One concept has helped me understand marketplaces more than any other.
I just published an essay in which I explore transaction costs, and how they explain:
1. Which industries make a good fit for marketplaces
2. How marketplaces have evolved and where they will go next
3. What
The way that Jensen Huang runs Nvidia is wild:
40 direct reports, no 1:1s
- Believes that the flattest org is the most empowering one, and that starts with the top layer
- Does not conduct 1:1s - everything happens in a group setting
- Does not give career advice - "None of my
My experience has repeatedly proven this Paul Graham quote to be true.
Three implications:
1. Writing cultures tend to learn faster than those that use slides, dashboards, or other mediums.
2. They learn faster because they present denser information to readers, but even more
Short thread on something I wish I had understood much better 10 years ago:
If you are ambitious, there are basically two games you can play at a startup
Many people early in their careers (especially those from big tech, consulting, finance) play the wrong game, and lose both
The business school version of how Doordash gained market share so rapidly is through clever strategy: targeting the suburbs where others weren't operating, acquiring untapped supply.
That is not wrong, but it is also not all (or even most) of the real story.
The actual story
โFollow your passionโ is such a scam.
But itโs not as catchy as โfollow the thing youโre pretty good at and people need, and then work really hard at it for a long time, and then later realize youโve built a passion for it"
โCreative intensityโ is the rate at which products or services go stale and new ones must be created. It is the key factor in predicting the future of marketplaces.
On the left, the marketplace model will ultimately go extinct. On the right, it will persist.
A thread ๐งต๐
Most strategy books are a waste of time, but 7 Powers by
@hamiltonhelmer
gave me a genuinely new mental model for thinking about how to build a moat.
Here is a brief summary of Helmer's concepts, and five common ways they are misunderstood by startups:
To junior folks recruiting at startups:
If you only have a few years of experience and they're willing to give you a Director or VP title, that is usually not a reason to get excited, but actually a clear signal not to take the job
Peter Thiel on customer acquisition:
โThe kitchen sink approach doesnโt work. Most companies get zero distribution channels to work. If you get just one channel to work you have a great business. If you try for several but donโt nail one, youโre finished. Distribution follows
It seems likely that the winning formula is hybrid: 2-3 days a week in the office
Fully in-person companies get out-recruited
Fully remote companies get out-executed
Game 1: Optimize for company outcome. Spend all of your time finding and fixing the highest leverage problems you can
Game 2: Optimize for career outcome. Seek external signals like title, scope, and team size. Make sure you're "learning the right things"
Donโt let a/b testing take the place of critical thinking.
โLetโs just test itโ is sometimes code for โI donโt want to do the hard work of talking to customers, analysis, synthesisโ.
๐ New essay on the Future of Marketplaces
An exploration into why the marketplace model will ultimately go extinct in some industries but not others, and what this means for founders, operators, and investors in the next generation of marketplaces
@JosephZaghrini
100%. Would not work in most contexts. However I do think the way they operate is part of why theyโve been able to attract the talent they have.
1. Startups are an iterated game. Earning a reputation for playing Game 1 opens doors
2. It's actually very hard to know what is good for your career. Being pulled by the intersection of what the company needs and what you're good at works better than any "20 year career plan"
At the largest companies, it's more natural (and probably actually in your best interest) to play Game 2
This is close to the definition of why it is so awful to work at very large companies
Product gets all the attention
Iโm a sucker for clever financing or distribution strategies
โ[Pinduoduo] took advantage of down payments from suppliers and used stretched payment terms to create float out of customer transactions. It used that float to fund customer acquisition
The more your personal outcome and the company outcome are synonymous, the more obvious it is that Game 1 is the right choice
Founders and people at startups with <10 people almost always play Game 1
The scarce resource in almost every company is synthesis.
We are flooded with information: data, customer feedback, missing goals, beating goals, experiments that worked and didn't, competition emerging or changing tactics.
The most impactful people in every function can parse
Picking the right startup is far more important than picking your initial role within it. If I could only evaluate three metrics to assess a startup's long term prospects, here is what they would be (none of them are topline growth):
User and revenue retention
By far the most
I'm excited to share that I'm stepping into a new role as the Chief Strategy Officer at
@faire_wholesale
I've worked with this team for seven (!) years - first as an advisor and then full time. It feels simultaneously as if we've come a long way and we're just getting started.
I believe you can't fully understand a business until you can articulate it as an equation. It forces you to identify the key outputs and where you have leverage to drive them.
Lenny and I teamed up to create equations for the most common business models in tech ๐
Every startup can be distilled into a simple equation.
And until you can express yours as one, you donโt fully understand your business.
Having this equation gives you a map for understanding your biggest growth drivers, your key inputs and output, and once your teams are
โถ๏ธ AI reading list
I added a section on AI to Long Form Reads. Because of the pace of development, many of these were written in the last few months, and some in the last few days.
Link below. Iโm sure it is highly incomplete. What must be added?
Huge thank you to the many
@provisionalidea
I find that slides tend toward a choppier, bulleted form of writing that allows for more gaps in thinking
Any time I've started with slides and then put into writing, I learn something in the process, but the inverse is not usually true
We are entering the age of the IC.
Individual contributors will be able to have >10x the impact they've had historically, and this will mean fewer total managers and layers at companies and many more people on senior IC career tracks for the long haul.
Three big drivers of this
A recommendation, especially for new managers:
Keep both of these books on your desk. They both have indexes in the back. When you come across a topic youโre struggling with (comp, hard feedback, hiring, firing) just see what they have to say.
They will rarely have perfect
But at most startups, even those with hundreds or thousands of people, it's almost always the right choice to play Game 1.
Doing so will result in a better outcome for the company, but also a better outcome your career, for two reasons:
How to get growth ideas:
- Talk to customers
- Unpack counterintuitive test results
- Talk to customers
- Articulate your growth model and where it's weak
- Talk to customers
- Tear down analogous companies
- Talk to the PM who has great ideas... because she talks to customers
When marketplaces entered, why did the TAM for taxis expand massively, the TAM for lodging expand marginally, and the TAM for housing not expand at all?
The answer is mostly transaction costs: the cost of coordinating and executing a purchase. The more you can reduce those costs
New contender for best definition of strategy:
โStrategy is accepting that you are doing something better than the other and the other is doing something better than you. You have to pick your fight.โ
- Axel Dumas, CEO of Hermes
(via latest
@AcquiredFM
pod on Hermes)
๐ Long Form Read of the Week:
Market Share, by Michael Mauboussin and Dan Callahan
If you flipped traditional competitive analysis and started only knowing market share, could you infer anything about power (ability to charge a markup over costs)?
Short answer: yes
3 ideas:
If I'm understanding the thought leadership correctly:
All saas will ultimately become marketplaces
All marketplaces will ultimately become fintechs
All fintechs will ultimately become banks
๐ค
1/ The Hierarchy of Engagement by
@sarahtavel
A framework for sustainable growth: (1) users completing the core action (2) accruing benefits and mounting losses to drive retention (3) virtuous loop by which user engagement fuels future growth.
"Four steps: choose a field, learn enough to get to the frontier, notice gaps, explore promising ones. This is how practically everyone who's done great work has done it, from painters to physicists."
- Paul Graham in "How to Do Great Work"
A framework shared with me many years ago when navigating a career decision: do you want to build, sell, or understand?
There are countless variations, but really only three fundamental paths, and figuring out how you want to spend your time is very useful.
๐จ Build:
Junior folks recruiting at startups: after you've received an offer, if the company is not willing to go 100% open book on their metrics, do not take the job.
I talk to so many people who are joining startups for the first time that want to do โhigh level strategy workโ.
This is a bad goal and they wonโt get to do much of it even under the very best circumstances.
There is just very little of that kind of work to go around for
@TheKamilFranek
Hard to know today (once in a lifetime tailwind)
But I think there is a good argument that if they didn't operate with this level of speed and nimbleness there are a few times in the early days that they would have died.
New essay: ๐๐ก๐๐ง ๐ญ๐จ ๐ฃ๐จ๐ข๐ง ๐ ๐ฌ๐ญ๐๐ซ๐ญ๐ฎ๐ฉ
In an AMA with the
@shreyas
product community, someone asked: "When is the best time to join a startup to maximize your economic outcome?"
My immediate reaction was to think about risk. Not a company's stage or how much money
Three of the most value-dense pieces you can read on startup strategy, by
@ganeumann
:
๐ก Schumpeter on Strategy - Argues that companies create excess profit through innovation, and continue to make this profit by protecting their innovation from being copied through moats. In a
New post on
@firstround
!
@lennysan
and I were riffing on the most common customer acquisition mistake we see, and created a framework to help consumer founders and operators avoid it.
@patrick_oshag
1-800-COLLECT was hugely popular in the 90s, and in response AT&T launched a competitor, 1-800-Operator
However AT&T later discovered many people misspelled Operator with an 'er' at the end... and that 1-800-COLLECT had bought the misspelled number and was siphoning customers
Why aren't there huge marketplaces in services like there are in products, despite massive TAM and highly fragmented buyers are sellers?
I just published an essay on what is holding them back, how they might soon get unstuck, and why this would produce some of the largest
The same factors that make marketplaces hard to build often make them incredible businesses at scale.
Riffing with
@onecaseman
@lennysan
and
@eriktorenberg
inspired me to dust off some of my favorite learnings on how to build a marketplace ๐งต
For anyone using the holidays to search for a new startup role: here is a ~2,000 word, ~5 min read on how I would approach it:
There is a common failure mode in early startup careers:
Step 1: Get your dream job. Doing product or growth or whatever it is that makes your heart
Incredibly interesting quote I heard recently from
@tylercowen
thatโs going to require some processing: โI think if you've lived an optimal life, there is actually a lot of regret at the end of it. The best path for a life is not regret minimization... it's something else. (1/4)
I am so excited to announce that I will be joining the team at
@faire_wholesale
to lead analytics and strategy!
Why didn't I join in 2017 when I first started advising and Faire had just raised their seed round? Good question ๐ฌ๐
Weโre thrilled to share Faire has acquired strategy & analytics consulting firm, Basis One, founded by
@danhockenmaier
. Dan will join as Head of Analytics & Strategy, bringing along an incredible team to help scale Faireโs analytics and strategy function.
Less content, consumed more deeply.
Most of what you learn from a book, an essay, or a podcast is not from the information it gives you but from your attempt to make sense of it and use it to solve the questions you are personally trying to answer.
You have to fit it into your
It's a big day at Faire as we announce our Series G at a 12.4B valuation.
I continue to be blown away by how fast this team ships. We are just getting started, and we're hiring on every team.
Here is the very hard thing about talking to customers (at least for me):
In the moment, it often feels low signal. You spend a lot of time talking to someone for just a few salient points, or perhaps it doesn't map well to anything you're focused on at all. It feels higher
In virtually every growth model I've ever seen, customer retention is the single most important input into improving both top and bottom line performance. But there is a major catch:
In almost as many cases, it is also the single hardest metric to move. If you think about a
โก๏ธ New essay on The Unbundling Fallacy
In which I argue that the expectation of mass unbundling of platforms like Reddit and LinkedIn was largely a zero interest rate phenomenon.
And how to predict the few that will break out and become category-defining businesses
Link below
๐ Long Form Read of the Week:
How to Lose a Monopoly, by Benedict Evans
If written today, it could have been titled "Alphabet, AI, and Antitrust"
The central metaphor remains a highly relevant one:
2/ Growth Loops are the New Funnels by
@bbalfour
@onecaseman
@kevinakwok
@andrewchen
The fastest growing products are driven by one or more compounding systems where inputs generate outputs that can be re-invested in the input.
Improving onboarding is the highest leverage way to improve customer retention.
One reason it is overlooked: most people on the team simply don't remember what it was like to be a new user of their product - they have spent too much time with it
Great read on why strategy matters, even for startups.
It draws on probably the best two books on strategy written in the last two decades:
- 7 Powers by Hamilton Helmer
- Good Strategy Bad Strategy by Richard Rumelt
3/ Product Channel Fit by
@bbalfour
Distribution follows a power law: most products get >70% of growth from a single channel: virality, paid marketing, sales, or content/SEO. To win, a product must be built to be highly competitive in a given channel.
"Perhaps modern time travel is our ability to take a loan out from our future success to ensure we achieve it."
Finally got to new
@kevinakwok
essay and it is worth your time:
The most common mistake marketplaces make when choosing new markets to enter: focusing too much on TAM.
You need some minimum viable TAM to justify the investment, but it's just not that important. This is because (1) most potential markets are quite large and (2) even if they
There is a part of the food delivery market story that is mostly about getting the strategy right, which is the shift from the old guard like Grubhub that did not offer delivery to the new guard of heavily managed marketplaces like Doordash and Postmates, and all that this
There is a "hierarchy of needs" when choosing a startup: start at the bottom and move up when a condition is met.
Many people apply these in reverse, optimizing for comp and title and then wasting years at a company that goes nowhere.
More on each step:
One of the things you learn working on businesses that serve SMBs: because of the high turnover rate, you can build a big business quickly without convincing anyone to switch. You just have to be the way new SMBs do business.
This has never been more true than it is today:
7/ Why Onboarding is the Most Crucial Part of Your Growth Strategy by
@onecaseman
Retention is the most important metric in your growth model. Onboarding (getting users to stick) is often the most effective way to work on retention.
5/ Increase Funnel Conversion with Psych by
@dariusmc
Conversion rate optimization is not necessarily about reducing friction, it is about managing a user's psychology.
Perhaps the strongest signal that someone will be successful at a startup: they minimize the amount of energy spent worrying about who gets credit or what is fair, and plow it into just doing the work.
Excellent posts from
@lennysan
and
@onecaseman
with benchmarks on the most important part of your growth model: retention.
One addition to this is the increasingly common scenario where low user retention can actually be a good thing ๐
10/ The Most Powerful Growth Hack Of 2019 Isnโt Sexy
@KateBour
The best growth teams use both quantitative and qualitative signals to guide decision making. For many, incorporating customer research is an untapped and potent weapon.
About as good of a one-line description of strategy as you can get:
"Strategy is making trade-offs, and aligning those trade-offs so that they are a force multiplier rather than in conflict with each other."
-
@gilbert
on the Benchmark dinner episode of
@AcquiredFM
Summary of 2-hour Huberman pod on how to prevent and treat colds and flus:
- people are contagious for longer than you think
- wash your hands and don't touch your face. Viruses mostly enter though your eyes
- moderate exercise (<60 min) boosts immune system, tons of exercise
How to actually calculate payback periods for a marketplace (without deluding yourself):
A good payback period for marketplaces serving consumers is 6-12 months. Early stage marketplaces virtually always have to operate way above that, because they don't have the liquidity to
Network effects are not created equally. Two dimensions that are highly predictive of the power of network effects in generating long term defensibility for marketplaces:
X-axis: How much supply does it take before the benefit to buyers begins to asymptote?
Is supply relatively