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!

Welcome to this week’s 🌮 Product Tapas, your go-to for the tastiest bites from the world of Product and Tech!

If you’ve been forwarded this or just stumbled upon it, you’re in for a treat. For the best reading experience, check out the web version and sign up for future editions here.

What’s on the menu this week? 🧑‍🍳

📰 Not Boring – Messaging, AI, and product strategy are all in flux: WhatsApp and TikTok are rolling out new features, OpenAI’s latest models are powerful but not without quirks, and Figma is quietly building for the next era of web design. Meanwhile, Tesla’s troubles deepen, and Bain is already thinking about how SEO will change in an AI agent world. (And if you’re feeling lost in the LLM arms race, you’re not alone—see this week’s Pod Shot for more.)

⌚️ Productivity Tapas – This week’s tools help you track product rankings, transform your clipboard, and keep fake signups at bay. Pro subscribers can access our full database of nearly 400 time-saving tools for product managers and founders 🔥.

🍔 Blog Bites – Tim Herbig reframes OKRs as a product, Peter Ramsay is back with some more UX Bites that delight and clarify, and Paweł Huryn makes the case for pragmatic, flexible product frameworks over textbook dogma.

🎙️ Pod Shots – Benedict Evans and Toni Karen-Brown tackle the “AI identity crisis,” asking how product teams can stand out when every chatbot looks the same and core tech is increasingly commoditised.

Plenty to get stuck into - off we go! 🚀

📰 Not boring

• WhatsApp is rolling out a stack of new updates including document scanning, tappable reactions and transcription for voice notes

• TikTok begins testing Footnotes, a new Community Notes-like feature

• OpenAI in talks to pay about $3 billion to acquire AI coding startup Windsurf

• OpenAI also released o3 last week and it’s pretty damn good (and 04-mini) both combine reasoning and tool use for the 1st time

  • But beware, OpenAI’s new reasoning AI models hallucinate more

• Grok gains a canvas-like tool for creating docs and apps [moats are narrow and shallow in the LLM world it seems… see this week’s podshot for more on this]

• Here’s an interesting “practical guide to building agents” for you

• Tesla’s in dire straights and leak shows that Musk's vision is trembling

• Figma is working on Figma Sites - “Design responsively, use pre-built blocks, and add preset interactions—then launch with a click”

• Bain has an interesting piece on the future of SEO in an AI Agent world. TL;DR, optimise content for LLMs

Smarter Sales Starts with Voice AI

AI SaaS Smartcat boosted answered calls by 24%, reactivated 15% more leads, and cut booking costs by 70%—all by deploying Synthflow Voice AI Agents. Automated calls and let their sales team focus on closing deals, not chasing leads.

This case study breaks down Smartcat’s strategy and results, offering practical guidance for teams like yours looking to scale outbound efforts, reduce costs, and turn more conversations into conversions.

Click below to read their story and get the full guide.

Get the full story

⌚️ Productivity Tapas: Time-Saving Tools & GPTs

• Product Rank: See where top products are ranked by AI models

• TabTabTab: Magic Copy and Paste; your clipboard transforms content to match wherever you’re pasting

• Trueguard: Stop fake signups to your product

Remember, as a Product Tapas Pro subscriber you can access the full time saving tools database for fast approaching 400 time-saving tools relevant for product managers and founders 🔥.

Check the link here to access.

🍔 Blog Bites - Essential Reads for Product Teams

Strategy: Make OKRs Drive Decisions, Not Spreadsheets

Tim Herbig challenges the conventional use of OKRs in product teams, arguing that they should drive real decision-making rather than serve as a box-ticking exercise or reporting tool. He reframes OKRs as a product in themselves, urging teams to focus on their practical value in guiding strategy and daily work. Read the full article here.

💡 "OKRs have become one of the most popular frameworks in product development, with abundant advice on 'doing them right.' But this focus on correctness often overshadows their actual purpose. Let's reframe OKRs by treating them as a product."

UX: Built for Mars UX Bites

Peter Ramsay is back again with another sharp set of UX insights from his Built for Mars site/newsletter. From WeTransfer’s timely upsell prompts to Instagram’s dynamic CTA labels, these small touches clarify, personalise, and smooth the user journey in ways that feel both helpful and delightful.

💡 "The best micro-interactions don’t just delight users—they serve a purpose while adding personality to the product experience."

Product Frameworks: The Pragmatic Reality

In this recent piece, Paweł Huryn offers a pragmatic take on product frameworks, challenging the rigid application of textbook methodologies in favour of adaptable approaches that drive real impact. Hallelujah. Read the full article here.

💡 "Your customers and business couldn't care less about 'the right way to do product.' The only thing that matters is whether your work drives impact."

🎙️ Pod Shots - Bitesized Podcast Summaries

Remember, Product Tapas Pro subscribers get access to an ever growing database of all the top Podcast summaries we’ve ever created. 

Check it out here

🤔 The AI Identity Crisis: Why All Chatbots Look the Same

In a recent conversation, tech analysts Benedict Evans and Toni Karen-Brown explored the puzzling landscape of generative AI, questioning how companies can differentiate when their core products are increasingly commoditised. As AI models rapidly improve, they're all climbing the same mountain from different directions—but for users, the experience remains remarkably similar across platforms.

"I've got a folder on my phone called AI, and there's eight or nine apps in it," Evans noted. "What's the difference between them? Some of them have got an icon that's a line drawing, and some of them got an icon that's like a bluey purple swoosh."

This identity crisis raises fundamental questions about product strategy, user adoption, and the future of AI products. Is OpenAI a technology company or a branding company? And how will any AI company capture lasting value when their competitors offer essentially the same capabilities?

🎙️ Listen to the full episode here

📆 Published: April 5th, 2025

🕒 Estimated Reading Time: 2 mins. Time saved: 25 mins🔥

🧩 The Commoditisation Puzzle

Unlike previous technology shifts where companies could differentiate through clear product strategies—think Amazon vs. eBay or iPhone vs. Android—generative AI presents a unique challenge. The underlying models are becoming commodities, with improvements happening simultaneously across competitors.

"You've got all this fine model building on one side, which is fundamentally commoditised," Evans explained. "And on the other side, you've got people building individual, very specific pieces of vertical software."

This creates a strange middle ground where the consumer-facing products—chatbots like ChatGPT, Claude, and Gemini—struggle to articulate meaningful differences. When a new model launches, users often can't tell why it's better unless they're running specialised benchmarks or logic puzzles.

Evans compared this to a late-night TV bit where people were shown last year's iPhone but told it was the newest model: "Wow, wow, this is much better," they'd say, unable to tell the difference.

Key Takeaways for Product Leaders:

• Technical improvements alone may not drive user preference or loyalty

• Consider how your product can solve specific problems in ways competitors can't

• If working with AI, look beyond model capabilities to user experience and integration points

💼 Corporate Strategies: Different Mountains, Same Summit

The conversation highlighted how major tech companies are approaching AI with different corporate strategies:

1. Meta and Apple believe differentiation should come from features built on top of AI, not the models themselves. This is why Meta makes its models open source—they want AI to be a commodity infrastructure.

2. Amazon aligns with this view since AWS sells generic commodity infrastructure at marginal cost.

3. Microsoft and Google have cloud businesses that would prefer models not become cheap commodities, plus product businesses where AI enables new features.

"They're all building models in different ways and sort of climbing the same mountain from different directions," Evans observed. "But as a user looking at five screenshots, if I hit the brand and you prompted the same, put the same thing in, how would you know?"

Not sure if it’s quite this black and white (for example comparing ChatGPT variants vs. Claude on copy I certainly have a preference. But it’s nuanced for sure.

Key Takeaways for Product Leaders:

• Understand where your company sits in this strategic landscape

• Consider whether you're differentiating on the model itself or what you build with it

• Recognize that corporate strategy may not translate to user-perceived differences

🔄 The Stickiness Challenge

If all AI models offer similar capabilities, how do companies create user loyalty? The conversation suggested two potential paths:

1. Branding and habit - Users stick with what they're familiar with

2. Personalisation - Models that learn from users over time, creating a history that makes switching costly

"We get to a point that these models know more about you, that you've used them sufficiently for long enough that you can ask them stuff that requires knowledge of the other questions you've asked over the last six months or a year," Evans explained.

This personalisation creates stickiness that's not quite a network effect but could evolve into one if models learn what "people like me tend to do."

Key Takeaways for Product Leaders:

• Invest in features that improve with continued use

• Create clear migration paths for user data and preferences

• Consider how your AI tool becomes more valuable the longer someone uses it

🎭 Marketing vs. Technology

Perhaps the most provocative idea discussed was whether AI companies are fundamentally technology companies or marketing companies. Evans suggested that OpenAI might be viewed as "a nice marketing company with a lab attached"—a description he acknowledged was "provocative and cynical and unfair" but raised important questions.

This perspective draws parallels to other commoditised industries where differentiation happens primarily through branding:

"How do you differentiate a McDonald's from a Burger King?" Karen-Brown asked.

"It's advertising where you're not actually trying to communicate something different about the product," Evans replied, comparing it to beer advertising or certain car segments where functional differences are minimal.

Key Takeaways for Product Leaders:

• Don't underestimate the power of brand identity and emotional connection

• Consider how your AI product makes users feel, not just what it helps them do

• Look to other commoditised industries for differentiation strategies

❓ The Regulation Question

They also briefly touched on AI regulation, highlighting the challenge of defining what "AI" even means in a regulatory context.

Evans recounted a conversation with a political journalist who asked about regulating AI to ensure it follows societal values: "I gave my example of configuring telco billing systems. Do you think Brussels needs to pass a law to make sure that the company that's helping a big telco reconfigure its billing systems to incorporate a different pricing plan needs to reflect our democratic values?"

This illustrates the need for more specific conversations about regulation rather than broad generalisations about "AI."

Key Takeaways for Product Leaders:

• Be precise about what aspects of AI might require regulation

• Prepare for regulatory frameworks that may treat different AI applications differently

• Contribute to nuanced policy discussions rather than broad generalisations

💰 The Value Capture Question

A final critical question emerged: How will AI companies make money if their core technology is commoditised?

Evans drew a parallel to telecommunications companies, which have seen data usage increase 100,000x since 2010 while their stock prices remained flat. Just because people use more of something doesn't mean companies can charge more for it.

"5G networks are these incredibly amazing things and they've changed our lives and we all use them all the time, every day. And guess how much money the telcos make from it? Like, f*ck all," Evans noted.

This raises existential questions for AI companies about where value will ultimately be captured in the ecosystem.

Key Takeaways for Product Leaders:

• Consider business models beyond charging for the AI itself

• Look for opportunities to create value through complementary products and services

• Study other industries that went through similar commoditisation cycles

🤖 What's Next for AI Products?

As AI models continue to improve at a rapid pace, the conversation suggests that product differentiation will need to happen at levels beyond the model itself. Companies may need to focus on:

1. Creating compelling user experiences that make AI accessible and useful

2. Building personalisation that increases in value over time

3. Developing strong brand identities that resonate emotionally with users

4. Finding specific verticals or use cases where they can excel

The current state of AI products—where purple sparkles and chatbot interfaces reign supreme—may give way to more specialised and differentiated offerings as the market matures.

What do you think? Are we headed toward an AI landscape dominated by a few major brands, or will we see meaningful product differentiation emerge? How might AI companies avoid the fate of telecommunications providers, who built revolutionary infrastructure but captured little of the value it created?

🎙️ Listen to the full episode here

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 🍽️.