š± 5-Bit Fridays: Comparative advantages, the anatomy of a Super Bowl ad, superlinear returns, 10 strategy questions, and optimizing your fundraising
#60
Todayās 5-Bit Friday is brought to you in partnership with:
One of the most important tools in my stack as a PM is Amplitude, which is why Iām honored and excited to recommend them as an absolute must-have to you guys.
Every single day, Iām inside Amplitude to better understand how our customers are using our product, and to help me make data-driven decisions around what to build, and how to prioritize things in our roadmap.
We also use Amplitude to find interview candidates, run experiments, power job recommendations for our users, and most recently, for predictive cohorts to find customers most likely to churn.
With actionable reports, powerful ways to cut and segment data, and their new Session Replay product that brings qualitative behavioral context to my data, Amplitude makes me better at my job.
If you read my last deep dive, you know theyāre just a game changer that brings my team a constant stream of insights. If you want best-in-class product analytics, you can give Amplitude for free. Or just play around with their free interactive demo to see why the best trust them.
Plus, their new AI feature set is just absolute š„
Happy Friday, everyone! š
Big welcome to the 364 people whoāve joined us this week, including George, Damian, Ellen, and Ruby! Youāre now part of a 19,414 strong bunch of folks who like to make things.
Iāve made some very minor updates to the format of this edition of the newsletter.
First, to increase the visibility of my select partners (thatās just how the lights stay on, folks). Second, to try and keep the posts a little shorter while still conveying key insights (Iāve merged Insight and Action into one), and finally, Iāve added some details at the end around how I can help you.
If you have any thoughts, just drop a reply to this email.
Hope ya learn something new, and of course, enjoy :)
ā Jaryd āļø
ICYMI this week šļø
Anthropic, the OpenAI rival, is in talks to raise $750M in funding at an $18.4 billion valuation. Thatās still far below OpenAIās $80 billion valuation, but with their latest model release, Clause Opus, being incredibly performant, that gap could continue to close. (Take a look)
Elonās company that aims to allow us to control computers with our minds, Neuralink, has started clinical trials with a real human having an implanted brain-chip. The man is paraplegic, so it is incredible to see him able to do previously impossible thingsā¦like staying up all night playing Civilization. You can watch the patient speak at a Neuralink all hands, where he says āIt sounds kind of crazy, but it makes being paralyzed not that badā. (Take a look)
Meet Aitana! The first Spanish AI model (insanely realistic) earning around $11,000 per month. Designed because the creator didnāt like working with real people. The future of advertising, folks. (Take a look)
Historical rivals Apple and Google might be getting into bed together. Apple are in talks to build Googleās Gemini AI engine into the iPhone, setting the stage for a blockbuster agreement that would shake up the AI industry. They already have a lucrative search deal, so this may be the next piece. (Take a look)
Trump has decided to start selling $60 bibles. This is not any bible though. No, the Trump bible also features a copy of the handwritten chorus to āGod Bless The USAā. In promotion, he posted, "We must make America pray again!". If nothing else, this man loves weird merch. (Take a look)
Oh, and if youāre brand new here, or just plain missed it, my latest analysis went out on Tuesday. I unpack the product and growth playbook of Amplitude. (Take a look)
Todayās 5-Bits ā
+ Quote, tools, chart, and rabbit holes of the week
Letās get to itā¦
1ļøā£ Plentiful, high-paying jobs in the age of AI
A lot has been said about us losing our jobs to AI.
The argument for why we would is pretty simple: Weāre literally inventing AI to do stuff for us, and itās getting better everyday. And by continuously automating our labour, that puts us on some trajectory to eventually have nothing to do, right? Simplyā¦
The same argument of human labour being squeeze by AI implies, because we live in a supply-and-demand world, that the surplus of humans looking for something to do will bring down wages. Soā¦.AI automation leads to less jobs, and less pay once you find one.
Sure, we could invent new things for us to do (Ć la the Agricultural and Industrial Revolutions), but that will take a lot more time to figure out (and retrain ourselves) than it will for AI to be capable of replacing you.
Gloomy news for a Friday morningā¦.
Well, thank the Pope, not so fast.
raises an excellent point in this essay that makes me feel a lot happier: While AI certainly will be able to do most conceivable things better than you, it wonāt make sense for it to because of a two simple yet powerful principles, (1) comparative advantage, and (2) opportunity cost.The result, we should have plentiful, high-paying jobs in the age of AI dominance ā probably doing much the same kind of work that weāre doing right now.
Key quote
When most people hear the term ācomparative advantageā for the first time, they immediately think of the wrong thing. They think the term means something along the lines of āwho can do a thing betterā. After all, if an AI is better than you at storytelling, or reading an MRI, itās better compared to you, right? Except thatās not actually what comparative advantage means. The term for āwho can do a thing betterā is ācompetitive advantageā, or āabsolute advantageā.
Comparative advantage actually means āwho can do a thing better relative to the other things they can doā. So for example, suppose Iām worse than everyone at everything, but Iām a little less bad at drawing portraits than I am at anything else. I donāt have any competitive advantages at all, but drawing portraits is my comparative advantage.
The key difference here is that everyone ā every single person, every single AI, everyone ā always has a comparative advantage at something!
To help illustrate this fact, letās look at a simple example [ā¦] illustrating how someone can get paid ā and paid well ā to do a job that the person hiring them would actually be better at doing:Imagine a venture capitalist (letās call him āMarcā) who is an almost inhumanly fast typist. Heāll still hire a secretary to draft letters for him, though, because even if that secretary is a slower typist than him, Marc can generate more value using his time to do something other than drafting letters. So he ends up paying someone else to do something that heās actually better at.
ā Noah Smith, via Noahpinion
Insight & Action š ļø
Comparative advantage is already everywhere. At work (or at home where you divvy up responsibilities with your partner), everyone ends up doing the thing theyāre best at relative to the other things they could be doing, rather than the thing theyāre best at relative to other people.
This all boils down as Noah points out to producer-specific constraints, which is significant in the context of AI and our jobs, because the computing power necessary for AI to work is limited and expensive.
Yes, compute is on a path to getting more plentiful and cheaper, but it will always be finite, meaning there is a limit on amount of AI-executed tasks.
Which leads to the opportunity cost of AIās time. As Noah explains, āWhen AI becomes so powerful that it can be used for practically anything, the cost of using AI for any task will be determined by the value of the other things the AI could be used for instead.ā
Now, outside of just learning a new perspective, what can you practically do with this concept? š ļø
Practice switching your thinking in terms of competitive advantage to thinking in terms of comparative advantage. This can bring you a more nuanced approach to strategy, market positioning, and how you use your team resources. Tactically, this means:
If youāre a manager, get the maximum leverage from your team by considering how to delegate based on relative strengths and weakness.
Think about not just your productās strengths, but how those strengths compare to its other capabilities and the market at large, allowing for more precise market positioning.
Read the full (paid) post by Noah
p.s If you want the flip side to this takeā¦this post is great by
2ļøā£ The Founders Guide to Optimizing Your Fundraise
I have mixed feelings at the moment on raising VC money. I always thought it was given step when building a company, but my views have continued to change and lean towards building bootstrapped.
But thatās absolutely a personal thought and just relates to how Iād want to build a businesses if I ever did it again.
There are of course many reasons to raise money, and in a recent essay,
spoke to 8 founders whoāve collectively raised over $4B to share some concrete lessons. I really enjoyed the adviceā¦Key quote
Fundraising is a skill. At the earliest stages, storytelling ability is a big differentiator between being a successful founder and an unsuccessful one. When your startup is at an early stage, you often donāt have metrics or even a product to point to, so you have to be able to tell a narrative about what youāre doing and why youāre doing it. For many founders, storytelling doesnāt come easily, but itās a skill you can build. In later stages, company metrics drive most investment decisions. There are a lot of really smart growth investors out there, and if you feel that youāre getting shut down by them, ask them to be specific about the elements of your companyās metrics that are causing them to be cynical about an investment.Ā
Founders often treat fundraising like a sales process where they try to use leverage and create scarcity/urgency. In my experience, this doesnāt work. Scarcity and demand are important to driving an oversubscribed process to completion, but those rounds donāt become high demand because the founder has created the illusion of high demand. They become high demand because they are companies that people are excited to invest in. Itās obvious to investors when a founder is playing a game (though itās often not obvious to the founder how transparent theyāre being). Itās incredibly off-putting to investors when they realize youāve inserted artificial scarcity into a process.Ā
ā Trae Stephens (Exec Chairman & Co-founder at Anduril), via
Insight & Action š ļø
Keep investors warm with build-in-public updates and casual talks
Run a tight process to raise as fast as you can
Donāt bullshit and create artificial urgency, it will be sniffed out
Raise on a great story, or strong metrics ā not both
Have a strong narrative of your āwhy nowā and your unique advantage
The best way to get investment is to not need investment
Optimize for the right partners, not the highest valuation
Know your businessās risks, ordered from high to low
Donāt get distracted by fundraising. Focus mostly on delivering customer value
You only need a few yeses
Read the full (paid) post by Mario.
3ļøā£ The anatomy of aĀ SuperĀ BowlĀ ad
Iām not a sports guy. And as a South African, Iām especially not an American sports fan. But, every Monday after the Super Bowl, I read up on and watch all the Super Bowl ads.
Itās the one time of year where I donāt stare at an ad and wonder who keeps getting paid to create such garbage. Because seriously, 99% of general ads are so bad that I cannot wait for AI to take some of those jobs.
Sorry, somehow I just started typing out a rant thereā¦.
My point: this year I saw Duolingoās 5-second commercial. Here it is incase you missed it; their weird āDo you lesson, no butsā spot.
As a social-first brand, our approach to marketing is to generate word-of-mouth and conversation. And the goal [with this ad] was to create a WTF moment.
ā James Kuczynski, Global Creative Director at Duolingo
In our How Duolingo Grows deep dive we chatted about their unique marketing. But today, I want to share Jenny Xieās abstraction of this ad into what makes a successful Super Bowl commercial. Because from here, thereās some obvious lessons on how anyone can make effective content.
Key quote
Creativity thrives in constraints. The constraint of the five-second spot, plus collaborating with minds from all different teams, added to the diversification of ideas.
ā James Kuczynski (Global Creative Director at Duolingo), via Shortcut
Insight & Action š ļø
You donāt need to spend money to create viral content, because production-value isnāt what makes something shareable or memorable
Virility boils down to creating surprise in your content, and eliciting a strong reaction
Create constraints, even if artificial ones, to push creative thinking. Vineās and TikTokās are successful because of their simplicity
Think about how you can create a meme moment, even if just because the internet will share it
The best campaigns actually extend beyond the campaign through additional content (like Duolingo sending a push notification right after the ad)
Even āboringā software can be sold in a fun way
Donāt be scared to try new things and āpush the envelopeā in your marketing ā nobody is going to share your bland B2B blog post about security. But, tell them you do security in a fun, human, and different way, and you may be surprised.
Read the full post by Jenny.
If you want to a) get the most actionable and influential insights into your users, b) make data-driven decisions, and c) help contribute to me and Juliaās honeymoonā¦then be sure to hit this link and check out Amplitude, todayās partner.
4ļøā£ Superlinear returns
Crossing thresholds leads to exponential growth, which as Paul Graham (founder of Y Combinator) argues, leads to superlinear returns, which leads to a "winner take allā scenario.
Key quote
One of the most important things I didn't understand about the world when I was a child is the degree to which the returns for performance are superlinear.
Teachers and coaches implicitly told us the returns were linear. "You get out," I heard a thousand times, "what you put in." They meant well, but this is rarely true. If your product is only half as good as your competitor's, you don't get half as many customers. You get no customers, and you go out of business.
It's obviously true that the returns for performance are superlinear in business. Some think this is a flaw of capitalism, and that if we changed the rules it would stop being true. But superlinear returns for performance are a feature of the world, not an artifact of rules we've invented. We see the same pattern in fame, power, military victories, knowledge, and even benefit to humanity. In all of these, the rich get richer.
You can't understand the world without understanding the concept of superlinear returns. And if you're ambitious you definitely should, because this will be the wave you surf on.
ā Paul Graham
Insight & Action š ļø
The first takeaway here is thisāabsolute numbers are a vanity, distraction, and are generally unactionable. Rates are what matters, and more specifically, growth rates.
So, focus almost entirely on that metric, especially in the earlier innings of building or doing something measurable. For one, it will help prevent discouragement when the absolute numbers are low early on. But more importantly, it will help you prioritize what areas to focus on.
The second takeaway is more broad and philosophical. When picking a startup idea or deciding where you spend your time, try look for things that donāt have thresholds, but rather, a practically infinite potential to compound. And from that list, itās key to pick things that you care a lot about, because, āThe most obvious way to take advantage of superlinear returns for performance is by doing exceptionally good work, and to do something exceptionally well, you have to be interested in it.ā
There are two obvious ways work can compound.
Directly, in that doing well in one cycle causes you to do better in the next. For example, building infrastructure or growing an audience
By teaching you, in that learning has an unreachable ceiling. Even if feel you're doing badly while youāre doing something, if you're building knowledge, then you're getting exponential growth.
Thatās because knowledge grows exponentially. So, go forth and always be learning. If you're not, you're probably not on a path that leads to superlinear returns.
Read the full post by PJ.
ā ļø This post is about to get cropped in your email. Keep reading the full thing here.
5ļøā£ This is the only āStrategy Frameworkā you need
A lot has been written about how to do strategy well. And with it, a lot of frameworks and templates have been pumped out and pushed, all promising that if followed, will bring you a brilliant strategy.
Iāve said it before, copying and pasting any template usually means key context specific to your product, team, and market, is missing.
However, guiding questions are not a template. Rather, they help create a mental model for thinking strategically.
shared 10 questions that can help with just that.Key quote
A strategy document serves as a comprehensive but sharp-like-knife narrative that effectively communicates the directions necessary to succeed in the targeted market.
It goes beyond merely informing leadership, but extending to all levels of the organization, such that team members at every level understand their role in contributing to the strategy.
ā Bandan, via
Insight & Action š ļø
Tactics are not strategy. Rather, strategy is a simple narrative that should be able to to be condensed to a memorable soundbite. Strategy is clear message that sets the directions necessary to succeed in the targeted market.
To help craft that narrative, Bandan suggests answering this list of 10 questions as comprehensively and clearly as you can.
Where do we want to be in 3 years, and how does this goal compare to our current state as a company?
Which market opportunities (and needs), do we need to address to achieve our goals, and why are these the right opportunities?
Which target customers do we believe we can effectively serve within these opportunities? What are the problem areas of these people?
What makes our organization better suited than others to address these opportunities?
What specific actions will we take to maximize our chances of success in these opportunities?
Which risks do we need to mitigate along the way, and how will we do so?
Under what circumstances would we reconsider our chosen actions during the 3-year period?
What distractions must we avoid to maintain focus on our selected actions?
What support do we require to execute these actions effectively?
How will we measure our progress and know that we are moving towards success?
Read the full post by Bandan.
Quote of the week š”
If you want to make everyone happy, don't be a leader, sell ice cream.
ā Steve Jobs
Tools of the week š ļø
Every week, Iām going to start highlight products I actually use and give them an honest review. Today, Iām highlighting:
Sukha: I use Sukha to reduce stress and burnout, and have a happier and more productive work day. p.s you can join me and 76 other readers in our group
Notion: I use Notion at work to organize research and manage our roadmap, and personally, for all things this newsletter. If youāre not using Notion yetā¦š¤·
Arc: I use Arc as my browser because itās just so much better than Chrome :) You can read why I think so here.
š View all my recommendations.
Chart of the week š
Consumers are spending more than theyāre making, especially the younger ones. And for many, this is intentional and based on their worldview that life is worth living now.
Die With Zero, popular among Gen Zs, promotes the idea that once youāve saved enough to fund your retirement, you should focus on having memorable life experiences. Why die with a bunch of money? Net fulfillment over net worth. Bill Perkinsās book Die With Zero came out during the pandemic and has fueled the new worldview. This line of thinking is especially predominant among young consumers, and we see the effects in the rise of payment forms like BNPL and in increasing credit card debt.
ā Rex Woodbury
ByteĀ on one of these posts š§
My favorite reads from the weekā¦
The State of Culture, 2024, by
Here lies the internet, murdered by generative AI, by
Jeff Bezos Deep Dive, by Bill Kerr
How to reframe capitalism to make sales and marketing work (better) for you, by
Whenever you are ready š¤
Here are 2 ways I can help you:
Are you looking to grow your team? If so, Iāve invested in a company (Athyna) that can help you find incredible talent and build out your global team. Their product, service, and talent pool is just amazing. Just reply to this email if curious, or learn more here ā
Are you looking to grow your product by reaching an audience of founders, PMs, marketers, and product builders? If so, maybe we can partner up. You can reply to this email to get started, or learn more here ā
And thatās everything for this weekās edition, folks!
As always, thanks so much for reading and being part of this growing community. It means a lot.
Until next time.
ā Jaryd āļø
Jaryd, thank you for posting, and Noah is right on the money. After reading his original post a few weeks ago, I wrote the following note to myself:
Humans have a huge comparative advantage over AI in terms of cost. Consider making a robot that can do the same job as a line cook.
The cost of acquiring, maintaining, supporting, and servicing this robot would be infinitely higher than the cost of employing a line cook who is less efficient at any single task but more effective at performing a sum of tasks. Here is how:
1. A robot canāt go to the basement to fetch a box of meat from a freezer;āØ
2. A robot canāt instantly switch from grating onions to frying an omelet;āØ
3. A robot canāt clean itself because electrical appliances, no matter how well done, arenāt compatible with water and soap;āØ
4. A robot canāt repair its own mechanical and electrical components;āØ
5. A mechanical robot will āwear and tearā much faster than a human. āØ
6. Robots canāt make micro decisions the way humans can because robotsā inputs are limited;
Thatās why all of the efforts and VC investments to build automated pizzerias have failed. The cost of automation, combined with the cost of infrastructure and support, ended up being higher than the cost of running a human-staffed pizzeria.
Or consider a DaVinci (robot-operated) surgical room vs. non-robotic:
Even though a DaVinci-run surgical room is theoretically capable of working 24/7, it is still limited by the capacity of people who employ it, and the cost of Da Vinci surgery vs. human surgery is many times greater:
Initial cost:āØāØA basic 4-arm Da Vinci Si system may have a purchase price of around $1.5 million. On a 5-year lease, the monthly lease for this system could be in the $30,000 to $40,000 range.
The more advanced Da Vinci Xi system, with additional features, may have a purchase price closer to $2 million. On a 5-year lease, the monthly lease for a Da Vinci Xi system could be $40,000 to $50,000.
āØPer surgery cos:
ā The disposable robotic instruments: from $500 to $2,500 per instrument.
ā For a typical surgery, the total instrument costs can range from $3,000 to $8,000.
ā The sterile drapes and other accessories required can cost $500 to $1,000 per procedure.
ā Hospitals can expect to pay $150,000 to $300,000 per year in maintenance and service contracts for a Da Vinci system.
ā Initial training for the surgical team can cost $50,000 to $100,000.
ā Ongoing staffing costs for a dedicated robotic surgical team can add $200,000 to $500,000 annually to a hospital's operating expenses.
In total, the extra costs associated with using the Da Vinci robotic system for a single surgical procedure can range from $4,000 to $10,000, with the annual maintenance and staffing costs adding several hundred thousand dollars to that.
Also, the Da Vinci-assisted robot requires specialized doctors, meaning that a doctor who can operate on a kidney canāt operate on a liver because this requires a completely different set of skills. On the other hand, a general surgeon can perform just about anything except highly specialized surgeries such as brain or heart transplants.
Summary:
Humans have a huge comparative advantage over robotic solutions in terms of cost, flexibility, adaptability, availability, longevity, decision-making, and many other factors. Opportunity cost ā forget about itā¦
So glad you included Noah Smith's take on this - much needed pragmatic antidote to all the dramatic tropes that people are flinging around!