage metering made simple
🍔 Blog Bites → AI-powered personalisation examples, strategic investment reality checks, AI job search tactics
🎙️ Pod Shots → Impact First Product Teams: Escaping the low-impact death spiral and the CEO question that changes everything
Let's go 🚀
📰 Not boring
The Intelligence Race
GPT-5's rollout fell flat for consumers, but it was OpenAI's stealth enterprise play. Startups like Cursor, Vercel, and Factory have already made GPT-5 the default model due to faster setup, better results on complex tasks, and lower price. Less popular with customers but more profitable overall. Interdasting…
Life post GPT-5: Sam Altman hints at a future beyond a single universal product, with Instacart's Fiji Simo leading the launch of multiple consumer apps outside of ChatGPT and whilst we’re on what’s next: Sam Altman on GPT-6: ‘People want memory’ (no shit Sherlock)
Gemini now learns from your past conversations over time
Google's new open source model can run on your smartphone whilst barely impacting battery. Edge computing eh?
But whilst we’re in the middle of all this hype, worth noting GenAI FOMO has spurred businesses to light nearly $40 billion on fire. With next to ZERO return
The Attention Economy
Threads now has 400 million monthly active users (same as Twitter/X), obviously great for ads but even more so for data. Meanwhile X's declining Android app installs are hurting subscription revenue. The network effect giveth and taketh away
Substack writers can now direct US readers to cheaper web-based subscriptions on iOS, joining Spotify, Patreon, and Amazon Kindle. Apple's 30% tax slowly crumbles
Disruption Roulette
Firecrawl announced an open source AI web app builder that can transform any web URL into a working editable clone. If your competitive advantage can be cloned in minutes, you don't have one. Network effects and brand loyalty seem the logical refuge, but as agents take over customer interactions, how do you even build those? 🤔
Grammarly gets a design overhaul and multiple AI features hot off its acquisition of productivity startup Coda
Character AI has over 20 million MAU for its LLM-based virtual friends. Loneliness is a growth market it seems
Everything Else
77% of product managers expressed uncertainty regarding responsible development of new AI features. Everyone's building, nobody knows what they're doing
A mind-reading brain implant that turns imagined words into speech. Unsure what is more ridiculous, A) we can do this B) the lack of shock I have we can do this
Everything announced at Made by Google, including Pixel 10, Pixel Watch 4, and Pixel Buds 2a
Every Apple Secret That Leaked Wednesday - Apple accidentally exposed hardware identifiers revealing HomePod mini with S-series chip, Apple TV with A17, Studio Display 2, iPad mini with A19 Pro, Vision Pro with M5, plus a Ring-style security camera and tabletop robot. Seems nothing remains secret any more!
Duolingo's roller-coaster week shows how AI model upgrades can threaten existing products overnight. Progress is a double-edged sword
ChatGPT's mobile app has generated $2B to date, earning $2.91 per install. The money printer goes brr
A16z on the future of e-commerce: Is Google screwed? Maybe. But not how you think. On a similar topic, ChatGPT is sending less traffic to websites – down 52% in a month
Spotify's latest feature lets you add your own transitions to playlists.
Google pays $30M to settle lawsuit over children's YouTube data
AirPods Pro 2 seem set for another surprise upgrade at iPhone 17 event - spoiler alert: it’s live translation
Wyoming's 'Frontier' Stablecoin Debuts on Ethereum, Solana and Avalanche. And whilst we're on good old crypto, Coinbase is betting on the rise of AI payments fuelled by digital dollars. Pretty sure I talked about the rise of AI automated crypto stablecoin payments about a year and a half ago….
Acrobat Studio rolled out by Adobe to help you manage multiple PDFs
Listen to your documents using Gemini in Google Docs
TikTok's latest feature lets college students find and connect with classmates.
Finally…here's a great explanation of MCP with some nice visuals too
Better inputs. Sharper outputs. Download the guide to premium AI.
Building or refining generative AI models? This guide shows why scraped data falls short—and what to use instead. Learn how real-world behavior signals, clustering, semantic scoring, and visual diversity improve output. Plus, see how Shutterstock’s licensed data and services reduce risk and boost performance. Train smarter, faster, and more responsibly.
⌚️ Productivity Tapas: Time-Saving Tools & GPTs
Reeroll: Create videos by chatting with AI. AI video editor for creating short, engaging videos in minutes. Choose a template, add your assets and links, and Reeroll will customize it for you. No complex editors, just a simple chat that turns ideas into ready-to-publish videos.
Airbook AI: “Cursor for Analytics” Airbook AI connects to your CRMs, marketing, support, and product databases - understands your schema, auto-writes SQL, joins across sources, creates charts, builds dashboards, and activates workflows - In one place, on command.
Autumn: Price, meter, and control AI usage in just three API calls. Built on Stripe, it handles subscriptions, metering, and access from one place—no webhooks, no backend glue. Perfect for early-stage teams shipping LLMs and image models
Remember. Product Tapas subscribers get our complete toolkit - 460+ personally tailored, time-saving tools for PMs and founders. Your shortcut to efficiency and what's hot in product management 🔥
Check the link here to access.
🍔 Blog Bites - Essential Reads for Product Teams

UX Design: Five Brilliant Micro-Interactions That Show AI-Powered Personalisation in Action
Peter Ramsay is back with another five standout examples of how leading platforms use AI, dynamic content and more to create more personalised user experiences. Each example demonstrates thoughtful design that anticipates user needs and adapts interfaces in real-time. Jump to a great onboarding example here.
💡 "Design is thinking. It's anticipating intent. It's understanding nuance. It's experimenting."
Key Takeaways
• AI-Powered Onboarding: Slite crawls user domains to create realistic, personalised questions within onboarding slides; Dynamic content makes abstract features feel immediately relevant and useful
• Contextual Personalisation: Devin uses company domains in email invitation placeholders for seamless team setup; Headway asks for preferred notification times during onboarding, then applies those preferences automatically
• Interactive Comparison Tools: Vercel's hover states show exactly what usage limits would change when upgrading plans; real-time feedback helps users make informed decisions without leaving the current page
• Smart Search Functionality: BookBeat allows multiple category selection whilst maintaining separate suggestion breakdowns; avoids the common "OR" merge that dilutes search relevance across categories
Strategy: Meta's AI Manifesto and the $100 Billion Reality Check
Not an article per se this one, just a good take from Benedict Evans re Mark Zuckerberg's strategic pivot from metaverse to AI where he covers how Meta's founder-led advantage enables radical shifts that professional CEOs couldn't execute. Evans argues that Zuckerberg's pattern of strategic copying reflects a deeper rigour in surfing user behaviour changes, comparing Meta's approach to Bill Gates' Microsoft rather than Google or Apple.
💡 "Social media is pop culture, and it's protean, always changing, and Mark Zuckerberg is extremely good at surfing user behaviour - at keeping up with how much and how fast everything is changing all of the time."
Key Takeaways:
• Strategic Investment Scale: Meta's 2025 capex planned at $66-72bn, up from $39bn in 2024; Reality Labs has burned through nearly $100bn since the 2014 Oculus acquisition; Company hiring individual AI researchers for tens of millions to regain AI leadership
• Founder Advantage: Zuckerberg's authority and preference shares enable radical pivots impossible for professional CEOs; Previous bold moves include "crazy price" acquisitions of WhatsApp and Instagram; Ability to make long-term bets that would destroy typical corporate leaders
• AI Market Dynamics: Big tech earnings show AI driving enterprise cloud demand beyond current capacity; Google and Meta seeing significantly improved ad revenue through better AI targeting; Microsoft-OpenAI relationship strained over AGI contract terms and access rights
• Industry Transformation Patterns: Generative AI enabling massive creative output scaling (Unilever: 20 to 400 assets per campaign); Robotics reaching inflection point with $6000 humanoid devices, though intelligence remains limited; - Autonomous vehicles progressing in constrained environments (Waymo, Aurora's Texas freight runs)
• Market Positioning: Meta's approach compared to Microsoft's historical rigour rather than Google/Apple innovation models; OpenAI reaching $12bn annualised revenue with 700m weekly active users; Anthropic raising at $170bn valuation versus OpenAI's $300bn
Career Strategy: How to Land 7 Final Round Interviews Using AI-Powered Job Search
Not sure whether this is going to step change your job hunt, but given how competitive the market is, I figured it's worth sharing as it looks interesting…
Brian Kemler explores how to leverage artificial intelligence to achieve "candidate market fit" and bypass traditional application processes. His strategic approach focuses on using AI to craft targeted narratives, research companies deeply, and connect directly with hiring managers rather than relying on mass applications. Read the full article here.
💡 "Instead of brute-force applying to companies and hoping something sticks, I focused on my candidate market fit and using AI to bring it to life."
FYI it’s paywalled - so you’ll need to sub to get he full article
Key Takeaways:
• Candidate Market Fit Strategy: Find where your experience meets market demand rather than chasing dream roles; Create a clear one-line professional summary that communicates both depth and target level; Work backwards from market needs instead of personal preferences to reduce competition
• AI-Powered Resume Optimisation: Use Claude to critique and refine your resume for specific roles and companies; Cut lengthy multi-purpose resumes down to laser-focused 2-page documents; Include proof of work like blog posts and patents as key differentiators, despite conventional advice
• Strategic Company Selection: Research companies where your specific background creates unfair advantage; Target large, public tech companies with clear business needs matching your expertise; Focus on quality over quantity - select companies strategically rather than applying everywhere
• Network Activation Through LinkedIn: Use targeted searches to find 1st and 2nd degree connections who are hiring managers; Search for "Director of Product" or "Group Product Manager" titles at target companies; Bypass online applications entirely by connecting directly with decision-makers
• AI-Enhanced Company Research: Generate comprehensive company dossiers using ChatGPT and Claude's Deep Research; Convert research documents into podcasts using Google's NotebookLM for better information retention; Focus research on trust, safety, privacy, regulatory, and AI-related company developments
• Interview Preparation with AI: Create company-specific interview dossiers predicting likely questions from each interviewer; Demonstrate deeper business understanding than other candidates through AI-assisted preparation; Prepare thoughtful questions that show more insight than current employees
🎙️ Pod Shots - Bitesized Podcast Summaries
Remember, we've built an ever-growing library of our top podcast summaries. Whether you need a quick refresher, want to preview an episode, or need to get up to speed fast - we've got you covered.
Check it out here
🎯 The Low-Impact Death Spiral: How Product Teams Can Avoid Layoffs by Becoming Impact-First
Matt LeMay is ace and I love his book Impact First Product Teams so was stoked to see him on Lenny's podcast this week. If you haven't got his book, definitely go get it. It's a breath of fresh pragmatic air and every product team should read it. The podcast is also worth a listen in full and a solid summary of what's going on in the book. And this below is a summary of that podcast…
Product managers and teams are getting laid off at unprecedented rates. According to Spotify CEO Daniel Ek, the core problem is clear: "We still have too many people doing work around the work" instead of driving real business impact.
Matt LeMay, longtime product leader and author of "Product Management in Practice," has identified the root cause: the Low-Impact Death Spiral. This insidious pattern starts with small, seemingly safe features that gradually increase complexity without delivering meaningful business results—until the next round of layoffs hits.
In the pod, Matt shared his three-step framework for becoming "impact-first" and introduced the CEO question that changes everything about how product teams think about their work.

Matt LeMay | Lenny's Podcast
🎥 Watch the full episode here
📆 Published: 14 August 2024
🕒 Estimated Reading Time: 4 mins. Time saved: 95 mins! 🔥
🤔 The CEO Question That Changes Everything
Before diving into frameworks, Matt poses a critical question every product manager should ask themselves: "If you were the CEO of this company, would you fully fund your own team?"
"Frankly, most of the people I ask that question to don't know the answer right away," he observes. This hesitation reveals a fundamental disconnect between what product teams think they're delivering and what the business actually needs.
The harsh reality? Companies won't keep writing paycheques for two years just because your OKRs hit 0.6 or 0.7. You need to demonstrate clear, measurable business impact—or risk becoming expendable.
Key Takeaways:
Ask yourself the CEO question honestly - would you fund your own team?
OKR scores don't guarantee job security - business impact does
Most product managers can't immediately answer this question - that's the problem
🌀 Understanding the Low-Impact Death Spiral
The Low-Impact Death Spiral follows a predictable pattern that Matt has observed across countless product teams:
Stage 1: Small, Safe Features
Teams focus on low-risk improvements—minor UI tweaks, small feature additions, cosmetic changes that feel productive but don't move business metrics.
Stage 2: Accumulating Complexity
Like rhinestones weighing down a car, these small features accumulate technical debt and operational overhead without proportional value creation.
Stage 3: Resource Drain
The team becomes increasingly busy maintaining and supporting low-impact features whilst struggling to deliver meaningful business results.
Stage 4: Layoffs
When budget cuts come, leadership can't justify keeping teams that haven't demonstrated clear business value, regardless of how busy they've been.
Matt shares a telling example: "I worked with a product manager who broke out of this cycle by focusing on one clear, impactful goal instead of juggling multiple small improvements. The difference in business results was dramatic."
Key Takeaways:
Small, safe features accumulate into big problems over time
Being busy doesn't equal being valuable to the business
Focus beats fragmentation - one impactful goal trumps multiple small improvements
📈 Step 1: Set Team Goals No More Than One Step Away From Company Goals
Continued here and on the link below 👇️
Hey everyone. As mentioned last week I'm starting to move the Pod Shots to standalone web articles because they're getting a bit long for the newsletter. So to continue reading follow the link below!
Don’t worry - NO sub requirement, just making the newsletter more compact and the Pod Shots more easily searchable.
That’s a wrap.
As always, the journey doesn't end here!
Please share and let us know what you would like to see more or less of so we can continue to improve your Product Tapas. 🚀👋
Alastair 🍽️.
Reply