Decision-making Question: Facebook’s Video vs Events Strategy

Welcome to the twelfth edition of PM Interview Prep Weekly! I’m Ajitesh, and this week we’re diving into one of the most instructive strategic failures in tech history: Facebook’s “pivot to video.”

The Context

It’s easy to predict the future in broad strokes. Video will dominate. AR/VR will be transformative. AI will change everything.

The hard part? Knowing when, in what format, and how much to bet.

This is the dilemma that keeps me up at night as a founder. At Tough Tongue AI, I’m convinced that AI roleplay coaching is going to be critical for training and interview prep. But when exactly? Should we go all-in on our current format, or will someone crack a slightly different implementation that wins the market? Do we bet 100% of our resources on this vision, or do we hedge?

In a startup, you can pivot quickly if you’re wrong. You make a bet, learn fast, and adjust. But at a company like Facebook or Google, the stakes are different. You have investors scrutinizing every move. Wall Street demanding quarterly results. Dependencies across dozens of teams. You can’t just pivot next week. Your decisions create momentum that’s hard to reverse.

This is why Facebook’s video decision is so fascinating. In 2015-2017, it was obvious that video was the future of content. Just like text dominated early Facebook, and Instagram cracked image-first social, video was clearly next. The question wasn’t “if” but “when” and “how much.”

The setup: Facebook has 2 billion users and the world’s most powerful content distribution algorithm. Video consumption is exploding. Every competitor—YouTube, Snapchat, Instagram, Twitter—is betting heavily on video. Wall Street wants to see engagement growth.

The pressure: Video watch time is the metric that impresses investors. YouTube shows it works at massive scale. Snapchat’s Stories prove ephemeral video drives teen engagement. Publishers are desperately seeking the next platform.

The dilemma: Should Facebook go all-in on video to compete with YouTube and Snapchat? Or should they double down on unique social features like Events that no competitor can replicate?

Today, we’re tackling: Should Facebook prioritize video content or Events in the newsfeed?

My Learning About Strategic Bets

Before we dive into the framework, let me share what I’ve learned about making these kinds of strategic bets—both from my time at Google and now building Tough Tongue AI.

1. The Format Risk is Often Bigger Than the Timing Risk

When I look at our AI roleplay coaching product, I’m confident the category will matter. The question is: will it be our format that wins, or something adjacent?

Facebook faced this exact challenge with video. They were right that video mattered—but they bet on YouTube-style long-form content created by friends. What actually won? TikTok’s short-form, algorithm-driven content from strangers. They had the right category, wrong format.

This tells me: even when you’re directionally correct about the future, the specific implementation determines success. You can’t just ask “should we bet on video?” You have to ask “which type of video, for what use case, in what format?“

2. Measured Bets > All-In When You’re Not in Startup Mode

At a startup, you can bet 100% on a vision. If you’re wrong, you pivot. The whole company can turn on a dime.

But at an established company—whether it’s Facebook with billions of users, you can’t do that anymore. You have dependencies. Stakeholders. Momentum.

This is why I’ve come to believe in measured betting for established players: you can’t go 100% into an unproven format (too risky, as Facebook learned with AR/VR’s Wall Street punishment), but you also can’t go 0% (you’ll miss the wave entirely).

The right move is often something like: 70% on your defensible strengths, 20% on maintaining parity in emerging areas, 10% on experimental bets that might be the real future. A forumale mathematically proven to be most optimal by Google Cofounder Sergey Bin.

Mark Zuckerberg invested tens of billions in AR/VR ahead of revenue, and Wall Street hammered Facebook’s stock. Was he wrong? Maybe not long-term—but the timing and resource allocation created real costs. As a founder with control, he could weather the criticism. Most PMs can’t.

3. Commodity Features Benefit Those With Distribution

When a feature becomes commoditized, it actually helps players who already have distribution.

If video becomes a commodity—anyone can build a video player and recommendation engine - then the differentiator becomes who has the users to distribute to. Facebook’s 2 billion users meant that video commoditization would help them compete with YouTube in the short term.

When Facebook tried to copy YouTube’s video model, they were solving the wrong problem. It wasn’t that video was commoditized—it was that YouTube’s video format was built for a different user intent entirely. The winning video format for social media (TikTok’s entertainment-driven, stranger content) was radically different from YouTube’s educational model.

Approach to Solving Decision-Making Questions

As I’ve discussed in previous newsletters, I follow Goal-Oriented Decision-Making—because it’s easier to gain agreement on high-level goals than specific solutions.

It has five steps:

1. Clarify the Context: Understand the situation and constraints
2. Define Goals and Success Criteria: What are we optimizing for?
3. Generate Options: Think beyond binary choices
4. Evaluate Systematically: Assess each option against your criteria
5. Recommend and Next Steps: Make the call and address risks

This framework mirrors how decisions actually get made at companies like Google and Facebook. You align on objectives first, then let strategic analysis guide you to the best option.

The Case

Interviewer: “You’re a PM at Facebook in 2015-2017. We have 2 billion users and the newsfeed algorithm determines what content they see. We’re facing a strategic choice: Should we prioritize video content in the newsfeed to compete with YouTube, Snapchat, and Instagram Stories? Or should we double down on unique social features like Events that leverage our network effects? Wall Street is pressuring us for time-on-site growth, and video watch time looks great on those metrics. What’s your recommendation?”

My Solution Using the Goal-Oriented Decision Framework

Step 1: Clarify the Context (2-3 minutes)

Let me understand the situation better:

Current state: Facebook is the dominant social network with 2B+ users globally. The newsfeed algorithm is our core product—it determines what content gets distribution and engagement. We’re in growth mode, but the competitive landscape is intensifying.

The content evolution pattern: Looking at the historical trajectory helps frame this decision:

• 2004-2010: Facebook nailed text-based status updates with some images
• 2010-2012: Instagram cracked image-first social with filters and mobile-native experience (we acquired them in 2012)
• 2015-2017: Video appears to be the next frontier—but in what format?
• Future: AR/VR is on the horizon, and we’re already investing heavily despite Wall Street pushback

Competitive pressure:

• YouTube dominates video with 1B+ hours watched daily
• Snapchat is winning teens with the Stories format
• Instagram (our acquisition) is growing rapidly with video
• Twitter launched Periscope for live video
• Every platform is betting on video

The broader strategic context: This isn’t just about one feature—it’s about understanding where to place our bets. We’ve seen this pattern in AR/VR: Meta has invested tens of billions ahead of clear returns, and Wall Street punished it. As a founder with control, he can weather that criticism. But it shows the risk of going 100% into emerging formats before there’s clarity on timing and format.

Key clarifying questions:

• “What’s driving this decision now?” → Video consumption exploding, competitive pressure, Wall Street wants engagement growth
• ”What data do we have on video vs Events engagement?” → Video watch time is 5x higher, but Event attendees have 3x better retention
• ”What’s our monetization landscape?” → Video CPMs are high initially but declining as market floods; Events monetization is unexplored

(Interviewer: “Exactly. We need growth, but we also need sustainable competitive advantage. Video looks good on Wall Street metrics, but Events might be our unique moat.“)

Step 2: Define Goals and Success Criteria (3-4 minutes)

Let me establish what we’re optimizing for:

Primary Goal: Deliver value to users in Facebook’s context while discovering what video means for social media

• Not just time-on-site (vanity metric that can mislead)
• Focus on what actually works for users in our social context

Key Constraints:

• Wall Street pressure for engagement growth (can’t completely ignore)
• Limited engineering resources (can’t do everything equally well)
• Need to balance stakeholder expectations with product experimentation
• Video format for social media is still unclear (YouTube-style? Something else?)

Success Metrics:

• Long-term user retention and repeat engagement
• User value: Does this fit what people come to Facebook for?
• Context fit: Does this leverage Facebook’s unique social graph and real identities?
• Discovery potential: Are we learning what video means in social media?

Time Horizon: 3-5 year strategic outlook—need to balance immediate stakeholder needs with format discovery

Step 3: Generate Options (5-7 minutes)

Here’s where we avoid binary thinking. Let me lay out five distinct approaches:

Option A: Status Quo (Balanced Approach)

• Continue current mix of content types in newsfeed
• No major algorithmic prioritization shifts
• Incremental improvements to both video and Events
• Let natural engagement patterns determine distribution

Option B: Full Video Prioritization

• Dramatically boost video in newsfeed ranking algorithm
• Tell publishers and creators video is the future
• Invest heavily in video infrastructure and creator tools
• Chase YouTube and Snapchat head-on with comparable features
• Maximize video watch time as the primary metric

Option C: Double Down on Events and Unique Features

• Aggressively invest in Events discovery, creation, and notifications
• Build event monetization (ticketing, promoted events, local business ads)
• Prioritize features that leverage Facebook’s unique social graph
• Bet on defensible differentiation over commodity features

Option D: Measured Investment (70-20-10 Split)

• 70% resources on Events and unique social features (competitive moat)
• 20% on maintaining video parity with existing formats (table stakes)
• 10% on experimental video formats (hedge against format risk)
• Strategic bet on differentiation with tactical coverage of emerging areas

Important clarification: These percentages refer to new content ranking priorities in the newsfeed algorithm, not replacing existing friend updates and core social content. The majority of what users see will still be updates from friends, family, and pages they follow.

Step 4: Evaluate Systematically (7-10 minutes)

Let me work through the key options against our criteria:

Option A: Status Quo

Impact on Long-term Value:

• Maintains current trajectory but doesn’t address competitive pressure
• Risk of death by a thousand cuts—losing ground gradually across multiple fronts
• Neither builds moat nor captures emerging opportunities

Feasibility: Simple—requires no major decisions or resource reallocation

Key Risk: Passive approach while competitors make aggressive moves

Option B: Full Video Prioritization

This is what Facebook actually did. Let me analyze why it was tempting but ultimately flawed:

The Case FOR Video:

• Competitive pressure is real: YouTube, Snapchat, Instagram all betting heavily
• Engagement metrics look incredible: Users spend 5x more time watching video than reading posts
• Wall Street loves watch time: Easy to report, impressive numbers, drives stock price
• Distribution advantage: With 2B users, we can push video consumption even if it’s not the right fit

The Critical Context Mismatch:

• YouTube’s success ≠ Facebook’s opportunity: YouTube works because people go there to learn—long-form tutorials, how-tos, educational news coverage. Facebook users come for quick social headlines and updates, not educational consumption.
• We’d be copying the wrong model: Betting on YouTube-style long-form, friend-created educational video in a context where people don’t want to learn
• Format completely unclear: Even if video matters for social, we don’t know what format works. (Spoiler: TikTok’s short-form entertainment from strangers, not YouTube-style education from friends)
• No evidence of product-market fit: Users aren’t showing genuine enthusiasm for long-form video on Facebook

Financial Impact:

• Short-term: +40% watch time, impressive Wall Street metrics
• Long-term: Unclear if this fits Facebook’s context, high risk of format mismatch

Option C: Double Down on Events and Unique Features

Why Events Work in Facebook’s Context:

• Users find it valuable today: People are actually using Events to coordinate real-world activities—birthday parties, community meetups, local gatherings

• Fits Facebook’s social context: Unlike educational video, Events leverage what Facebook is uniquely good at—real identities and local social connections

• Proven retention: 3x higher retention than average users—they’ve invested real-world time and social capital

Why Facebook Can Uniquely Deliver This:

• Requires the social graph Facebook built over 10+ years
• Real identities matter—you need to know these are actual people, not anonymous profiles
• Local network density—critical mass of real friends in geographic areas
• No competitor has successfully replicated this (LinkedIn tried and failed)

Monetization Potential:

• Untapped market: promoted events, local business ads, ticketing partnerships
• More sustainable than competing in crowded video ad market

Key Risk:

• Doesn’t address Wall Street’s hunger for engagement metrics
• Video opportunity might be bigger than we think
• Could appear defensive rather than innovative

Option D: Measured Investment (70-20-10)

This balances what works today with discovery and stakeholder management:

The Philosophy: We need to:

• Focus on what users value in Facebook’s context today (Events)
• Show stakeholders we’re exploring video (even if context mismatch likely)
• Discover what video actually means for social media (experimental budget)

Resource Allocation Rationale:

70% on Events and Unique Social Features:

• Users find this valuable today: Real people using Events for real coordination
• Fits our context: Leverages Facebook’s social graph and real identities
• Proven value: 3x retention shows this resonates with users
• We can deliver it uniquely: No competitor has replicated Events successfully
• Build on this strength rather than chase context mismatch

20% on YouTube-Style Video (Existing Formats):

• Satisfy stakeholder pressure: Wall Street wants to see us competing on video
• Try it even though likely won’t work: YouTube-style long-form educational video doesn’t fit Facebook’s social context
• Learn from the attempt: Even failure teaches us what doesn’t work
• Table stakes: Can’t completely ignore video category
• Keep teams exploring basic video features and Stories

10% on Experimental Video Formats:

• This is the critical discovery budget: What does video mean in social media context?
• Don’t assume YouTube’s model: Educational long-form might be completely wrong
• Explore radically different approaches: Short-form entertainment? Algorithm-driven stranger content? New interaction patterns?
• This is how we might discover the TikTok format before TikTok does
• Empower PMs and product teams to experiment even if we don’t know what will work

Why This Works:

Delivers Value Today: 70% on what works for users in our context (Events)

Manages Stakeholders: 20% shows we’re trying video, addressing Wall Street pressure

Enables Discovery: 10% lets us find what video means for social (could be transformative)

Balances Risk: Not betting 100% on unproven format, not ignoring category entirely

Critical Clarification on Implementation: This resource allocation affects new content ranking priorities in the newsfeed algorithm. We’re not replacing friend updates—those remain the core. We’re deciding how to prioritize new content types and where to invest platform development resources.

Step 5: Recommend and Plan (3-5 minutes)

My Recommendation: Option D - Measured Investment (70-20-10)

The Decision: “I recommend a measured investment strategy: prioritize Events and unique social features (70% of new content resources), try YouTube-style video to address stakeholder pressure (20%), and reserve experimental budget to discover what video means for social media (10%). This balances delivering proven user value today with managing stakeholders and discovering the right video format for our context.”

Key Risks:

• Wall Street pressure: Communicate long-term strategy; can shift to 60-30-10 if severe
• Video bigger than expected: 20% parity investment prevents catastrophic loss; experimental budget gives options
• Wrong video format: 10% experimental budget hedges against TikTok-style disruption
• Events don’t monetize: Still provide retention value as defensive feature

What Actually Happened: The “Pivot to Video” Cautionary Tale

Facebook chose Option B—full video prioritization. It became one of their most infamous strategic missteps.

What Facebook Did (2014-2017):

• Dramatically boosted video in newsfeed algorithm ranking
• Told publishers and media companies that “video is the future of our platform”
• Reported metrics showing massive video engagement growth
• Encouraged content creators to shift from written content to video production

The Inflated Metrics: Facebook counted a “video view” as someone watching for just 3 seconds. This artificially inflated engagement numbers and made video look far more successful than it was. Publishers saw these numbers and believed the market was there.

The Publisher Casualties: Major media companies went all-in on Facebook’s video vision:

• Mic: Pivoted to video, laid off journalists, eventually collapsed
• LittleThings: Viral publisher pivoted to video, shut down entirely
• Mashable: Heavy video pivot, sold at a fraction of previous valuation
• MTV News: Shut down written content division for video
• Hundreds of journalists lost their jobs as companies bet on Facebook’s video promise

The Reality Check (2017-2018):

• Ad revenue didn’t materialize as Facebook promised
• Video CPMs were lower and more competitive than expected
• Facebook quietly admitted the metrics were inflated
• The algorithm was adjusted to deprioritize video
• Publishers who had pivoted were left stranded
• Many video-focused media companies collapsed

The Deeper Irony: The Context Mismatch and TikTok Plot Twist:

Here’s what makes this even more instructive: Facebook wasn’t wrong that video mattered. They were wrong about what video means in social media context.

While Facebook chased YouTube-style video (long-form, friend-created content), TikTok launched in 2016-2017 with a completely different model:

• Short-form clips (15-60 seconds), not long videos
• Algorithm-driven feed of strangers’ content, not friends’ videos
• Highly produced, entertaining content from anyone, not casual friend updates
• Addictive scroll experience, not deliberate video watching

TikTok cracked what video means in social media context—and it was nothing like YouTube. Facebook was copying the wrong model for the wrong context.

By the time Facebook realized their mistake, TikTok had captured the teen market and was growing exponentially. Facebook had to play catch-up with Reels in 2020—years behind and copying a competitor rather than leading.

Meanwhile, Events Quietly Thrived:

While the video pivot was failing, Events became:

• Critical for small businesses and local community organizing
• A feature no competitor successfully replicated (LinkedIn tried and failed)
• Foundation for Facebook’s local business advertising strategy

Practice This Case

Want to practice this strategic decision-making case with an AI interviewer that challenges your assumptions and tests your ability to balance short-term pressure with long-term moats?

Practice here: PM Interview: Facebook Video vs Events Strategy Decision

The AI interviewer will push you on your strategic thinking, challenge your resource allocation, and test whether you can navigate the tension between Wall Street metrics and sustainable competitive advantage—just like a real Facebook PM would.

Further Reading

Explore more decision-making resources I’ve created:

• Netflix Ad-Supported Tier Decision - Previous newsletter on decision-making under competitive pressure
• Decision-Making Interview Guide - Complete guide to decision-making questions in PM interviews
• The Pivot to Video: A Cautionary Tale - Deep dive into what went wrong

PM Tool of the Week: Dub.co

As PMs, we’re constantly sharing links—product updates, feature launches, demo videos, blog posts. This week, I’m sharing Dub.co—a link tracking tool I’ve been using that’s refreshingly simple.

Here’s what I like about it:

• Dead simple: Just paste your link on the landing page and it makes it available everywhere. No complex setup, no learning curve—it just works.
• Optimized for what matters: Completely focused on driving traffic and tracking it. The interface is intuitive and built around the two things you actually care about: getting clicks and understanding where they come from.

I’ve been using Dub to track links for our newsletter, product launches, and social media posts. It’s the kind of tool that gets out of your way and just does what you need.

Using any link tracking tools that work well for you? Hit reply—I’d love to hear what you’re using!

---

How would you have approached Facebook’s video decision? Would you have had the courage to resist Wall Street pressure and stick with Events? What metrics would you optimize for? Reply and let me know—I’m especially curious how people think about balancing short-term stakeholder pressure with long-term moat building.

---

About PM Interview Prep Weekly

Every Monday, get one complete PM case study with:

• Detailed solution walkthrough from an ex-Google PM perspective
• AI interview partner to practice with
• Insights on what interviewers actually look for
• Real examples from FAANG interviews

No fluff. No outdated advice. Just practical prep that works.

— Ajitesh
CEO & Co-founder, Tough Tongue AI
Ex-Google PM (Gemini)
LinkedIn | Twitter

---

P.S. The Facebook video decision haunts me because it’s so relevant to what we’re building. I’m convinced AI roleplay coaching is the future - but am I making the same mistake Facebook made with video? Am I understanding what works in our context, or am I copying someone else’s success without thinking about fit? Should we be reserving more experimental budget to discover what the right format actually is?

The honest answer is: I don’t know. Nobody does when they’re in the middle of it. That’s what makes these decisions so hard and so important.

What I do know is this: Facebook’s mistake wasn’t betting on video. It was copying YouTube’s educational video model without asking if it fit Facebook’s social context. They optimized for vanity metrics (watch time) that impressed Wall Street instead of asking if users actually wanted long-form learning on a platform built for quick social updates.

The next time you’re in an interview (or making a real strategic decision), remember: it’s not about predicting the future perfectly. It’s about understanding your context, focusing on what delivers value to your users today, and reserving discovery budget to find the right format—not copying someone else’s format blindly.

That’s the real job.