Vaibhav Sahgal on Value Connection, Not Growth Hacking

Growth Teams Should Be Renamed Value Connection Departments

Vaibhav Sahgal hates the term "growth." He thinks it's misleading, overused, and fundamentally misunderstands what growth teams actually do. After nearly a decade at Zynga and five years at Reddit—first as Head of Growth, now as VP of Consumer Product—he's settled on a different framework: growth isn't about creating value. It's about connecting value that already exists to the people who need it.

"Growth really should be renamed to something like value creation. Connection," Sahgal says, crediting the insight to Casey Winters. "There is value that exists in your product and you need to make sure that value gets connected to each and every single user so that they may retain, they may engage, they may use the product for a really long time."

It's why tactics like push notifications or tactics like new user onboarding typically sit with growth team. Because what is new user onboarding? You didn't create anything new in the product. There's no new value being created. What you're doing is you're trying to organize your product to better connect it to the user coming in so that they retain, engage, etc.

— Vaibhav Sahgal

At Reddit, this distinction plays out in concrete ways. When the growth team sends a push notification about activity in a subreddit you follow, they're not creating content. They're connecting existing content to someone who cares. When the core product team builds a real-time chat feature for subreddits, that's value creation—a net-new capability designed to attract different users or change behavior. The confusion between these two functions, Sahgal argues, leads founders to hire growth too early, expect the wrong outcomes, and burn out channels they can't afford to lose.

The Facebook Notification Channel Zynga Burned to the Ground

Sahgal's obsession with channel discipline didn't come from reading best practices. It came from watching Zynga systematically destroy one of the most powerful distribution levers in internet history. When Facebook first opened notifications to developers, Zynga went all in. They sent hundreds of push notifications per user, per day. No guardrails. No CTR thresholds. Just volume.

"We burnt out so many fucking channels," Sahgal says. "We spammed the hell out of our users. We didn't have any guardrails set up, and we essentially burned that channel out. And not only did we burn that channel out, we made Facebook change their platform so people like us wouldn't spam it, which was a really bad thing."

When Sahgal joined Reddit, he made sure history wouldn't repeat itself. The team implemented a hard rule: you can only send push notifications if they hit a minimum 10% CTR. Anything below that threshold is spam. If your notification underperforms, you don't get to send more—you have to fix targeting or improve relevance first. The policy forced product managers to treat notifications as a scarce, valuable resource rather than a free growth lever to exploit.

The broader lesson Sahgal took from Zynga is that channel diversification isn't just about hedging risk. It's about personalization. Some users prefer push, others email, others nothing. Frequency tolerance varies wildly. "You have to be able to diversify and have an arsenal of channels at your disposal so you can personalize it for each and every user, both on frequency level and which channels you use for each user," he explains. The goal isn't multi-channel spam—it's building enough infrastructure that you can match channel and cadence to individual preferences without burning anything out.

Mary from Ohio and the Fitbit for the Brain

For four years at Zynga, Sahgal ran product for Words with Friends. He's a guy in his 30s in San Francisco who loves sports. The core user base? Women in their 50s and 60s in the Midwest who thought the game was "Fitbit for the brain." The cognitive dissonance nearly derailed him.

"In the beginning, when I worked on Words with Friends, I was just littered with all of my biases and I was trying to figure out, I really love sports. Can we do something with sports and Words with Friends? And that's not helpful," Sahgal admits. The breakthrough came when the team created a persona—Mary from Ohio, in her 50s, with grandkids—and printed her picture in the office. Every product decision went through the filter: would Mary want this? Would Mary understand this?

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 making a decision, think about Mary and would Mary want this? Would Mary like this. And that was such an important exercise.

— Vaibhav Sahgal

But the persona exercise only worked because Sahgal put in the unglamorous hours. He found real Marys inside the product and talked to them every day. Not focus groups—actual ongoing relationships with users who fit the profile. He asked about their lives, how they discovered Words with Friends, why they stuck around for years. The qualitative work transformed the persona from a demographic sketch into a living mental model he could interrogate on every feature decision.

The lesson he took to Reddit: get to know your user "in and out. Who are they? What are their motivations? Why do they use your product? And I mean like really get to know them, not just like, oh yeah, it's this person, this is why they use it. I mean like you should be in their shoes, try to be living in their life somehow." It's the difference between knowing your user is a 25-year-old male software engineer and understanding that he uses Reddit to lurk in niche hobby communities at 11 p.m. because it's the only time he gets alone after his roommates go to bed.

Hire Senior, Hire Slow, Fire Fast

When Sahgal built Reddit's growth team from scratch, he made three mistakes that nearly broke him. First, he didn't hire fast enough. He estimates he should have spent 30 to 40% of his time recruiting in the early years. He didn't. The gap forced him to write specs, run analyses, and operate tactically—which left even less time for hiring. The spiral compounded.

Second, he didn't let go of underperformers quickly enough. "We wanted to move at a breakneck speed and there are only certain types of people who are okay with that. And in the beginning we didn't really have that," he says. His current rule: for senior hires at the VP level, you should know within three to six months if it's working. If it's not, act.

3 to 6 months for someone who you've hired who's hopefully of the right caliber should be more than enough to make a decision. But of course you need to ask why it's not working out.

— Vaibhav Sahgal

The third mistake was cultural. Sahgal came in "guns blazing," imposing a new team culture and operating model without building relationships across the company first. "What I didn't do is sit down with people from across the table, form a relationship with them," he says. The result was tribal: us versus them, growth versus product, new guard versus old. It created organizational drag that took months to unwind. His advice now: bring people along for the ride. Make them understand what you're building and why it matters.

On the profile question—junior generalist or senior specialist—Sahgal is unequivocal: hire senior. A head of growth needs to analyze data, form strategy, write specs, operate, and hire simultaneously. That's not a junior job. He recommends finding someone who rode shotgun on an early growth team—someone who learned from a Casey Winters type but is hungry to lead their own build. Moneyball, but with operators who've seen the movie before.

The A/B Test That Started It All

Sahgal stumbled into growth by accident. He was a software engineer at Hi5, a pre-Facebook social network no one remembers. His PM, Mike Starbird, had him coding "people you may know" emails—pictures of potential connections, sent to users en masse. The images were huge, which limited how many people fit in each email. One night, Sahgal had an idea: shrink the images, fit more people in the frame. He ran the test. The next morning, click-through rate jumped 20%.

From that point on, I just kind of became a junkie for that process. The idea that you can have some idea, whether it might be a lousy idea or whatever, test it against real people, have them tell you if it's good or not, and read in the data if it's good or not. To me, it's just romantic.

— Vaibhav Sahgal

That one test rewired his brain. He switched from engineering to product management and joined Zynga, which he describes as "this school for product management. And of course some bad habits as well, which took a while to shake off." The romantic part—hypothesis, test, data, repeat—never left him. It's why he pushes teams to experiment constantly, why he built CTR thresholds into Reddit's notification strategy, and why he thinks growth is a long game, not a hack. The 20% lift didn't come from genius. It came from curiosity, speed, and willingness to let users vote with their clicks.