Oct 26, 2022 · 58m · news
Vaibhav Sahgal: How We Scaled Reddit to 55M Users; Lessons from Zynga | 20VC #942 · 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
Vaibhav Sahgal, VP of Consumer Product at Reddit, shares foundational lessons on scaling products, building growth teams, avoiding distribution burnout, and implementing cheap user-validation techniques.
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 19.1% of the talking time here. How this is scored →
Harry as informed peer 2.9 Guest teaching 4.3 Guest disagreement 1.3 Harry pushing back 3.1
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 52:03 Vaibhav insists depth is Reddit's non-negotiable differentiator
When Harry asks if Reddit can pursue both depth and breadth simultaneously, Vaibhav firmly asserts that depth is their primary defense and core differentiation over rival networks.
Hardest push from Harry ▶ 36:16 Harry calls out the 'bull in a china shop' onboarding approachHarry forcefully pushes back on Vaibhav's demand for immediate 30-day execution, pointing out that coming in too aggressive pisses off existing heads of product and customer success.
Biggest teaching moment ▶ 11:45 Vaibhav clarifies value creation vs value connectionVaibhav provides a clear conceptual framework differentiating core product feature creation from growth-oriented value connection using concrete Reddit examples.
Harry holds his own ▶ 16:28 Harry challenges senior growth hiring assumptionsHarry draws on his knowledge of early-stage startups to counter Vaibhav's advice, arguing that senior growth execs are often too specialized and resistant to doing gritty early work.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Vaibhav Sahgal's Journey into Growth Product Management | 1 | 3 | 1 | 1 | The conversation begins cordially as Vaibhav shares his origin story from Hi-Five to Zynga. Harry misidentifies Hi-Five as a video conferencing platform, which Vaibhav gently corrects by explaining it was an early social network. | |
| Empathizing with Users and Shaking Bad Habits at Zynga | 3 | 4 | 1 | 4 | Harry actively presses Vaibhav for practical execution details on how an SF product manager actually understands users like 'Mary in Ohio'. Vaibhav explains user personas, direct communication, and avoiding bad channel-spamming habits. | |
| Preventing Channel Burnout Through CTR Guardrails | 3 | 5 | 1 | 3 | Vaibhav reframes growth as 'value connection' rather than 'value creation' using Reddit examples. Harry pushes for clear distinctions between core product feature creation and growth notification loops. | |
| Designing the Growth Team and Avoiding Management Pitfalls | 4 | 4 | 2 | 5 | Harry directly challenges Vaibhav's advice to hire senior growth executives, arguing that young generalists are better suited to get their hands dirty. Vaibhav defends his position while admitting to major early team management mistakes. | |
| The Growth Interview Process and Equity Alignment | 3 | 4 | 1 | 2 | Vaibhav outlines his growth interview strategy and equity expectations for key growth hires. Harry asks about metric stability, leading Vaibhav to explain downstream North Star metrics versus upstream indicators. | |
| Effective Onboarding and Resourcing for Growth Roles | 4 | 3 | 2 | 5 | Harry confronts Vaibhav on the risk of aggressive new hires acting like a 'bull in a china shop' and alienating existing leadership. Vaibhav acknowledges the danger and stresses the need to bring cross-functional peers along. | |
| Cheap Idea Validation and the Fastplay Failure | 2 | 7 | 1 | 1 | Vaibhav shares a humbling story of being reprimanded by Mark Pincus at Zynga for building 'Fastplay' without cheap pre-validation. The segment highlights a classic product mistake and the lesson of using low-cost dialogues to test demand. | |
| Cultivating Team Culture, Reviews, and Investing Rubrics | 3 | 4 | 1 | 4 | Vaibhav covers team morale during downturns, product review structures, and angel investment rubrics. Harry playfully mocks Vaibhav for breaking his own investment rubric by investing in a San Francisco hotel. | |
| Quickfire Insights: The Social Landscape and Exceptional Growth | 3 | 5 | 2 | 3 | In a rapid-fire session, Vaibhav shares insights on interest-based social networks, growth anti-patterns, and Slack's bottoms-up growth. Harry presses on Reddit's premortem and trade-offs between depth and breadth. |
Statements from this episode (26)
Sahgal: Shrinking image sizes in Hi5 emails boosted CTR 20%
“I did that one night, and then the next morning I came back, I looked at the data, and the click-through rate had gone up 20%”
Sahgal: Zynga was a PM school that taught bad habits
“Zynga at that time was basically this, School for product management. And of course, some bad habits as well, which took a while to shake off.”
Sahgal: Words with Friends core users were 50-to-60-year-old Midwestern women
“For us, it was Words with Friends, which was the product I've worked at for about four years, my term in Singa, and our users were not me, mostly not me. They were 50 to 60 year old women in the Midwest or somewhere in America, and they were looking to use thi…”
Sahgal: Product teams should anchor decisions to a visual persona
“What worked for us is putting a picture of this person that's called her Mary from Ohio, who is in her fifties and has Grandkids. Putting that picture in your office with, here's what she likes doing. Here's why she uses your product. And every time you're mak…”
Sahgal: Zynga spammed users with notifications until Facebook altered platform rules
“The worst habit I would say is we burnt out so many fucking channels. And what I mean by that is Facebook in the early days opened up notifications for the first time for developers. So you could, as a developer, just send literally hundreds of push notificati…”
Sahgal: Reddit required a minimum 10% CTR for push notifications
“One of the rules actually, before I talk about diversification, that we did the push notifications, which is a massive lever for Reddit, is we said you're only allowed to send push notifications if it hits a minimum CTR. And if it doesn't hit that CTR, it's sp…”
Sahgal: Growth teams should be defined as value connection, not creation
“Growth really should be renamed to something like value connection. And mean, and while value connection isn't exactly a very sexy term, I do think it defines what growth teams do way better than the term growth does.”
Sahgal: Growth is about multi-year impact, not short-term growth hacking spikes
“It's also not growth hacking, which is a term I absolutely hate because it makes it seem like it's this kind of short term hacky thing people, people do to get a little spike. That's absolutely not true. Growth is really about the long term. Right. And it's ab…”
Sahgal: Startups should wait until post-product-market fit to hire for growth
“I think ideally you want to do it post product market fit. In the beginning, you're trying to create value in your product, right? You're not, it's not so much about connection. I mean, there's a minimum amount of connection, obviously that's important, but th…”
Sahgal: Early growth leaders should spend 30-40% of time hiring
“Probably 30 to 40% of my time in those early years should have just been about hiring.”
Sahgal: Managers should evaluate senior growth hires within 3-6 months
“You should know within, within three to six months where the wind is blowing. And of course, like people are unsuccessful for lots of reasons. I've been unsuccessful in my career plenty of times. You really need to ask yourself why are they unsuccessful becaus…”
Sahgal: Execution ability is crucial but nearly impossible to interview for
“The number one thing you're trying to figure out is this person can execute, right? Can they actually get their hands dirty and get shit shipped and make an impact? That is also incredibly hard to find out in interviews.”
Sahgal: Reference checks must include candidates' former direct reports
“A lot of times when you back do backchanneling, you only talk to folks who were their previous managers. But I think it's actually equally important to talk to people who are on their team.”
Sahgal: Daily active unique users is Reddit's North Star metric
“For Reddit, for example we, our North Star metric is DAUQ, DAU Uniques. Probably the most important metric for the company. Leading indicator of revenue, all of that.”
Sahgal: Growth teams should prioritize retention over daily active users
“Daily active users is so, so downstream, and by itself, it really does not mean anything. So really, you don't want to focus on daily active users. You want to focus on way more upstream metrics, like retention, or even more upstream than that, what drives ret…”
Sahgal: Startups shouldn't expect growth leaders to clean messy data infrastructure
“It would be very good to be disciplined around it because you, what you want is you don't want this person to be your head of data and to actually fix whatever's happening with your data. That is not their job. What you want is for them to be able to actually …”
Sahgal: A new growth hire taking no action in month one is a red flag
“In the first month, the red flags have got to be people who are doing a lot of talking and a lot of listening and not a lot of action. The right candidate just wants to get going. And I mean, get going as in they want to launch something their second week ther…”
Sahgal: Growth teams require full-time engineers rather than contractors
“As I mentioned in the beginning, you just need a few engineers who can collaborate, who can work with this person. Full-time, much better than contractors or part-time. Again, you want people who have skin in the game, who are in it. This is a high leverage ar…”
Sahgal: Zynga's Words with Friends Fastplay feature saw only 5% adoption
“So we built a board and it was actually hundreds and hundreds of hours of engineering time. And we had launched this thing and five percent of DAU used it.”
Sahgal: Pre-testing feature interest via in-app dialogues saves hundreds of dev hours
“That would have been hours of engineering work would have saved us hundreds of hours of engineering time. We could have deployed on, on something else.”
Vaibhav Sahgal: Individual contributors generated the turnaround idea for Words with Friends
“The idea that took us out of the mess was actually from an individual contributor. It was not for me. It was not for product manager. It was from an engineer and a designer who came up with it and brainstormed meeting.”
Vaibhav Sahgal: Reddit runs monthly product reviews versus Zynga's weekly cadence
“So when I was at Zynga, we did them weekly because it was just a weekly product cadence. I mean, gaming is Gaming is brutal, right? It is a daily, weekly business. And you only have so much of a shelf life. So for gaming, that was the right cadence. For Reddit…”
Vaibhav Sahgal: Investing in a San Francisco hotel was a horrible decision
“I invested in a hotel in In San Francisco, which is a horrible investment.”
Sahgal: Social networks are shifting from friend graphs to interest-based communities
“What you're seeing now is the broad social networks are becoming all about communities and all about interests, and they just have different take on it. Like that's where tick tock is. That's what Instagram is becoming pretty quickly. That's what Reddit obviou…”
Sahgal: Reddit's biggest existential risk is imitating other social networks
“I think how we fuck up is by not, instead of doubling down where we are strong, we try to cover our weaknesses, or we look at what's happening with other social networks, and we try to follow.”
Sahgal: Hiring growth leaders based solely on past brand titles is a major error
“The number one mistake is just not going through that process and just being like, oh, this person was the head of growth at X, Y, and Z. Yeah. They will know what to do. Let's just hire them. No, it's very different. Every single situation, every single compa…”