Nov 21, 2024 · 22m · y-combinator
Twitter vs. X: Product Lessons For Startup Founders · Y Combinator
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 inaugural episode of Y Combinator's 'The Breakdown', hosts Tom Blomfield and Dave Lieb analyze Twitter's transformation into X to extract essential product design and strategy lessons for startup founders. They evaluate the risks of single-metric optimization, algorithmic feed design, rebranding pitfalls, and the vital role of founder taste in building enduring consumer software.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
The partners as informed peer 6.0 Guest teaching 2.6 Guest disagreement 1.4 The partners pushing back 1.4
speaking balance: gold is the partners, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 4:19 Countering the ad-dwell thesis
Dave challenges the premise that maximizing time spent on site benefits ad-based businesses, arguing it triggers churn when user sentiment turns negative.
Hardest push from the partners ▶ 17:23 Pushback on the X rebranding rationaleDave rejects the justification for the X rebrand, arguing Musk forced an abstract future vision onto users before delivering a coherent replacement product.
Biggest teaching moment ▶ 7:40 Recalling Facebook's newsfeed backlashDave educates on product history by reminding how universal user outrage accompanied Facebook's newsfeed rollout even while engagement metrics surged.
The partners hold their own ▶ 15:29 Monzo and Bump on brand verb equityTom demonstrates deep consumer product acumen by explaining why owning a cultural verb is the ultimate competitive advantage that should never be abandoned.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The partners as informed peer | Guest teaching | Guest disagreement | The partners pushing back | Why |
|---|---|---|---|---|---|---|
| Setting the Stage: Twitter Under New Ownership | 5 | 1 | 1 | 1 | Tom and Dave open with a collaborative breakdown of Twitter's algorithmic shift towards high-engagement, sensationalist content. Both share aligned personal observations about rising dwell time coupled with declining user satisfaction without conflict. | |
| Single-Metric Optimization vs. Long-Term Retention | 6 | 3 | 2 | 2 | The conversation explores how PMs over-optimize for single metrics like minutes watched. Dave introduces a counter-argument to Tom's point about ad models, emphasizing that high engagement on disliked content destroys long-term retention. | |
| Live Feed Audit and Reddit Troubleshooting | 6 | 4 | 1 | 1 | Tom conducts a live feed audit while Dave walks through the historical arc of feeds from Facebook's original newsfeed launch to Instagram and TikTok. Both analyze how growth mechanics gradually dilute niche community quality. | |
| Founder Lessons: Defining Product Purpose and Constraints | 6 | 2 | 1 | 1 | The discussion turns to product purpose, Dunbar's number, and Twitter's remaining core strength in real-time breaking news. Tom highlights curated lists as an underappreciated tool for bypassing noisy algorithmic recommendations. | |
| Evaluating Recent Feature Changes: Paid Blue Checks & Grok | 7 | 3 | 2 | 2 | Dave and Tom critique paid verification, Grok integration, and the rebrand to X, drawing from their founder experiences at Bump and Monzo. They emphasize that discarding established verb equity like 'to tweet' is a rare strategic blunder. |
Statements from this episode (20)
Lieb: Twitter's biggest change is shifting to an engagement-based feed
“By far the biggest change that I have seen in Twitter over the last year or so is this shift from a, you know, chronological-ish feed of topics that you have expressed that you want to follow to now kind of like more of a TikTok feed of stuff that you might Be…”
Lieb: Musk said Twitter should make far-left and far-right equally unhappy
“I think Elon explicitly said, when you're using Twitter, you should feel equally bad if you're on the far left or on the far right.”
Blomfield: Optimizing for dwell time ignores product quality and taste
“When a team and a product manager has a number to work towards, and in this case, like dwell time or engagement or number of minutes each user uses a service, like a well-meaning product team, in my experience, like tends to optimize and optimize and optimize …”
Lieb: Engaging users with content they dislike damages long-term retention
“If I am highly engaged in content that ultimately I really don't like long-term retention will be poorer, right? And so eventually, even if your business model is ads and you're getting me to spend a bunch of time on the site, I'm gonna eventually stop using i…”
Blomfield: Products requiring user-made tutorials indicate flawed UX design
“When people need to, like, record long videos telling other people how to use your product, you've probably done something wrong.”
Lieb: News Feed was the single best product change Facebook ever made
“Probably the single biggest change that Facebook, the product ever made for the better was moving to this newsfeed where basically anytime someone would modulate their wall, it would show you the content that you had chosen to see.”
Blomfield: Changing familiar consumer UX always triggers massive user backlash
“You change something that people are used to, you're going to get millions and millions of people screaming about it, whether it's good or bad.”
Lieb: TikTok skipped social graphs to rely purely on algorithmic discovery
“And TikTok, the most recent one I think just skipped straight to the end game. Which is like, let's just show all the possible content in the world and let the algorithms figure out what the human brain wants to see.”
Lieb: Building real human connection requires resisting engagement optimization
“If the product you're trying to build is like connection between real people, then probably you need to very explicitly like fight against this urge to just drive more engagement.”
Blomfield: Small-group social networks require non-ad monetization models
“A social network that kind of constrained itself to that smaller group might be really interesting, but Then you have to find a different way to monetize, because you're not getting as many eyeballs, as many hours of viewership, therefore your ad revenue goes …”
Lieb: Twitter excels in breaking news due to direct first-party publishing
“I think it is that it is so easy for first party information owners to put content onto the platform. So like we saw, like during the assassination attempt, we saw a bunch of firsthand videos that The normal media organization would not have produced for us so…”
Blomfield: Community Notes has done a good job flagging fake news
“And I feel like community notes have done a relatively good job of calling out the stuff that's fake news versus real news actually. So in, in one sense, I think that's been a really positive change.”
Lieb: Tying paid subscriptions to blue checkmarks destroyed identity verification
“That makes sense, and if they connect that to the blue check, though, they now overload this term, and now I don't know if it's just a twelve-year-old in his basement paying for Twitter premium, or an actual legit person that I should pay attention to.”
Lieb: Integrating Grok into X makes zero sense for normal users
“So this is basically ChatGPT inside of Twitter, which to me doesn't make a whole lot of sense, and I'm sure to any normal person makes zero sense.”
Lieb: A founder's job is clearly articulating product purpose and usage
“I think this is a great example of probably the job of a founder or product leader, which is to clearly articulate to the entire team and to the world What this product is for and how you're supposed to use it.”
Lieb: Elon Musk's hardware taste does not translate to software
“Jeff Bezos once said that, like, taste is not transferable across domains, and to me, this is like a great example of that. I would say, Elon, like, if I drive a Tesla, I think it's the best car that I've ever sat in, but the taste and the decisions they made …”
Blomfield: When a product becomes a common verb, it has won
“When your product gets the verb, you, you've won.”
Blomfield: Rebranding Twitter to X was total insanity
“I think when you get something short and catchy, and you have the verb, just like total insanity to change, especially for something as meaningless as X, right?”