Sunday Reads #147: How Amazon, Zoom and InMobi became antifragile (and how Nokia became fragile)
The two ways to build an antifragile company.
Hey there!
Hope you’re having a great weekend.
I did a Twitter Space yesterday (for those of you that tuned in, thank you!).
The title was “Building Antifragile Companies”. Ravi (CEO - Scale, ex-CPO Tinder), Shravan (Swag Wala PM), and KC (VP & GM - InMobi) brought their A-game 🔥.
I went into the conversation with a 20,000 ft view. Just a broad strokes opinion on what makes companies antifragile. But they made it so REAL. Like the difference between a pixelated website on a 90s dial-up internet, and high-speed broadband.
170 people joined live, and 300+ people have listened in so far. You can listen to the recording by clicking on the image below.
Before we get into what I learned:
What exactly is an antifragile company?
I’ll let Ravi say it (source):
Fragile companies are like glass. Under the right amount of the right kind of stress, they break. We saw this with the banks in 2008. Today we see this same fragility in the airlines, many of which will go bankrupt in the next couple months.
Resilient companies are like rubber. They deform when subjected to stress, but bounce back once the stress is removed. Zillow is an example of a resilient company. They weathered the 2008 financial crisis and built a system that can withstand downturns in real estate — a system that includes $2.5B stashed under their mattress.
Antifragile companies are like muscle. Bodybuilders know that muscle tears under the right kind of stress, but it builds back up — stronger than before. Amazon is an antifragile company. Yes, Amazon has benefited from increased demand, but that radically increased demand could have easily broken a fragile company. Instead, Amazon has leveraged its diversified business portfolio and rapid adaptability to seize the moment and emerge stronger.
An antifragile company isn’t merely resilient. It doesn’t just rein in chaos. It reigns in chaos.
Amazon is antifragile. It posted record profits during COVID, despite massive investment in PPE across its supply chain.
Roblox is antifragile. By giving its users (kids) the tools to design whatever they want in the Roblox ecosystem, it is ready for the future. No matter what the kids do, Roblox will benefit. Wild experimentation is a feature, not a bug.
In his book Antifragile, Taleb says, “I want to be happy in a world I don’t understand”.
That’s the essence of how a business becomes antifragile. By not having to worry about predicting the future. No matter what happens, you win.
So what makes a company antifragile?
There were two big lessons that came out from the discussion.
#1: Empower your people. In one very specific way.
The companies that thrived during COVID had one thing in common. Decentralized decision making, by a very adaptive workforce.
Remember the Trader Joe’s toilet paper anecdote from last week?
From the company’s podcast in May 2020:
Tara (Marketing Director): Let’s go back a little bit because I want to talk about some of the things that the folks on our buying teams, very specifically, were doing during that phase of this that was really the huge impulse. Almost like a panic buy.
Matt (VP Marketing): The buying team was calm and methodical and really looking at ‘what do we need’ and ‘how can we best fill that need’. And it turns out by looking in places other than the usual places. You might see a five pound bag of rice where we previously only sold a one pound bag of rice.
Tara: I’m going to share the toilet paper story. (laughs)
I opened up my email one morning and there was an email from someone I didn’t know who was an executive at an international hotel chain saying, “Our business is down. We’re not using a lot of the things that we’ve contracted for. We have some of this, we have some of that.”
So I forwarded that email to the folks who manage our buying teams and I instantly got a message back from one of the folks on our buying teams who just said, “Toilet paper!!” with big exclamation marks at the end. Within a week and a half, we had made a deal to buy toilet paper from this large international hotel chain that suddenly didn’t have guests staying in their hotel rooms. You get to a Trader Joe’s when they have it in, because it comes and it goes and it’s, you know, it’s there and then people buy it and it’s gone. And it comes back. But we’re selling individual rolls of toilet paper that were originally intended for use in hotel rooms.
Matt: Those weren’t retail ready packages and specifically they didn’t have what’s known as a universal product code, the barcode, the UPC, so these didn’t scan at the register. And for a lot of retail businesses, that would be a make or break deal. But we figured out that, you know what, our crew is smart, they’re capable, we can figure out how to do this. We can ring it up manually. And that’s what we’ve been doing. I just love how it’s summarized in this store sign that I saw…I’ll just read the sign to you. “April break getaway canceled? Don’t worry. Now you can enjoy a hotel toilet paper experience in your own bathroom.”
That’s the power of decentralized decision making.
When you empower the people who are closest to the ground to make decisions, they will surprise you.
But empowerment is not enough. You also need to staff accordingly.
In a fast-changing environment, specialists tend to lose their specialization. They become out of step with the times.
So you need to hire generalists. Foxes, not hedgehogs. People who are happy to say, “OK I don’t know the answer to this question. But ask me again tomorrow.”
Shashank Mehta, the founder of Whole Truth Foods, said something similar when I recorded a podcast episode with him (will be out soon, I promise!). I paraphrase:
One of the big 80/20 decisions that we made, with outsize impact, was: Hire generalists.
We made an active choice not to hire veteran specialists for any function. Instead, we hired people who were more adaptable, even if less experienced. So, someone who wants to start their own company later, would join us as an “Entrepreneur-in-Residence”.
And we’d tell them - OK, we need to develop this. Go figure it out.
And they usually did.
But empowering your people to make decisions is not enough. You need to empower them in the right direction.
Which brings us to the next big learning…
#2: Capacity > Efficiency.
One thing common about companies that are antifragile is: they are able to jump when they see an opportunity.
From Ravi’s article:
Video conferencing is a crowded and commoditized space. Yet, it is Zoom that seized the moment by scaling from 10M to 200M daily users in a few weeks. Why has Zoom benefited disproportionality relative to Google, Facebook, and other companies that should have done better?
…How were they able to meet crushing demand for a service that is technically taxing even in the best of times?
“Zoom traditionally keeps about 50 percent more capacity on its network than its maximum actual usage, and the team has been busy in recent weeks maintaining that cushion.”, said Alex Guerrero, senior manager of SaaS operations at Zoom, in a recent interview with DataCenter Knowledge.
So, when other players’ servers crashed with the crazy demand in March 2020, Zoom was reliable. And once you set up an account on Zoom, why would you ever move to another provider?
Amazon was antifragile in COVID too. With the amount of free cash flow it generates, it was ready to invest when everyone else was hunkering down for the coming retail bloodbath.
Now, it’s trite to say: “OK, got it. To be antifragile, you need to have a margin of safety. You need to have extra cash in the bank, bandwidth in the server, etc.”
But that’s only one way of creating capacity. And it’s not enough. Just ask Jeff Katzenberg how Quibi crashed and burned despite having a $2.5 billion war chest.
Far more important than funds, is capacity for innovation. And InMobi is a great example of creating such capacity.
InMobi’s unicorn parade.
InMobi was India’s first unicorn. It’s valued at USD 10+ billion today (and may IPO at USD 15 billion soon). But it also has multiple breakout companies within it.
Inmobi started Glance and carved it out as a separate company three years ago. Glance raised $200M from Reliance last week, at a USD 1.8 billion valuation.
The company also acquired Roposo - a video tech platform - in 2019. And when TikTok was banned in India in 2020, Roposo swooped in to capture users. It’s another unicorn in the making.
KC is an InMobi lifer (11 years!), so I asked him on the Space, “What makes InMobi so innovative? The ad world has been hit again and again by tidal waves. Even Facebook slumped with Apple’s new privacy policies. How has InMobi stayed ahead?”.
KC’s answer was as simple as it was powerful. InMobi always ensures there’s enough capacity for innovation.
Only 80% of InMobi’s workforce is working on business-as-usual at any time. That’s a hard-and-fast rule. 20% of the workforce will always be working at the frontier of technology - whatever is next.
That’s how you create capacity for innovation.
Google’s famous “20% time” gets you some of the way there. Give every employee 20% free time to work on projects that interest them. But that’s not enough. You need to structurally push smart people to work at the bleeding edge of whatever comes next.
And you can never stop doing that.
The moment you stop to bask in your success, you’re fragile. The moment you start resting on your laurels, when your capacity turns to efficiency, you’re fragile.
As fragile as the over-optimized global supply chains today. Which crashed to a standstill because one ship ran aground in one tiny canal.
But I digress. Because the textbook case of building massive antifragility… and then losing it… is Nokia.
Nokia - the antifragile giant that became fragile.
Nokia in the 20th century was the archetypal antifragile company. A small company in a small country (Finland population: 5.5 million) somehow became the market leader in mobile telephony. At its peak in 2007, Nokia had 49.4% share of the mobile phone market.
And a mere 5 years later, it was languishing at 3.1% and almost bankrupt.
It was in such a dismal state, that it felt like Microsoft had paid too much, when it acquired Nokia for USD 7Bn (down from its peak of USD 155 bn in 2007).
How does something like this even happen?!
The story of Nokia’s rise is the story of a company that innovated in new technologies, without a clear goal in mind.
Invested in digital data transmission tech in the ‘60s, because it seemed… interesting.
Got into a JV with a radio telephone supplier, without a clear market visible.
Invested in digital switches because that was a new technology in the ‘70s. And because Nokia didn’t have the funds to develop its own chips (like its bigger competitors like Ericsson did), it used an x86 microprocessor from some company named Intel in California.
End result - when different countries came together to formulate the GSM standards in the ‘80s, Nokia positioned itself center-stage to define those standards. And wrote its existing capabilities in!
After all that, winning was inevitable.
But then, over time, David became Goliath. The company stopped innovating. It was far more fun to just enjoy the market dominance it had built.
The company lost its muscle memory for innovation.
And with the sudden rise of iPhone and Android, the company found itself on a burning platform. But… the company had forgotten how to jump.
So, to summarize:
If you want to be antifragile, then create capacity for innovation. And empower your people.
Now, I won’t lie, this advice feels a little incomplete.
It makes sense for big companies, who have the ability to create innovation teams. Who have enough cash in the bank to invest in interesting technologies.
But what about startups?
How can a startup become antifragile?
The truest answer is an unsatisfactory one. As KC said during the Space,
A startup has no business being antifragile.
What a startup needs to do is score.
Paul Graham said this best, in his celebrated essay, How to make wealth:
A startup is like a mosquito. A bear can absorb a hit and a crab is armored against one, but a mosquito is designed for one thing: to score. No energy is wasted on defense. The defense of mosquitos, as a species, is that there are a lot of them, but this is little consolation to the individual mosquito.
Startups, like mosquitos, tend to be an all-or-nothing proposition. And you don't generally know which of the two you're going to get till the last minute.
But you can get close to antifragility, by doing three things:
Have at least some cash in the bank. Be default alive, not default dead.
Hire A+ people and empower them to figure things out. And make them owners - don’t be a miser with equity.
Experiment. Always try things. In a power law world, you need to try different things, and try them fast, to see what sticks.
Ravi sums this up as “Latency > Speed”.
As a startup, just saying “I’ll move fast” is not enough. It’s hardest to change direction when your car is hurtling down the highway.
Instead, run quick experiments. Learn, iterate, and move to the next one.
Speed is not what’s important. What’s important is frequency. Frequency of feedback.
As Shravan said in the Space, a system which has high frequency of feedback thrives. A system with bad feedback loops dies.
Yes, this is the Lean Startup method. As I mentioned in The Fat Startup Experiment (the article where I predicted Quibi’s demise):
The first insight about Lean Startup is that it bootstraps leverage from scratch.
There are many types of leverage a startup can have.
Proprietary IP / best-in-class technology
Regulatory capture
A viral product
Network effects
The revolutionary suggestion of the Lean movement was that the process itself could generate leverage.
Create hypothesis → Test → Observe results → Refine hypothesis → Repeat.
Such a simple process, but generates such strong momentum.
It’s like you try to lift yourself by your shoelaces. You wouldn’t expect it to work, but magic, it does!
As Ravi said yesterday, the schools that adapted to the lockdown best were the ones that empowered their teachers to try different things. Some teachers tried teaching on Zoom. Some created interactive exercises. Some made students do puzzles.
And when something worked, the entire school was able to double down. While all the other schools were trying to figure out how to get kids to sit put while the teachers droned on on Zoom.
I usually don’t like sports analogies, but this one translates well:
So build more MVPs. Throw things on the wall and see what sticks.
And remember - your MVP can be more minimum than you think!
Golden Nugget of the Week.
If you’ve been scratching your head as to why NYT bought Wordle for $8M, here’s your answer:
The Kanye Story.
I love to say, “There are no gatekeepers anymore. Except yourself, that is”.
But I only say it. Harry Dry actually demonstrated this.
Read The Kanye Story. It’s a suspenseful, yet heart-warming account of how a 22-year old just out of college brokered a business deal with Kanye West.
Or as he says it: “From lying in bed with zero ideas to negotiating with the biggest superstar on planet earth.”
That’s it for this week. As always, stay safe, healthy, and sane.
PS. My next few emails will be a little lighter, as I try to get a project off the ground. More details soon!
Would love if you could share today’s edition on Twitter so more people can see it. Thanks a lot!
Cheers,
Jitha