Amazon's move into full service mode
💡JC's Newsletter #69
Dear friends,
Welcome to the 15 people who have joined us since last Tuesday! If you're reading this and haven't subscribed yet, please join the 3 436 people who have already joined the move. It's right here!
In JC’s Newsletter, I share the articles, documentaries and books that I enjoyed the most in the last week, including a must-read.
Let’s talk about this together on LinkedIn or on Twitter. Enjoy!
💡Must-read
👉 Why and How Amazon went to full Service mode (chitchcock)
The fascinating story of how Amazon decided to go from a monolithic approach to teams that communicate through services.
Jeff Bezos is an infamous micro-manager. He micro-manages every single pixel of Amazon's retail site. He hired Larry Tesler, Apple's Chief Scientist and probably the very most famous and respected human-computer interaction expert in the entire world, and then ignored every goddamn thing Larry said for three years until Larry finally — wisely — left the company.
Being aware of who you are helps you understand what people you need to hire.
Jeff’s Big Mandate went something along these lines:
All teams will henceforth expose their data and functionality through service interfaces.
Teams must communicate with each other through these interfaces.
There will be no other form of interprocess communication allowed: no direct linking, no direct reads of another team's data store, no shared-memory model, no back-doors whatsoever. The only communication allowed is via service interface calls over the network.
It doesn't matter what technology they use. HTTP, Corba, Pubsub, custom protocols — doesn't matter. Bezos doesn't care.
All service interfaces, without exception, must be designed from the ground up to be externalizable. That is to say, the team must plan and design to be able to expose the interface to developers in the outside world. No exceptions.
The rules are clear and kind of simple. They are really limiting in some ways forcing big decisions for the organization. What I like is the “technology you use doesn’t matter”, it is really about thinking about the organization.
Over the next couple of years, Amazon transformed internally into a service-oriented architecture. They learned a tremendous amount while effecting this transformation. Discoveries included:
pager escalation gets way harder, because a ticket might bounce through 20 service calls before the real owner is identified.
every single one of your peer teams suddenly becomes a potential DOS attacker.
monitoring and QA are the same thing
you won't be able to find any of them without a service-discovery mechanism. And you can't have that without a service registration mechanism, which itself is another service. So Amazon has a universal service registry where you can find out reflectively (programmatically) about every service, what its APIs are, and also whether it is currently up, and where.
debugging problems with someone else's code gets a LOT harder, and is basically impossible unless there is a universal standard way to run every service in a debuggable sandbox.
There are dozens, maybe hundreds of individual learnings like these that Amazon had to discover organically. There were a lot of wacky ones around externalizing services, but not as many as you might think. Organizing into services taught teams not to trust each other in most of the same ways they're not supposed to trust external developers.
Service-oriented architecture is very hard, and has a lot of drawbacks. In the case of Amazon it had also many advantages enabling them to build a platform in many ways (third parties, AWS,…):
From the time Bezos issued his edict through the time I left, Amazon had transformed culturally into a company that thinks about everything in a services-first fashion. It is now fundamental to how they approach all designs, including internal designs for stuff that might never see the light of day externally.
There are without question pros and cons to the SOA approach, and some of the cons are pretty long. But overall it's the right thing because SOA-driven design enables Platforms.
When software — or idea-ware for that matter — fails to be accessible to anyone for any reason, it is the fault of the software or of the messaging of the idea. It is an Accessibility failure.
That one last thing that Google doesn't do well is Platforms. We don't understand platforms. We don't "get" platforms. Some of you do, but you are the minority. This has become painfully clear to me over the past six years. I was kind of hoping that competitive pressure from Microsoft and Amazon and more recently Facebook would make us wake up collectively and start doing universal services. Not in some sort of ad-hoc, half-assed way, but in more or less the same way Amazon did it: all at once, for real, no cheating, and treating it as our top priority from now on.
The problem is that we are trying to predict what people want and deliver it for them. You can't do that. Not really. Not reliably
The two are basically the same thing, because platforms solve accessibility. A platform is accessibility.
If you want to go the full road towards being a platform, you will have to make very hard choices. That is why it is super important to know if being a platform is going to be the key enable of your business.
🏯 Building a company
In addition to selected articles, I share one of Alan's leadership principles every week - the same one that I share internally and with our investors every Wednesday.
👉 Alaners hire and develop the best missionaries (Healthy Business)
We give a lot of ownership to the Alaners, and trust them with important decisions. For that to be possible, we need to build an organization made of high performers, with very high talent density.
Alaners raise the performance bar with every hire and promotion. We hire only Missionaries: they believe in Alan's mission and also embody our Leadership Principles.
👉 How to know if you've got product-market fit (Lenny's newsletter)
1/ Companies that experienced a sudden and significant pull from the market:
Netflix: "It took us at Netflix 18 months to finally find the repeatable scalable business model that worked. [...] I called my book "That Will Never Work" because everyone I pitched that original idea had that reaction. (Including my wife!). But they were right. The original idea didn’t work. But hundreds of failed experiments later, and after many a sleepless night of worrying, we finally tested the unlikely combination of No Due Dates, No Late Fees, and Subscription that ultimately was the thing that ended up working. And boy did it work. Within days of testing it we knew we had a winner."
Your strategy should help you learn & iterate as fast as possible.
Stripe: "What indicated to us that there was something interesting here was that our friends who were using it asked if they could invite their friends, and those people invited their friends, and it spread through word of mouth process."
Are you measuring word of mouth somehow for your inbound leads? Starting to get leads from word of mouth would be a great indication of success.
Carta: “About six months after we launched we changed our phone provider I think to DialPad or RingCentral. We misconfigured the system and for some reason after it prompted the user to "Press #1 for Sales" and the user pressed the button, it would just hang up on the caller. We didn't know about it for three weeks because we were just trying to fill the non-phone orders. The way we found out was a prospect was able to contact us through chat support and wrote "I've been trying to get to your sales people for weeks and you keep hanging up on me!!! Why won't you take my money???" That's what product market fit feels like.”
2/ Companies that experienced a steady and compounding pull from the market:
Substack: “Early on I had product-market fit anxiety. Do we have it? How will we know? There isn't really a moment I can point to where that changed. We've just been growing fairly consistently, and gradually the how-do-we-keep-up anxiety got bigger and bigger until there wasn't time left in the day to worry about whether we had product-market fit. We did have retention curves that looked pretty good from early on, especially for paying readers, and writers with paying readers."
At Alan, we try to make sure that all the members we sign are incredibly happy with us
Instacart: "For Instacart, product-market fit happened across a series of moments. we found product-market fit very early on with people who wanted groceries delivered as soon as possible and didn’t care which store they came from. This made us feel like we had achieved product-market fit but it was only with a small sub-segment of customers. The average customer wanted to shop from their favorite grocery store. So, we formed partnerships with top retailers. As a result, customers started to seek us out and word of mouth grew. We then signed more partnerships, reached a larger scale with customers, and in turn attracted more partners. "
How can we better segment the market to focus our distribution? First-timers vs. switchers, cash-rich vs. cash-strapped, admin-focused vs. member-focused, ...?
3/ Companies that hit a meaningful milestone that proves the idea is working:
Caviar: "We knew we had product market fit when we had our first large company (Kabam) replace their lunch service with Caviar for the entire company. That was when we became sustainable and ramen profitable.”
🗞In the news
📱Technology
👉 Clubhouse and AirPods; Twitter, Meerkat, and Clubhouse; Clubhouse Monetization (In Silico)
Live streaming is huge in China; what is notable is that the biggest use case is for e-commerce.
Still, for now I am more bullish about the audio streaming use case than I am the video streaming one.
Speed and information density, meanwhile, are not exactly the hallmarks of live audio!
My suspicion is that most heavy Twitter users end up not being heavy Clubhouse users, and vice-versa. They are just two very fundamentally different mediums, and that is going to limit Twitter’s market for Spaces more than it will Clubhouse’s.
It makes all kinds of sense for Clubhouse to focus on creator monetization from the beginning.
The App Store guidelines now allow “individual users to give a monetary gift to another individual without using in-app purchase, provided that (a) the gift is a completely optional choice by the giver, and (b) 100% of the funds go to the receiver of the gift.”
It is very difficult to enable micro-transactions based on dollars and cents. That is why Twitch uses Bits, which you buy in packages (the smallest denomination is 100 bits for $1.40); instead of paying credit card fees on every microtransaction you pay the credit card fees for a package of a virtual currency that you can then pay out as you see fit.
👉 Twitter seeks the wisdom of crowds (Platformer)
One is an effort to crowdsource responses to misinformation, in hopes of setting the record straight in something close to real time.
“Birdwatch users are able to flag tweets from a dropdown menu directly within Twitter’s main interface”
Birdwatch, by eventually welcoming a large number of volunteers into that program.
Reddit has proven that reputation systems and voting can bring order and context to even the most unruly communities
Maybe you’re happy to see profanity, nudity, and offensive speech in your timeline. Maybe you want a timeline that consists only of images. Maybe you want a purely chronological timeline; maybe you want one ranked only by engagement. All would be possible in the world Dorsey describes, and you can imagine it being attractive to many Twitter users.
🏥 Healthcare
👉Digital Health Consumer Adoption Report 2020 - How COVID-19 accelerated digital health beyond its years (RockHealth)
How has telemedicine adoption changed?
More consumers used live video telemedicine than ever before. After leveling off between 2018 and 2019, adoption of live video care increased 11% (from 32% to 43%) in 2020, indicating the significant and swift shift to live video.
Telemedicine is not reaching new demographic populations in large numbers (yet).
Higher-income earners, middle-aged adults (aged 35-54), highly educated, and those with chronic conditions.
Use of non-video forms of telemedicine is down. Alongside the massive increase in reported live video use, fewer consumers reported using other forms of telemedicine (e.g., live phone visits, text messaging, and email) than in past years.
The future of tech-enabled care models may not be reflected in how consumers are currently using telemedicine.
How are consumers using digital health tracking tools differently during COVID-19?
More consumers reported use of digital trackers and wearables. Wearable ownership and use were both up in 2020, jumping up 10 percentage points from 33% in 2019 to 43% in 2020, after not growing at all between 2018 and 2019.Emerging spaces such as women’s health—for fertility and menstrual tracking—may have contributed to the proliferation of digital tracking: 83% of women who track their fertility and 67% of those tracking their menstrual cycle used digital methods.
It might be interesting to think again about trackers. I have never been a big fan, but the adoption rate seems to radically change in the US. Do you use trackers?
Digital health use in 2020 was a reflection of consumers’ responses to the coronavirus pandemic. Sixty percent of respondents searched for provider reviews online. The majority of respondents continue to search online for health information—on symptoms, treatments, medications.
Most live video and phone users accessed telemedicine through their doctor/clinician (not their employer, insurer, or other channels).
Prescriptions for the future: A look ahead for telemedicine
Telemedicine must shift from a transactional model to a continuous virtual care model. New innovations offer the possibility of “full-stack” services that are proactive, continuous, and better positioned to match limited provider supply to demand.
New care models are disintermediating the patient-provider relationship. The opportunity for consumers to access care outside of their existing provider relationships is growing. Many consumers will prefer their existing physician, but our 2020 data reinforced the degree to which younger consumers are more likely than their older counterparts to access telemedicine outside of their provider: The next generation of healthcare consumers is rapidly going digital.
How are consumers using digital health tracking tools differently during COVID-19?
The health metrics tracked (digitally or analog) by the most respondents were weight (50% of all respondents), medications (33%), blood pressure (32%), food and diet (32%), and physical activity (32%). The health metrics with the greatest portion of digital trackers were fertility (83% of fertility trackers are tracking digitally), heart rate (75% of heart rate trackers), physical activity (73% of physical activity trackers), and menstrual cycle (67% of menstrual cycle trackers).
Regarding their willingness to share health data, health insurers come third, behind physicians and family. Health tech companies only come seventh.
👉 🇫🇷 Comment i-Virtual mesure vos signes vitaux à partir d’un simple selfie (Maddyness)
i-Virtual met au point un dispositif à même de mesurer six variables physiologiques à partir d’une simple vidéo prise par l'utilisateur·rice : fréquences cardiaque et respiratoire, pression artérielle, saturation du sang en oxygène, variabilité cardiaque et niveau de stress.
D'après la startup, 30 secondes suffisent à faire remonter les données. Dans les faits, seuls les deux modules dédiés au bien-être sont en vente à cette heure. « Il s'agit de la variabilité cardiaque et de l'indicateur de stress, qui s'établit sur une échelle de 0 à 5 ».
💚 Alan
👉 🇫🇷 “Définir une culture d’entreprise” program is now live on Majelan
In this program, I advise on how to build a strong corporate culture. The one that brings people together around something strong. Ready to define your corporate culture?
👉 Alan has launched a Shopify platform: Alan Shop
You will find Alan’s sweaters, t-shirts, socks and our cuddly mascot.
🙋 Featured in this newsletter
Lenny Rachitsky 🇺🇸 Author, Lenny’s Newsletter | Twitter | LinkedIn
Casey Newton 🇺🇸 Founder, Platformer News | Twitter | LinkedIn
Arthur Le Denn 🇫🇷 Journalist, Maddyness | Twitter | LinkedIn
It’s already over! Please share JC’s Newsletter with your friends, and subscribe👇
Let’s talk about this together on LinkedIn or on Twitter. Have a good week!