Sep 13, 2023 · 59m · news
Suchit Dash: Scaling Dubsmash to 43M Users, Battling TikTok & Joining Reddit | E1060 · 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 insightful interview, Suchit Dash details the roller coaster of scaling Dubsmash to millions of users, surviving a high-stakes pivot to the dance community, competing against TikTok, and transitioning to a corporate executive role following its acquisition by Reddit.
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 18.1% of the talking time here. How this is scored →
Harry as informed peer 4.0 Guest teaching 4.6 Guest disagreement 0.9 Harry pushing back 2.8
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 53:11 Rejecting social capital leaderboards and Threads' graph porting
Suchit directly rejects Harry's assertion that leaderboards and social capital create value, arguing that ported graphs rig the system and destroy motivation for long-tail creators.
Hardest push from Harry ▶ 2:17 Harry challenges working on button colors at corporate giantsHarry forcefully challenges Suchit on whether bright young technical talent wastes their time working on minor incremental details like button colors at large companies like PayPal.
Biggest teaching moment ▶ 34:15 Explaining why friend-map social apps fail relative to aggregated dataWhen Harry cites Zenly to claim location social apps never work, Suchit educates him on why personal friend preference maps fail whereas aggregated community data networks like Google Maps and Reddit succeed.
Harry holds his own ▶ 8:35 Harry cites missed investments in Riverside and DescriptHarry demonstrates high domain expertise by connecting Suchit's creator motivation theory directly to his own missed startup investments in Riverside and Descript.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Finding a Passion for Product Management | 1 | 2 | 0 | 0 | Harry opens with an introductory question regarding Suchit's early passion for product management. Suchit details his start at PayPal and outlines his three core pillars of product love. | |
| Balancing Big Corporate Experience with Startups | 5 | 3 | 2 | 6 | Harry bluntly challenges Suchit on whether young talent wastes time working on button colors at corporate giants like PayPal. Suchit politely clarifies his experience and defends marrying big company frameworks with startup execution. | |
| Human Motivations and the Creator Economy | 6 | 4 | 0 | 2 | Suchit explains the 3 Cs of creator motivation (community, clout, cash). Harry demonstrates industry knowledge by bringing up his own failed investment theses in Riverside and Descript. | |
| Designing Dubsmash's Viral Core Mechanics | 2 | 5 | 0 | 0 | Suchit delivers a masterclass on designing viral mechanics without cohort data, highlighting human psychology, cringeless lip-syncing, and products making it into the mind. | |
| Extreme Virality and the Growth Hack Playbook | 4 | 5 | 1 | 4 | Harry asks a sharp question about whether 10 million users in 43 days actually constituted product-market fit. Suchit directly admits they did not have PMF due to dismal 30-day retention. | |
| The Retention Challenge and the Failed Messaging Pivot | 3 | 5 | 1 | 3 | Suchit breaks down why retention failed and explains their unsuccessful pivot into messaging against Facebook's dominant address book graph. | |
| Downsizing, Relocating, and Sustaining Company Morale | 4 | 4 | 0 | 2 | Suchit details downsizing Dubsmash from 40 people to 4, relocating to NYC, and managing board expectations. Harry brings in a military commencement quote regarding leadership under pressure. | |
| Identifying the Retentive Use Case: The Dance Community | 2 | 5 | 0 | 1 | Suchit explains how observing an organic user behavior on Instagram—Atlanta teens using Dubsmash for dance challenges—led to the company's major product pivot. | |
| Achieving the Smile Curve and Product-Market Fit | 3 | 5 | 0 | 1 | Suchit explains reaching 30-35% D30 retention and achieving the coveted smile curve, while breaking down informal creator monetization tactics on CashApp. | |
| Democratizing Creator Platforms and the Failure of Threads | 6 | 6 | 4 | 6 | Harry challenges Suchit's aversion to leaderboards, arguing social capital drives platform value. Suchit refutes this framing, explaining how ported graphs and top-heavy leaderboards rig the game and suppress new creators. | |
| The Evolution of Social Platforms: Beyond the Social Graph | 6 | 6 | 3 | 6 | Harry pushes back on location-based social networks, pointing to Zenly's failure. Suchit educates on the shift from friend-graph location to aggregated community trust models like Google Maps and Reddit. | |
| Competing with TikTok's Aggressive Ad Strategy | 3 | 5 | 2 | 2 | Harry shares a personal $10k bet against Threads before Suchit breaks down TikTok's unprecedented ad arbitrage strategy of buying market share via Facebook's ad network. | |
| TikTok's Ad Arbitrage and the Creator Arms Race | 5 | 5 | 0 | 2 | Harry prompts Suchit on whether TikTok's playbook would work post-ATT Apple privacy changes. Suchit analyzes ByteDance's capital advantages and the creator fund arms race. | |
| The Reality of Startup M&A: Emotional and Confidential | 5 | 4 | 0 | 2 | Suchit outlines the emotional and values-based reality of M&A versus VC fundraising. Harry strongly agrees on the necessity of maintaining secrecy from the broader team during sales talks. | |
| Navigating Acquisition Diligence and Managing Investor Expectations | 4 | 5 | 0 | 3 | Harry asks how Suchit broke the news of a non-outcome acquisition to his VC board. Suchit details the difference between term sheets and diligence, and how he reframed the competitive landscape. | |
| Why the US Doesn't Need a TikTok Copycat | 3 | 5 | 1 | 2 | Suchit reframes whether the US needs a TikTok, emphasizing community interest graphs instead. He outlines the contrast between a founder mindset and an acquired executive mindset. | |
| How a Media Leak Sparked Reddit's Dubsmash Acquisition | 5 | 4 | 0 | 2 | Suchit reveals how a press leak catalyzed Reddit's outreach. Harry references Adam Mosseri's insights on video feed ad unit constraints, which Suchit addresses by presenting a conversation-first model. | |
| The Hardest Product Decisions at Dubsmash and Reddit | 5 | 4 | 3 | 6 | Harry bluntly challenges Reddit's current messaging positioning. Suchit defends the company's community orientation, shares tough product lessons, and runs through a rapid quick-fire round. |
Statements from this episode (35)
Dash: Reddit community consensus is more trustworthy than friends' opinions
“People actually weigh in, and it's become so trusted, because the aggregation of the community and the community actually giving it weight on what is valuable and not, is more trustworthy than necessarily what your friends think about something.”
Stebbings: Passing on Riverside and Descript were my biggest investment misses
“You've articulated why I've missed my biggest misses in investing, which is Riverside and it's Descript. And I miss them both because I said, the only thing creators care about is do you bring me more cash or more fans?”
Dash: Self-consciousness over voice hinders asynchronous video adoption
“Humans hate the sound of their own voice. There's a reason why we don't send asynchronous videos to one another. It's because whenever you're playing a video back to yourself, you kind of cringe at your own voice.”
Dash: Dubsmash team built three failed applications before finding success
“The early team that worked on multiple applications before even landing on DevSmash, so overnight successes are never really actually overnight successes. They were three failed applications that the team had worked on before landing on that first Area.”
Dash: Dubsmash offloaded video hosting to WhatsApp and iMessage to save cash
“We couldn't do things like host video, so we did a bunch of lightweight hacks, like the whole product was a creator app. You could just send a video, and because we couldn't host video because it was too expensive, we just sent it to WhatsApp, Facebook Messeng…”
Dash: Early Dubsmash had poor retention of about five percent after 30 days
“The hard part was that we saw the retention really early and it was awful. After about 30 days out of a hundred people, maybe five stuck around. It was more like a party trick than anything else.”
Dubsmash Gained 1M Brazil Downloads in a Day After Neymar's Post
“Neymar posted a dub smash. We saw a million downloads on Brazil in a single day.”
Suchit Dash: Lip-Sync Products Are Weekly Party Tricks, Not Daily Habits
“DubSmash's itself of doing a lip sync product was not a daily use case at its core. It was a fun party trick. It was maybe a weekly use case.”
Dubsmash cut team from 40 to four and moved to NYC
“We'd expanded to about 40 people at this point and had raised our series B. You know, the user base was still large, even if the retention wasn't as strong. And so we realized we didn't have the team to do it, and so we made a really tough choice of scaling th…”
Dash: Dubsmash shut down major international markets during dance pivot
“We kicked out, we shut down markets. We literally shut down markets in places like India or in other areas.”
Dash: Dubsmash dance challenges directly drove rap song virality on Spotify
“We started to see rap songs starting to become actually viral on, on Spotify as a result of this.”
Dash: Dubsmash's 30-day retention jumped from 5% to 35% post-pivot
“So the first iterations of Dubsmash as a lip sync product, after 30 days, it was like, out of a hundred people, five would stick around. With this new version, out of a hundred days, almost 30 to 35 would stick around, which was significantly better.”
Dash: Dubsmash maintained 20% user retention after 12 months
“And after almost 12 months, we saw about, literally about 20% of the users still sticking around, which is crazy to see.”
Dash: Creator monetization behaviors cascade down to micro-creators with 50 followers
“So what we saw was a user with 10,000 followers would say, Hey, I'll promote your account. Just hit me up on cash app for 50 bucks and I'll promote your account. But then we would see that behavior mimicked. So a user with 50 followers would say, Pay me five d…”
Dash: Dubsmash intentionally avoided leaderboards to protect new creators
“So for us, we were always careful about ever showing things like a leaderboard or ever showing, Hey, here are the top 10 creators on Dovesmash. We always wanted to make it feel like Dovesmash was accessible and everyone had a shot at being able to be famous.”
Dash: Threads rigged the platform in favor of established Instagram creators
“And I think the challenge here with threads was, on day one, if you're completely porting over the graph, it's very hard for someone to suddenly say, hey, now I'm gonna, I wanna start from zero and become really large on threads. Because already the game is ri…”
Suchit Dash: Connecting users around content vectors holds untapped potential
“And I think people, and actually oftentimes products, take sometimes lazy approaches to say, well, we need people to connect, so let's just make sure to put in a You know, a Facebook connection in here, or an address book connection in here, and actually more …”
Harry Stebbings bet $10,000 Threads would have under 1M users
“I made a bet that they would have less than a million users in a year. And I put 10 K on it.”
TikTok spent Dubsmash's entire $7M bank balance every week
“We were a startup that had just kind of raised seven million dollars. We had the, that was the cash in the bank. They were spending that in like a week”
Dash: TikTok's 2019 Ad Spend on Facebook and Snap Cannibalized Their Users
“The ad targeting at this point in 2019 had gotten so granular You could actually get to, like, a city level, a user demographic level, and TikTok leveraged all of that. And Facebook and Snap all benefited because they saw tons of money coming in in the form of…”
Dash: Apple Privacy Changes Made Buying Massive App Growth Much Tougher
“I think the targeting capabilities after what Apple has done has gotten reduced dramatically, so I think it would be much tougher, but If you spend that much money, maybe it's possible.”
Dash: ByteDance Used Chinese Revenues from Douyin to Buy US Market
“I think they did it at a scale leveraging all of the revenues that they had gotten in China already via Dao Win and To Tiao and actually leverage that to really buy out the U.S. Market.”
Dubsmash founders decided in January 2020 to sell the company
“My co-founders and I on the first week of January of 2020 said we are going to sell the business. And we said, so from that moment moving forward, we were really clear about this was the outcome that was going to happen for them.”
Most startup sellers should focus on closing one offer, not negotiating
“And really very few companies are in the position of fielding multiple offers. Your goal is really to optimize, to get an offer and get it over the line. And I think oftentimes entrepreneurs forget that.”
Dubsmash Endured Two to Three Failed M&A Processes Before Reddit Deal
“Openly we went through sort of almost two to three failed processes until we found a successful process with Reddit.”
Dash: Tech M&A Requires Product Leader Buy-In, Not Corp Dev
“Look, with technology platform companies, you have to really connect with the product leader or the CEO. It's not really ever with the corp dev team or the business development team.”
Dash: Term Sheets and LOIs From Tech Giants Mean Very Little
“Term sheets and LOIs, even from some of the largest companies in the world, Mean very little. I think oftentimes entrepreneurs can get really excited by it, but generating those is not very difficult for a large company to do.”
Dubsmash Was Saved From a Data Breach by Cyber Insurance
“We had a moment where there was a data breach in the company, and with 10 weeks of cash left, and we happened to find that we had insurance to actually save us.”
Dubsmash Avoided Shutdown by Negotiating an Acquisition Breakup Fee
“When we figured out that we were able to negotiate a breakup fee with one of the acquirers that kept us going, otherwise we would have shut down.”
Press Report Prompted Reddit's Initial Outreach to Acquire Dubsmash
“So, interestingly enough a journalist was able to snuff out that two acquirers were actually coming in and thinking about acquiring the company. And, you know, I denied it. As much as I possibly could. And then really the next day I realized, I really don't ha…”
Dash: Video feeds inherently constrain ad inventory compared to text feeds
“Yeah, look, I think Adam is right in that that dynamic generally exists with feed oriented products.”
Reddit created a 'Fix the Video Player' subreddit to collect bug reports
“We actually even created a subreddit called Fix the Video Player”
Dash: Facebook is an execution machine, not a copy machine
“No. I think they're an execution machine.”
Dash: TikTok's strength lies in disregarding privacy and conventional rules
“Probably a disregard for privacy and conventional rules.”
Dash: United Airlines leads iOS product strategy and mobile customer tech
“What United has done over the past few years has actually really been at the forefront of any new iOS feature with kind of their live action bar notifications, messaging, and actually has built a really robust kind of Tech stack to actually facilitate that.”
Shorts cut from this episode
▶ 3 Things Every Content Creators Wants on Their Social Networ (@7:57)
▶ POV: Jimmy Fallon shouts out your app on TV · 20VC with Harr (@15:30)
▶ Most Impressive Product Strategy of 2023 · 20VC with Harry S (@58:41)
▶ Reddit Video Broke and You Won’t Believe Who Helped Fix It · (@56:26)
▶ Origin of TikTok's OG Competitor 👀 · 20VC with Harry Stebbi (@24:04)