In partnership with

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!

Merry Christmas. Hope those that celebrate are enjoying some well deserved time off.

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.

What’s cooking this week? 🦃

Amazon rams AI into Kindles with zero opt-outs, ChatGPT hits $3bn faster than Disney+, and Notion discovers 50% of revenue now comes from AI features nobody asked for two years ago. Grok goes to the War Department whilst Gemini runs Boxing Day sales, Cloudflare breaks twice in two weeks, and Meta's AI team implodes after hiring a 28-year-old for $14.5bn.

Meanwhile, 84% of developers use AI but 46% don't trust it. Make it make sense.

Happy Boxing Day - turns out the AI arms race doesn't take Christmas off. Neither can I apparently. Over two years, never missed a newsletter. Goddamn streak has me in a chokehold. 🎄

Let's go 🚀

📰 Not boring

🤖 AI Product Blitz

Amazon's Kindle move shows raw platform power: no author consent, no opt-out, don't care. DoorDash's Zesty reveals existential fear - if you're just logistics, you're infrastructure (although not always a bad thing). But when Google Maps adds AI restaurant discovery, what's the moat? Grok going to the War Department whilst Gemini runs Boxing Day sales suggests we've hit the "race to ubiquity" phase.

🎅 The Money Suddenly Works

  • Notion $600M ARR - 50% revenue from AI products, 62% of Fortune 100 use it

  • ChatGPT mobile $3B - faster than TikTok or streaming apps (31 months), $2.48bn in 2025 alone

  • Cursor fastest-growing vendor - +1,000% YoY, developers are largest AI spenders. Lines of code per developer up 76% (4,450 → 7,839), medium teams jumped 7,005 → 13,227 lines

  • Levels.fyi 2025 salaries - Staff engineers +7.52%, AI/ML now "largest and highest-paid SWE tracks"

ChatGPT hit $3bn in 31 months - nearly twice as fast as Disney+ or TikTok. Companies aren't cutting headcount - they're hiring MORE engineers and shipping faster (Brex data). But Principal engineers saw the only pay decline (-6.58%) despite earning most. Is AI actually replacing that tier?

🎬 Platform Power Plays

Instacart insists their pricing tests "were not dynamic pricing"...yet killed them anyway. Perception beats methodology. Apple's Japan concession and debt collector terms show platform power cuts both ways - regulators force openness whilst developers get squeezed harder.

⚡ The Developer Trust Gap

84% of developers use AI tools but 46% actively distrust them (up from 31% in 2024). Experienced developers are MOST sceptical. Yet code output surged 76%. Figma's "prototypes are PRDs" shift creates new gatekeeping - those who can prototype (designers/PMs) vs those who can't.

🎯 Big Swings & Science Projects

Luma's Ray3: "shoot once, reimagine everything else" - but makes location scouts redundant whilst claiming to preserve "authenticity." Anthropic's vending machine became profitable...then fell for social engineering (onion futures, imposter CEOs). The helpfulness training that makes Claude useful directly conflicts with profit maximisation. Vitalik's "exit" privilege assumes mobility most people don't have.

🎁 Year-End Engagement

Year-end retrospectives create "data awareness" that locks users in. Once you see your year quantified, switching platforms means losing that narrative. Retention disguised as fun. But the shear volume has got a somewhat draining this year for me at least.

❄️ Everything Is Breaking

  • Cloudflare outage - 28% of global HTTP traffic down for 25 min, second outage in two weeks

  • Trouble in Meta AI team - LeCun leaving after 12 years, $14.5bn Alexandr Wang hire calling Zuckerberg's control "suffocating," 600 laid off from Superintelligence Labs

  • Threads communities + badges - Meta's Twitter alternative introduces engagement recognition

  • Benedict Evans agentic UX - "AI, networks and Mechanical Turks" asks if LLMs enable better search without massive user bases (adoption weekly not daily despite revolution claims)

  • AI won't take your job - McKinsey says 50% of work hours automatable but 70% of skills remain valuable (which is it?)

Cloudflare knew the fix but "implementation complexity" prevented it - classic execution gap when you're reliability infrastructure for 28% of the web. Meta's drama: LeCun quit after being placed under 28-year-old Wang (hired for $14.5bn), who calls Zuckerberg's control "suffocating." McKinsey claims 50% of work hours are automatable BUT 70% of human skills remain valuable - can both be true?

Become An AI Expert In Just 5 Minutes

If you’re a decision maker at your company, you need to be on the bleeding edge of, well, everything. But before you go signing up for seminars, conferences, lunch ‘n learns, and all that jazz, just know there’s a far better (and simpler) way: Subscribing to The Deep View.

This daily newsletter condenses everything you need to know about the latest and greatest AI developments into a 5-minute read. Squeeze it into your morning coffee break and before you know it, you’ll be an expert too.

Subscribe right here. It’s totally free, wildly informative, and trusted by 600,000+ readers at Google, Meta, Microsoft, and beyond.

Productivity Tapas: Time-Saving Tools & Workflow Automation

  • ManyPi: Turn any website into an API

  • Chathawk: Another llm agregator or sorts - Ask Once. Compare the Best AI Models. Get Consensus.

  • Pixalytica: AI-based KYC Reports Created Using Facial Recognition and Websites Where the Face Was Detected

    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

Product: Rethinking Team Dynamics Through a Restaurant Lens

John Cutler uses a restaurant operations analogy to tease out lessons product team dynamics (as only he could), highlighting the disconnect often found in software development. He notes that while kitchen staff thrive on immediate feedback, product teams frequently lack this critical loop, leading to inefficiencies. Read the full article here.

💡 "When diners aren’t getting good food, they just lower their expectations and stop coming back."

Key Takeaways

Customer Feedback Loops: Unlike kitchens, product teams often overlook real-time feedback. Restaurant staff gain immediate insights from diners, enabling quick adjustments that software teams lack due to a fragmented feedback culture.

Leadership Dysfunction: Many commenters noted that leadership can exacerbate inefficiencies by prioritising optics over function, leading to a focus on busyness rather than effective output. This disconnect can deter meaningful progress.

Motion vs. Impact: The perception of productivity can be misleading. While restaurant kitchens operate with clear output (food served), software teams may appear busy without tangible results, creating a false sense of achievement.

Unpredictable Customer Behaviour: In contrast to the clear requests of diners, software customers often have evolving needs, complicating the process of delivery and necessitating adaptive solutions from product teams.

The Role of the Entire Team: Successful restaurants rely on every staff member, including servers and hosts, to deliver quality service. Similarly, product success depends on cross-functional collaboration beyond just the product and engineering teams.

John Cutler, The Beautiful Mess

Innovation: Unleashing the Power of AI Prototyping for Product Teams

Brian Balfour discusses the transformative potential of AI prototyping in product development, emphasising that it's not just about speed, but about the capacity to explore multiple ideas before finalising a solution. This methodology allows teams to make better, more informed decisions by thoroughly comparing alternatives. Read the full article here.

💡 "The real value of AI prototyping isn’t that you move faster. It’s that speed gives you capacity to explore more ideas before you commit."

Key Takeaways:

Speed vs. Exploration: While traditional prototyping strategies were limited by time and resources, AI prototyping allows teams to generate multiple solutions in hours, fostering an environment for deeper exploration of ideas rather than rushing to a conclusion.

Change in Approach: Instead of committing to a single solution based on initial insights, AI tools encourage teams to present several prototypes to stakeholders, ensuring that the final decision is based on strategic comparisons rather than assumptions.

Disciplined Exploration: Successful teams resist the temptation to build upon the first prototype that appears adequate, instead systematically exploring at least three divergent solutions to guarantee that they select the most effective path forward.

Improved Stakeholder Feedback: Working prototypes enable stakeholders to interact with tangible designs, leading to more constructive discussions and feedback during the early stages of development, which clarifies expectations before major resources are invested.

Brian Balfour, Reforge

Product: Tailored Solutions for Shared Challenges

Tim Herbig explores the dynamics of addressing the same customer problem across different segments. He argues that distinct solutions tailored to various user needs are essential for successful outcomes. Read the full article here.

💡 "Repeating the outcome forces you to define success criteria for each segment, not to accept misleading averages."

Key Takeaways:

Segment Differentiation: When customer segments share the same problem, it’s vital to map out specific outcomes for each to clarify unique needs and avoid one-size-fits-all solutions. This approach helps teams prioritise functionalities based on segment requirements.

Measuring Success: Distinct success metrics for each segment prevent misleading averages that can obscure individual progress. By repeating outcomes, teams can tailor measurements that reflect the true experience of each user group.

Prioritising Solutions: Understanding that not all segments will require the same solution allows teams to ask informed questions about their priorities and outcomes. This clarity is essential for effective project management and resource allocation.

Forcing Function: Repeating outcomes is not merely procedural; it acts as a catalyst for essential discussions that might otherwise be overlooked. It encourages teams to engage in more profound conversations about measuring impact and tackling challenges.

Tim Herbig

🎙 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

🤖 Product Operations: The Role Transforming How 50% of Companies Build Products

As we wrap up the year, I've decided to take us back to two years ago this week to a great podcast on Product Operations.

Product operations has evolved from almost non-existent to standard practice at half of all scaling tech companies - and Melissa Perri saw it coming. In this prescient November 2023 conversation, Melissa Perri and Denise Tillis, authors of "Product Operations: How Successful Companies Build Better Products at Scale", laid out the three pillars of product ops, why PMs shouldn't fear this role, and how to roll it out successfully. Two years later, their predictions have proven remarkably accurate.

  • 🎙 Listen here

  • 📆 Published: November 16th 2023

  • 🕒 Estimated Reading Time: 9 mins. Time saved: 55+ mins! 🔥

Key insights from the full article:

  • 📈 Explosive growth — Product ops went from almost non-existent to deployed in ~50% of scaling tech companies in about 5 years (Uber, Stripe, OpenAI, Ramp, Deal all have it)

  • The 70/30 problem — Most PMs spend only 30% of time on strategic work; product ops flips this to 70%+ strategy by offloading operational busywork

  • 🏗️ Three pillars — Business & Data Insights (quantitative), Customer & Market Insights (qualitative), Process & Practices (how you build)

  • 🎯 PMs keep the important stuff — Decision-making, strategy, vision, prioritisation, tradeoff conversations, go-to-market—none of this gets outsourced

  • 👥 Start with one person — Don't build a massive team; hire one skilled person focused on your biggest pain point and demonstrate quick wins

  • 🚫 Not project management — Product ops increases speed and quality of decision-making; project managers deliver specific timeboxed projects

  • 💼 Reports to CPO — Product ops should be the right-hand person to the head of product, not scattered across other functions

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

or to participate