Feb 2, 2024 · 57m · news
Will Wu: Top Five Product Lessons from Creating Snapchat "Discover" and "Chat" | E1111 · 20VC with Harry Stebbings
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of 20VC, host Harry Stebbings interviews Will Wu, CPO of Match Group and former Snap executive, exploring his Silicon Valley origins, product design philosophies, hiring frameworks, and the future of AI in consumer technology. Wu shares critical lessons on human-centered design, the delicate balance between artistic instinct and data, and the absolute necessity of building high-empathy, mission-driven team cultures.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Harry holds 21% of the talking time here. How this is scored →
Harry as informed peer 1.5 Guest teaching 1.7 Guest disagreement 0.7 Harry pushing back 0.6
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 46:00 Rejecting UI Irrelevance Thesis
Guest directly rejects host's premise that UI will become completely irrelevant due to AI, firmly arguing that multiple interface paradigms continue to coexist.
Hardest push from Harry ▶ 42:31 Challenging the Mercenary StanceHost explicitly pushes back on guest's rejection of money-motivated candidates, asking whether being in it for the money really matters if the person delivers great work.
Biggest teaching moment ▶ 8:32 Repositioning Snap DiscoverGuest educates host on traffic pattern design, showing how a single UX shift moving Discover one screen left completely rescued the feature's adoption.
Harry holds his own ▶ 35:10 Predicting Fruit Ninja StudioHost demonstrates strong industry knowledge by immediately guessing the creator studio behind Fruit Ninja when guest describes the Australian game developer he acquired.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Welcome and Will Wu's Entry into Tech | 0 | 1 | 0 | 0 | Host opens with standard biographical questions and friendly banter. Guest explains his early fascination with technology and IRC chat rooms, educating the host on what IRC is. | |
| Transitioning from Engineer to Product Designer at Snapchat | 0 | 1 | 0 | 0 | Host prompts guest to tell his Snap origin story. Guest details meeting Evan Spiegel and shifting his self-identity from engineer to product designer. | |
| Working with Evan Spiegel: The Product Visionary | 1 | 2 | 0 | 0 | Host asks what makes Evan Spiegel a visionary and asks for lessons from Snap Discover. Guest explains how shifting Discover one screen over saved the product's traffic. | |
| Building Snap Games and Overcoming Imposter Syndrome | 0 | 1 | 0 | 0 | Host asks about Snap Games. Guest candidly shares his personal experience overcoming public speaking fears, stress, and imposter syndrome during the launch. | |
| Art vs. Science: The Dual Nature of Product Design | 1 | 2 | 0 | 0 | Host asks guest to quantify whether product is art or science. Guest explains his philosophy of art and science intersecting, citing Match Group's ASL team. | |
| Case Studies: Apple Watch vs. Enterprise Software | 2 | 2 | 1 | 1 | Host asks for good and bad product examples. Guest cites Apple Watch for good human-centered design and Concur for bad UI, which host enthusiastically agrees with. | |
| Operationalizing Empathy: How to Uncover User Needs | 1 | 2 | 0 | 0 | Host asks how founders can uncover user needs and whether simplicity is always key. Guest explains starting simple for novel products like the iPhone and adding complexity over time. | |
| Preventing Feature Creep Through Information Architecture | 2 | 2 | 1 | 1 | Host presses on how to prevent feature creep and whether rapid prototyping conflicts with pride in quality. Guest explains rapid prototyping accelerates high-quality releases. | |
| Fostering Creative Brainstorms: Safety, Structure, and Small Groups | 2 | 2 | 0 | 0 | Host asks how product reviews and brainstorms should be run, mentioning Miro boards. Guest clarifies he prefers physical whiteboards, small groups, and psychological safety. | |
| Engineering Creative Chaos: Diverse Teams and Global Inspiration | 1 | 1 | 0 | 0 | Host asks how to introduce creative chaos. Guest advocates for multi-disciplinary teams and global travel inspiration. | |
| Balancing Today's Revenue and Tomorrow's Innovation | 2 | 2 | 0 | 0 | Host asks how to balance short-term revenue with long-term innovation. Guest outlines structural team separation to avoid conflicting incentives. | |
| Evaluating Candidates Through Past Products | 2 | 2 | 0 | 0 | Host asks how guest evaluates product candidates. Guest shares acquiring an Australian game studio and host correctly guesses Fruit Ninja's creators. | |
| The Growth Mindset: Assessing Long-Term Ambition and Passion | 3 | 1 | 2 | 3 | Host pushes back on candidate motivation, suggesting modern product people care heavily about work-life balance and money. Guest insists genuine passion is required. | |
| Standardized Tests vs. Real Work: Interviewing Product Candidates | 2 | 2 | 1 | 1 | Host asks about case studies in interviews and hiring mistakes. Guest likens case studies to standardized tests and prefers evaluating past work directly. | |
| The Danger of Paid Mercenaries: Money vs. Passion in Product | 3 | 1 | 3 | 3 | Host challenges guest on whether being a money-motivated mercenary is actually bad if the employee is competent. Guest strongly rejects the mercenary mindset. | |
| Will Conversational Interfaces Make Traditional UI Irrelevant? | 2 | 3 | 4 | 2 | Host brings up industry claims that AI will render traditional UI irrelevant. Guest explicitly disagrees, explaining how CLI, GUI, and conversational interfaces coexist. | |
| Mission-Driven Success: Losing Users to Love on Match Group | 2 | 2 | 1 | 1 | Host notes that Match Group loses users when successful. Guest agrees, framing app competition around share of user time. | |
| Quickfire Round: Listening to Users and an Elaborate Nerd Setup | 1 | 2 | 0 | 0 | Host starts quickfire round. Guest describes his elaborate work-from-home setup including a custom artisan Leopold keyboard. | |
| Quickfire: Mechanical Watches as Art, Science, and Milestones | 2 | 2 | 0 | 0 | Host asks about watch collections. Guest shares his appreciation for mechanical watches, and host adds insight into Patek Philippe marketing. |
Statements from this episode (35)
Will Wu: Simplicity is paramount when building novel products
“When you are creating a novel product, something that's truly innovative and new, simplicity is paramount towards making it easy to use and easy to grok.”
Will Wu: Biggest hiring mistake is keeping poor culture fits
“Not getting rid of Poor culture fits fast enough.”
Evan Spiegel offered Will Wu a Snapchat job the day they met
“In that moment, he was like, I have the perfect job for you. And later that day, I got a job offer in my email, and it said product designer as a job title.”
Wu: Moving Snapchat Discover next to stories completely changed its fate
“We moved Discover one screen to the left, essentially, to be on the same screen as where you see your friends' stories, And that completely changed the fate of Discover.”
Wu: Some of the best product thinkers come from game design
“I think some of the best product thinkers actually come from game design.”
Wu: Imposter syndrome is a positive force pushing learning and growth
“Imposter syndrome, which is always something that I've felt my entire life, and I still feel to this day, is it can be a good thing. It always used to terrify me, but now I'm almost like I intentionally put myself into these imposter syndrome like uncomfortabl…”
Will Wu is building an 'Art and Science Lab' team at Match Group
“I'm just sort of beginning to build a new team here at Match Group called ASL, ASL, and it has two meanings, but one of the meanings is art and science lab”
Will Wu: Zero-to-one product development requires faith, not market research
“When you're taking a product from zero to one, you know, no amount of data or market research is going to tell you exactly whether it's going to work. You have to take a leap of faith. But then oftentimes when you're iterating beyond that, you can use data to …”
Will Wu: Over-relying on data in product design risks long-term damage
“I think it can get lost if you, again, sort of lean on data too much, because oftentimes, you know, a data, a signal that you're seeing in the data might tell you to do one thing, but if you don't stop and sort of think about how whatever feature change that i…”
Harry Stebbings calls expense software Concur 'terrible'
“Of course I know Concur. It's terrible. Terrible.”
Will Wu: Enterprise SaaS succeeds by serving buyers, not end users
“I mean, it's like an enterprise SaaS product, and they can have a really successful product because they're solving the right needs for an enterprise customer, another company that they're selling to, but I don't know if that product was designed from the grou…”
Will Wu: Early founders should talk to strangers on the street
“You can get really scrappy if you're an early founder, entrepreneur, you can go talk to people on the street.”
Will Wu: Apple used skeuomorphism on original iPhone to anchor user understanding
“The original iPhone, for example. It was very simple. Actually, they use design, user interface design patterns like skeuomorphic iconography to help ground people's understanding in real physical analogies.”
Will Wu: Products should layer in complexity only after finding product-market fit
“If you start simple, then you have a core product. If you find, if you're lucky enough to find product market fit, then you've got a simple core product that people can easily pick up and understand. And then once you've sort of captured their attention, once …”
Will Wu: Product teams should prototype with code, not just Figma
“I'm a huge fan of rapid prototyping, like not just creating interactive Figma mockups that you can click on, but literal prototyping with real software code so that you can just rapidly experiment and feel out using your intuition and your instinct”
Will Wu: Launch products with pride and vision over unattainable perfection
“I've always felt that when you're launching something brand new, It can't be perfect. Perfection is, it just takes forever, and you know, you might not actually, if you try to build perfection, you launch it after taking years, then you might find out that it …”
Will Wu: In-person presence is paramount for creative product brainstorms
“When it comes to creative brainstorming and design brainstorming, being in person is just of paramount importance.”
Wu: Observing target customers in person globally is paramount for product innovation
“Traveling around the world and seeing your target customer, for example, in the flesh, and observing the way that they're interacting with the world and understanding their environments in a really nuanced way. I think that, I think getting the right sort of c…”
Will Wu: Separate revenue and innovation into two distinct product teams
“I have found that the best approach, or at least what's worked in my experience, Is having two separate teams dedicated towards each of those two independent goals. So one team dedicated towards revenue generation, another team generated towards innovation, be…”
Will Wu: Past built products reveal candidate ability better than interviews
“I specifically look at products that people have designed or built in the past. So I think that someone's actual work, the products that they've created, or at least that they've imagined, speaks volumes, more than you can possibly extract out of like a, you k…”
Will Wu: Snap acquired Crash Club's studio after one intro call
“I confirmed my suspicion. I immediately flew over to them, like, the next day to Brisbane, Australia, to pitch them on essentially Snap buying their game studio. And we had, we made it happen incredibly quickly.”
Stebbings: Tech workers care more about work-life balance than money or mission
“People care less about money than ever before. Mission, nah, they say they do more because they think it's more moral. But they actually do care about the work-life balance, I find, much more than anything.”
Wu filters out product candidates motivated primarily by high pay
“I feel like it's this sexy, trendy job in tech that has an allure because, you know, it's known to sometimes be high paying. And I guess it's like, I try to weed out people who might only be interested in product management because of reasons like that.”
Will Wu rejects structured brain teasers in product management interviews
“If they're not good at brain teasers, they can still be good at a lot of other things. So I don't really take a very structured approach when it comes to my interviews with case studies or brain teasers or anything like that.”
Will Wu evaluates product candidates by probing specific UI design choices
“I look at what they've done. I look at what they've built, even if they're really early in their career, looking at concepts that they've created, and asking them about those concepts, asking them, oh, like, why did you put this button here, or why did you use…”
Will Wu: Over-optimism leads managers to give failing hires too long a leash
“For me personally, I mean, I think that I can get overly invested in people. I can give people a lot of room To fail because I like to see the best in people, I guess. I'm like a glass half full kind of guy, and my friends always actually give me a hard time a…”
Wu: Hiring product managers from outside industries drives first-principles thinking
“No, I think it's really oftentimes valuable to bring in folks from a different perspective, to bring in folks from a different perspective, to think through from a first principle point of view, or, you know, by leaning on the expertise, that unique perspectiv…”
Will Wu: Match Group designers use Midjourney and DALL-E for concepts
“I have already some product designers on my team who are using tools like mid journey or Dolly to come up with different UI concepts or visual concepts.”
Wu: AI tools will soon simulate instant persona-based user testing
“And I think there's probably going to be an AI tool in the near future that gives you instantaneous user feedback and testing. Like you create a mock-up or a prototype and you Hand it to the AI, and you say, hey, AI, give me feedback on this new experience as …”
Will Wu: Traditional screen user interfaces will not be replaced by AI
“I don't think that the previous human computer interaction paradigms are going away. For instance, look at, like, the command line interface. That was one of the original ways to interact with computers. Like, people are still using command line interfaces tod…”
Will Wu: Prompt engineering AI startups lack long-term defensibility
“I do think the hype is warranted, but I'll caveat by saying I wouldn't just invest in every single generative AI based company that comes across my radar. I think there's a lot of entrepreneurs right now who have a great idea maybe for how to apply Something l…”
Will Wu: Match Group considers losing users to successful relationships a product win
“I mean, it is a major success for us. If we never see someone again because they found a relationship that they're happy with for the rest of their lives.”
Will Wu: Every mobile app ultimately competes for finite attention against TikTok and gaming
“I mean, I think you're competing for time. You're sort of share of mind for people, whether it's, you know, gaming or TikTok or whatever it is. I mean, at the end of the day, there's just a finite, finite amount of time in the day. And I think, you know, at so…”
Wu: Ignore user requests only when they conflict with core values
“I think listening to users is normally extremely important, though I think the one exception might be when something that users are asking for Conflicts with one of your core product values. But then I still think you don't like ignore them, but I think it's j…”