I often have trouble describing exactly what Notion does because there are seemingly a limitless number of reasonable use cases. When Notion first launched, the co-founder, Ivan Zhao, described it as a “a unified tool for docs, wikis, and tasks.” Today, I would describe Notion as a fully customizable, all-in-one solution for note taking, databases, document storage, project & task management, team collaboration, and organization. Wow, that was a mouthful. Imagine the best parts of Google Drive, GitHub Wiki, Dropbox, Evernote, Trello, Asana, and Jira all combined into one. And the cherry on top? It’s actually fun to use.

Let’s dive into why I love using Notion.

Founded in 2016, the team at Notion has quickly grown to a base of over 4 million users and an impressive valuation of $2 billion. I was introduced to Notion less than 2 months ago and completely fell in love. As an individual user, Notion leads the charge as the best productivity tool because of its wide customizability. In working in small teams, such as a startup environment, the additional real-time collaboration features make it a fierce competitor to other popular task management platforms.

Imagine the best parts of Google Drive, GitHub Wiki, Dropbox, Evernote, Trello, Asana, and Jira all combined into one.

Notion for Individuals

Every Notion page is a blank canvas that is meant to be morphed to fit the individual.

Whether you’d like to manage your travel plans, blog posts, job applications, or weekly agenda, Notion is the perfect tool. This wide customizability begins with the powerful use of “basic blocks” and ends with the creation of pages, databases, and views. A block is similar to a Lego piece in that users can stack them to fully customize the final product to fit their specific desires. These blocks can be turned into over 50 different types of content including pages, bullets, to-do-lists, integrations, and databases (calendars, tables, boards). However, to its detriment, a blank page can cause a significant amount of confusion for new users in the most effective ways to interact with the product. Notion has created a series of tutorial videos and articles to help alleviate these pain points. Additionally, they have numerous starter templates, such as Figure 2 below, to give signifiers for how to use the product.

Notion makes organization delightful

As 100s of productivity tools have entered the market in the past decade, the companies that will succeed in the long run are those that place the most emphasis on the user experience. As seen in Figure 2 above, Notion pages are created with the use of emojis. This reduces the dread in the process of organizing information and adds strong visual effects to every page. In my opinion, this is one of the best features of the product. In fact, Dr. Linda Kaye, a cyberpsychologist, stated that seeing positive emojis such as the smiley face can be impactful through emotional contagion. Notion makes a particular effort to nudge users to use emojis to represent new pages. In fact, any page created without an emoji has an empty placeholder making it feel incomplete. Similar to Dr. Linda Kaye, they likely figured out that associating productivity with the lighthearted/playfulness of emojis leaves users feeling happy.

Notion for Teams

In a small group environment, Notion provides strong collaboration features. The most pertinent include note taking, task / project management, views, tagging, and integrations.

(1) Note Taking

The core feature of Notion is its powerful note taking capabilities. Through the use of blocks, you can quickly add to-do lists, bullets, snippets, images, and much more to any page. With rich text formatting, you also have a surprising amount of flexibility in organizing notes, creating headers, and customizing page structures. Similar to Google Docs, everything is real-time and automatically saved. Team members can leave comments on each other’s work and collaborate on note-taking.

(2) Task / Project Management, Views, and Tagging

While Notion for individuals is used primarily as a wiki, one of its powerful collaboration features is the ability to create projects, assign tasks to team members, and manage timelines / responsibilities. Every database can be stored in multiple different views that link all the processes together. While the default view is a simple table, users can easily add custom views to better fit their needs. These include a kanban board (as seen in Figure 3), calendar, list, or gallery.

(3) Integrations

One of the most useful collaboration features in Notion is its ability to embed information from third party applications into pages. For example, Google Drive, Twitter, GitHub, and Figma. This completely streamlines all outside sources of work into one central platform — easing the process of cross-platform note-taking and sharing.

Where Notion needs to improve

While Notion is an extremely effective tool for individuals and small teams, there are several key scalability issues that the product still needs to comb through. The most pertinent is the development of a public API. This will allow for outside applications to integrate with Notion and for users to pull additional data into databases. This has been one of the most highly requested user features for over a year. According to the Notion Team, “a public API is on the roadmap.”

Because of its broad customizability, for large teams, Notion is not as effective as a dedicated project management tool. For example, the real-time collaboration makes it difficult to establish consistent workflow. Additionally, the “ramp up” period for new users on Notion is significantly higher than some of the more rigid task management tools such as Jira or Trello. With an extremely large team, these problems can be magnified. Lastly, one of my personal pain points with Notion is that you can’t embed another Notion page within a page. While not a pressing issue, this can add difficulty in trying to link sub-tasks to projects.

With Spotify’s 248 million active monthly users (Oct. 2019), 124 million paid subscribers, and paid user base growing at over 30% year-over-year, we have to wonder how they’ve managed to differentiate themselves as the leading music streaming service in the world. Comparatively, Apple Music has ~60 million paid subscribers (last updated in June 2019). Although the two apps have very similar features, Spotify prevails in its ability to work cross-functionally with other devices and subscription packages.

How Apple plays an unfair game

(1) The “30% Apple Tax”

According to Spotify’s “Time to play fair” video, they claimed that they have to pay 30% of all Spotify Premium subscription fees to Apple as a result of being forced to use their payment system for in-app upgrades. This makes it extremely difficult for Spotify to sustain competitive pricing. Apple later rebutted against this point claiming that Spotify only pays a 15% “app tax” under it’s revenue sharing model, which only affected about 0.5% of their 100 million subscribers before Spotify started using their own payment system outside of the app.

(2) Apple is the creator of iOS and the App Store

As the owner of the App Store and iOS, Apple maintains a competitive advantage in controlling Spotify’s ability to submit updates to their app, communicate with customers, and provide excellent user experiences. Spotify has claimed for years that Apple has previously blocked their iOS upgrades or ability to send emails to Apple users. In fact, it wasn’t until October 2019 that Apple allowed Spotify and other third party music apps to integrate Siri into their user experience (a massive disadvantage for Spotify users). As Apple Music is the default music application for all iOS devices, Spotify is fighting an uphill battle that doesn’t seem to be ethical. Upon buying an iPhone or iPad, users that do not choose to sign up with Apple Music may receive frequent notifications similar to Figure 1 below.

This ability to serve push notification advertisements directly to potential users at no extra cost proves to be immensely advantageous (especially considering Spotify spent almost $1 billion on ads in 2019). Several Apple users have even reported having trouble getting these Apple Music notifications to disappear.

For Spotify to have the privilege of sending push notifications to users, they first have to fight the battle of getting them to download the application from the App Store. However, Apple doesn’t believe that they are in the wrong. In a statement released on their website in March 2019, they directly fought back against many of the complaints made from the Spotify team.

“After using the App Store for years to dramatically grow their business, Spotify seeks to keep all the benefits of the App Store ecosystem — including the substantial revenue that they draw from the App Store’s customers — without making any contributions to that marketplace.”

….

“The majority of customers use their free, ad-supported product, which makes no contribution to the App Store.

A significant portion of Spotify’s customers come through partnerships with mobile carriers. This generates no App Store contribution, but requires Spotify to pay a similar distribution fee to retailers and carriers.

Even now, only a tiny fraction of their subscriptions fall under Apple’s revenue-sharing model. Spotify is asking for that number to be zero.” — Apple

Is Apple being ethical?

This debate elicits a much larger conversation about where the line should be drawn between ethical business practices and sustained competitive advantages. Should Apple be allowed to make their music app the default on all iPhones and iPads? If not, what precedent does this set for Apple calendar, mail, and other utility apps that we know and love to use? Is the fact that Apple nudges users toward their own music streaming service on Apple products an unethical business practice or the outcome of an amazing business model?

Why Spotify excels

Apple Music and Spotify are nearly identical in their ease of use, pricing, music catalog, and social sharing features. For these features, big proponents of either product could make strong arguments to say why one is advantageous over the other.

The largest difference between the two platforms is their music recommendation algorithms. Spotify is famous for their “Daily Mix” and “Discover Weekly” playlists while Apple creates a culmination of music “For You.” From my personal experience with both products, I think that Spotify does a much better job providing new music recommendations tailored to my preferences. I find that while Apple does well recommending new music from the same artists already in my library, Spotify is able to introduce new artists based on my listening patterns that has immensely grown the breadth of my knowledge. While this experience is surely anecdotal, Spotify proves to lead the mark due to cross-functional compatibility and subscription packages.

(1) Cross-Functional Compatibility

For Spotify, the desktop app is intuitive on both Mac and Windows, and can be easily controlled from the mobile application. However, for Apple, the desktop application is built into iTunes which introduces increased complexity, especially trying to setup with Windows. In order to control the desktop app from mobile, users have to download the Apple HomeKit, which can be a cumbersome task. For Android, without control of the operating system, Apple Music also loses the many advantages it used to enjoy on iOS (free advertising, compatibility with Siri, serving as the default application, etc.).

(2) Subscription Packages

Spotify recently launched a partnership with Hulu, the popular video-streaming service, that grants free ad-supported accounts to premium subscribers. For students, they also partnered with SHOWTIME, another popular video-streaming service, providing a bundle deal priced at $4.99/month. These are remarkable luxuries that have set Spotify apart from Apple Music in a substantial way. Additionally, Spotify provides a free ad-supported version for users while Apple Music does not. Thus, for the millions of users who wish to avoid paying music subscription fees, Spotify is the clear option between the two.

Features Spotify should integrate to stay on top

Spotify, my favorite music streaming product, is an amazing platform that has completely revolutionized the variety of music that I enjoy. With their particular focus on college students and the younger demographic, there are three features that I would suggest implementing if I was running the product team: (1) “Play Next”: Allowing users to add songs to the beginning of the queue, (2) “Fork Playlist”: Allowing users to download their own versions of created playlists and alter them, and (3) “Song Requests”: Allowing users to push song requests to their friends’ queues.

As part of the younger demographic, the music listening experience is becoming increasingly social. Users are taking advantage of social media and group chats to share their favorite songs, new albums, or their own custom playlists. As these trends continue to spike, the long term winner of the Spotify x Apple Music battle is going to be whichever product is able to build an ecosystem around their platform that is enhanced by network effects. This means that the value of you downloading Spotify increases when your friends also download Spotify. In order to build that ecosystem, Spotify should implement the following features:

(1) “Play Next”

The “queue” is a feature used by both Apple Music and Spotify that allows the user to manage the order of songs being played next. For Apple Music, you can add music to either the front or the back of the queue depending on when you prefer it to be played. Additionally, they make it very easy to clear the queue or view its history. However, for Spotify, you can only add music to the back of the queue, the history is hidden, and there is no obvious way to clear. As a result, trying to add a song to the front of the queue becomes a big user pain point. This is only solvable by first adding to the back, and then dragging the last song all the way to the top — a frustrating process that can lead to many unintended consequences.

(2) “Fork Playlist”

In the current version of Spotify, if I download a playlist from my “Daily Mix” or from a friend, I can only listen to that exact version. If I wanted to contribute, I would have to either text him or her offline to add/delete music, or copy a new version from scratch (an extremely daunting task with large playlists). To solve this user pain point, similar to GitHub, Spotify should build a feature to allow users to “fork playlists” and edit their own versions. In an effort to build an ecosystem, this would promote increased social sharing amongst users in the community.

(3) “Song Requests”

As more and more users are listening to music in group settings, Spotify should implement a feature to allow friends to send each other song requests. These requests can then be added to the front or the back of the queue. This feature would make big strides in building out network effects, similar to Apple’s iMessage, by creating a fear of missing out: the right to send song requests only exists for those with Spotify.

Cheating growth metrics is a habitual mistake that many inexperienced founders make in the early stages of their startup. This can manifest itself in a variety of different methods: paying users to download your app, buying google/facebook ads, etc. When prompted, founders will declare that they are paying for ads to “quickly drive users to their platform.” In reality, most founders pay for ads as a shortcut for having to build great product — this will usually come back to bite them in the long run. Using Sakichi Toyoda’s, Five Why’s technique, you can quickly find the ultimate reason behind this common founder mistake:

Q1: Why are you paying for ads?

A: Because I want to quickly drive users to my app.

Q2: Why don’t you drive users to your app through organic channels (word of mouth, virality, etc.)?

A: I’m trying and it’s not working that well.

Q3: Why is it not working that well?

A: Because i’m not growing fast enough. I need to reach a larger audience.

Q4: Why aren’t you growing fast enough?

A: People aren’t sharing the app with their friends / colleagues.

Q5: Why aren’t people sharing the app with their friends / colleagues?

A: It’s early and there’s still a ton of bugs. People aren’t that excited about the app

“In reality, most founders pay for ads as a shortcut for having to build great product — this will usually come back to bite them in the long run.”

What’s so bad about paying for ads early on?

Paying for ads = cheating growth. Imagine that you’re given a pile of clay and told to make a functioning bucket. The purpose of this bucket is to hold and retain as much water as possible. When building the first iterations, you’d likely pour water into the bucket to test if it can do its job effectively. If there were massive leaks at any point, you’d likely spend time identifying and fixing the leaks before adding more water. This all seems intuitive, right?

Now imagine that your startup is that pile of clay. The purpose of your startup is growth — attracting and retaining as many users as possible. While building the first few iterations of your startup, you’re likely to experience low retention rates and software bugs (massive leaks in your bucket). This is where many founders lose their sense of intuition. Instead of doing the grunt work to identify and assuage the leaks in their company, founders will pay for users to download their app. This results in vanity metrics: numbers that look good on paper but don’t actually indicate any real value. These metrics often hurt founders in the long run because they mask the underlying issues until it’s too late and funding dries up. The only real benefit that these metrics provide are making founders feel good about themselves while working on their startup. But can you guess what happens to water that’s poured into a leaking bucket?

Priority #1: Finding Product-Market Fit

As a founder, your primary concern should be to find product-market fit (PMF), the way in which your product satisfies a strong market demand. This concept was developed by Andy Rachleff, Co-Founder of Benchmark Capital:

“If you address a market that really wants your product — if the dogs are eating the dog food — then you can screw up almost everything in the company and you will succeed. Conversely, if you’re really good at execution but the dogs don’t want to eat the dog food, you have no chance of winning.” — Andy Rachleff

Finding product-market fit proves that you’re building something that (1) tons of people want and (2) has the capability of reaching and serving all those people. Unfortunately, very few founders will ever actually find PMF. The ones that do are typically the ones that we read about in TechCrunch. Although there is no formulaic approach, building with a user-centric focus and consistently iterating on feedback will place you in the best position possible. According to Marc Andreessen, a successful entrepreneur and investor, founders should know exactly where they stand in relation to Product-Market Fit.

“You can always feel when product/market fit isn’t happening. The customers aren’t quite getting value out of the product, word of mouth isn’t spreading, usage isn’t growing that fast, press reviews are kind of ‘blah’…. And you can always feel product/market fit when it’s happening. The customers are buying the product just as fast as you can make it — or usage is growing just as fast as you can add more servers. Money from customers is piling up in your company checking account. You’re hiring sales and customer support staff as fast as you can. Reporters are calling because they’ve heard about your hot new thing and they want to talk to you about it.” — Marc Andreessen

What if there’s no other way for me to reach my target user than buying ads?

Founders often excuse cheating growth by saying that they don’t have organic contact with their target user. For example, a college male building an app to help middle aged women shop for clothes. While this may be a valid question, it elicits another major concern: Product-Founder Fit. This is the measure of how well-equipped a founder is to run his or her startup.

When product-founder fit is combined with product-market fit, you have founder-market fit. According to Forbes, this is the innate, unfair advantage that sets founders apart from their competitors. For example, Mark Zuckerberg had founder-market fit in building Facebook given that he was a college student at the time. If there is no organic way for you to communicate with your core users, you’re likely not the right person to be building that product. How do you build an effective solution without being able to empathize with your user’s pain points? Often times, the most successful founders are the ones that fit perfectly within their own target demographic.

“Paying people to download an app is easy. Building an amazing product that people want to pay for is a lot harder.”

Paying for ads is not always bad

Founders should invest in paid advertisement if (1) they’ve already found product-market fit, or (2) they have high switching costs and have clearly identified that the lifetime value (LTV) of a customer is worth far more than the customer acquisition cost (CAC).

(1) After finding PMF:

After you’ve found product-market fit, paying for ads can result in an extremely high return on investment. In the today’s advertisement world, you can optimize your marketing dollars efficiently by showcasing your product to your exact target user. Otherwise known as targeted ads — the crux behind the social media business model. This is the equivalent of building the perfect bucket and then filling it with a fire hose.

(2) LTV > CAC, with high switching costs:

As mentioned previously, LTV stands for the lifetime value of a customer. This can be calculated by the following formula:

(Rev. per customer — expense per customer) / (1 — customer retention rate)

If the average revenue of a customer is $100, the expense of having that customer is $80, and we retain 75% of our customers annually, then the lifetime value of that customer is (100 - 80) / (1 - 0.75) = $80.

The CAC, measures the cost of acquiring a new customer. This is measured by dividing the number of new customers by the amount of direct marketing dollars spent to acquire them. For example, if I spend $1,000 on a Google Ads campaign and get 100 new customers, the CAC will be $10. In other words, it costs me $10 to get a new customer.

In this scenario, the LTV/CAC ratio would be 8x. In order to create economic value, the LTV/CAC ratio of a business must be greater than one.

As an early startup, founders should pay for growth if the nature of their business induces high switching costs, and the lifetime value of users is much higher than the cost to acquire them. A great example of a company that fits within this mold is TurboTax, which boasts a 77% retention rate. Once users become comfortable with the software, they are very unlikely to switch platforms. Thus, they can justify spending a significant amount to acquire each new customer.

Stop paying for users and start getting scrappy

The art to being a successful founder is the ability to get scrappy — finding cheap and innovative solutions to get things done. In Lenny Rachitsky’s article, he wrote an amazing piece on how some of today’s greatest consumer apps acquired their first 1,000 users through seven main strategies. You’ll notice that none of these strategies include “Sat at home and paid for Google Ads.”

The rising inequality that exists within America started with the policies implemented by our very own government.

By definition, the term wealth gap, is the “the widening gap between America’s wealthiest and its middle and lower classes.” But when we speak about the wealth gap, it would be unfathomable to ignore any mention of race. This “widening gap” exists because of systemic racism, and the ramifications of these discriminatory policies that have pervaded our nation’s deep history are still alive and thriving today — the median wealth of a white family is ~$171,000, while for African-Americans it’s only ~$17,000.

So how did we get to where we are today?

Less than 200 years ago in America, slaves used to be printed on money — the same areas today where we hold the faces of U.S. Presidents. Unlike our presidents, the slaves weren’t being glorified.

This was a perpetuating reminder that the only purpose that black people served in America was to perform slave labor. Not only were they a physical representation of wealth, but they actually were the backbone of the economic system. By the time slaves were freed in the 1860s, they were generating over $3.5 billion of economic value, thus making them the “largest single financial asset in the entire U.S. economy, worth more than all manufacturing and railroads combined.” When the freedom bells finally rang, President Lincoln approved a bill that promised 40 acres of land to each family of freed slaves. This was estimated to set aside 400,000 acres of land for over 4 million families, an estimated economic value of over $6.4 trillion dollars in today’s worth. However, after he was assassinated, the bill was quickly revoked and thousands of slaves were evicted from their land. This was a major turning point that normalized the idea that black people don’t deserve to own property.

African-Americans have been barred from owning property for generations.

Owning property in America is the most influential factor in building generational wealth. Approximately two-thirds of all of the wealth of America’s middle class families comes from home ownership. This is because the wealth that properties generate for families grows over time with compounding interest. This is the idea that if I invest $100 today, then in 10 years, not only will my principal investment gain interest, but the interest that I gained on the principal will gain interest. To put this into perspective, you should realize that a $100 bill in 1863, is worth about ~$2,000 today. However, if you were to have invested that same $100, taking advantage of compounding interest, it would be worth over $3.5 million today.

The status quo of black people not having the right to property has induced some of the greatest tragedies that have affected our communities. These are the underlying motivations behind the massacre of Black Wall Street and federal processes such as redlining.

“This ‘widening gap’ exists because of systemic racism, and the ramifications of these discriminatory policies that have pervaded our nation’s deep history are still alive and thriving today.”

After the great depression, President FDR released over $100 billion dollars of funding in the form of mortgage credit to give Americans the opportunity to own their homes. This brought to life the idea of the “American Dream.” After one of the worst economic periods our country has ever faced, the opportunity to own a home was invaluable. But there was one small caveat. During the implementation of these policies, the Federal Housing Administration (FHA) decided that they wouldn’t give credit to any homes located in areas deemed to be “risky.” So how did they calculate which areas were risky? African-American neighborhoods. The result? 98% of the loans that were issued during this period were given to white families.

When we speak about the term “redlining,” it’s not just a mere figure of speech. The FHA would actually draw maps with entire neighborhoods marked in red where African-Americans lived. They then agreed to not ensure any of those homes. Thus, not only could African-Americans not afford to own the homes in their own neighborhoods, but they couldn’t move into white neighborhoods because that would bring the property values down. These policies were so entrenched in the system that even real estate agents could get their licenses revoked if they were found to be integrating neighborhoods.

The negative feedback loop of racist policy

“Racism begets more racism.” All of this began with policy that governed who can participate in home ownership — discriminatory practices against African-Americans. Essentially, federally enforced segregation. This gave birth to the pernicious effects of racism: low property values in African-American neighborhoods; businesses not willing to operate in those areas; low funding for public schools and programs; higher rates of crime and incarceration. I could break off the multitude of byproducts into an entire essay. All of these outcomes of racism then stack on top of each other and give rise to more racism. People see the current affairs of African-American life in a vacuum and give damaging comments: “African-Americans are just low achievers”; “most of them don’t even graduate from high school. They’re lazy”; “Look at their incarceration rates. They’re criminals.”

As a whole, African-Americans weren’t given access to mortgage credit up until the 1990s. However, because they weren’t financially literate, they had no idea that banks were engaging in predatory lending techniques with them at much higher rates than any other race. Instead of getting the same types of federally subsidized loans that FDR issued during the New Deal, they were getting subprime loans. It didn’t matter if you had a perfect credit score. Put simply, subprime loans are made to seem attractive because they start out either extremely cheap or free. However, the interest payments eventually compound at vicious rates — ones that African-Americans couldn’t possibly afford to pay back. As a result, when the housing market crashed in 2008, black communities lost over 53% of their wealth.

“When we speak about the term “redlining,” it’s not just a mere figure of speech.”

Why the wealth gap is systemic.

The reason that we use the word “systemic” when we talk about racism and the wealth gap is because the policies that discriminate against African-Americans were inscribed by the very same government that we have today. As a country, if we can ever dream about closing the wealth gap, we first have to truly understand and admit to the historical context behind it.

Grandma’s Secret Chocolate Chip Cookie Recipe

Imagine that your grandma gifts you her renowned recipe for chocolate chip cookies. This recipe has been passed down in your family for generations and are the most delectable cookies that you’ve ever tasted. Now that you have it, you decide to make some for your friends. They salivate nonstop after eating them. Every week they beg and plead for you to bake more and soon start inviting their other friends to your house. Your cookies are so appetizing that people are actually willing to pay you for them. You’re suddenly the talk of the town.

After a few weeks of strangers coming in and out of your house, you start to grow discontent. You enjoy selling cookies, but don’t want strangers in your personal space. Thus, you decide to open up a restaurant.

This is the beginning of basic systems design: the process of defining the architecture, interfaces, and data for a system. It all starts with an algorithm, a piece of code that takes in some input and generates some output. In our case, your cookie recipe. People are willing to pay you for this algorithm, but right now it only exists on your computer. For them to use it, they have to come to your house. In order to solve this problem, you decide to host your algorithm on the internet. For us, this translates to opening a restaurant.

But what does it actually mean to “host your algorithm on the internet?”

This means that you pay a cloud computing platform, such as Amazon Web Services (AWS), to store your code on a virtual computer, otherwise known as a server. This provides several advantages:

(1) Anybody can use your algorithm. Now, instead of people having to come to your house and use your computer, you can provide a website link that will redirect to your virtual computer.

(2) Security and Encryption. AWS provides incredible safety so that you don’t have to worry as much about hackers stealing your precious algorithm (keep in mind that nothing is ever 100% secure).

(3) Accessibility and reliability. No need to worry about the power going out in your house and crashing the algorithm. AWS is extremely reliable and accessible. If you want to make any changes, you can remote login from anywhere.

In order to host your code on the internet, you’ll need to follow something called a protocol, which is a set of rules that govern how you receive and send data over a network. For the internet, we will use the hypertext transfer protocol (HTTP), which I’m sure you already have mild familiarity with. For our restaurant, these are going to be the laws, rules, and regulations that we have to follow when serving cookies to our customers.

Within the first week of hiring staff and opening the restaurant, the place is packed. People are obsessed over these cookies. Based on increasing demand from customers, you intend on offering a larger variety of cookie flavors. In addition to chocolate chip, you now offer snickerdoodle, sugar, macadamia nut, and oatmeal raisin. At all hours of the day, the store is filled with a line wrapped around the block. You can no longer manage all of the traffic; it’s time to start thinking about expanding.

Basic Systems Design

Before we get too far, let’s examine how your cookie restaurant resembles the most basic architecture of a system. In basic systems, we have a client, a server, and a database (Figure 1). The client sends requests to the server through an Application Programming Interface (API). This is basically a messenger that passes information back and forth between the client and the server. Once the server receives the request from the client, it will process it and send back a response through an API. This response will be either a failure or a success message along with some data. In our cookie example, the client represents our customers. When our customers look at the menu and decide to order food, they will send a “request” through the waiter, who represents an API. The waiter will then bring this request to the kitchen, which represents the server. The kitchen then knows exactly what to prepare based on the request. In order to complete the order, the kitchen will have to pull some ingredients (data) from the food storage cabinet, which represents the database. When the kitchen finishes the order, they will send a “success” response back to the customer using the waiter as their messenger (an API). However, if the customer requests something from the kitchen that is not available, then the kitchen will respond with a “failure” message.

Scaling the Server

Now that we have an understanding of the basic infrastructure of systems, let’s talk about expanding our restaurant (also known as scaling our server). Why would ever want to scale our server? To handle more requests, of course. Ever since we added a variety of cookie flavors, our popularity has gone through the roof and we can no longer make cookies fast enough to handle all of the requests. This is what would happen if tens of thousands of users tried visiting your website at the same time (remember when you visit a website, you’re sending a request to a server for the information that you’re seeing). In order to solve this problem, we can either decrease the amount of time it takes for our customers to get their orders (latency), or increase the maximum number of cookies that we can produce in a given period of time (bandwidth). Both of these solutions have to do with making changes to the server. There are two ways to scale: Vertically and Horizontally.

“Why would ever want to scale our server? To handle more requests, of course.”

Vertical Scaling

Vertical scaling is adding more resources to the existing server (computing power, memory, etc.) so that you can process more requests. For our restaurant, this would look like hiring more manpower (another baker) and buying more ovens. This is the quickest solution but has prudent limitations: a single point of failure and hardware limit.

Single point of failure: If you decide to keep everything under one roof and add more resources, you’re consolidating the risk of your entire business on one server. This can be problematic for our restaurant if a fire burned down our kitchen.

Hardware Limit: There’s a limited number of new bakers and ovens that we can buy before we run out of space in our restaurant and will be forced to buy another kitchen.

Horizontal Scaling

Horizontal scaling is adding more servers to balance out the amount of requests that any given server has to handle. This would translate to building another physical location for our restaurant.

Overall, horizontal scaling is the better solution for growing with customers. However, this method has limitations of it’s own: load balancing and potential data inconsistency (the food might not taste the same at the new restaurant).

Diving into Load Balancing

If you decided to build a new physical location, you would ideally need some sort of application that directed customers to different locations in order to manage the traffic (you’d want to avoid a situation where one location is empty and another is packed). A load balancer is an algorithm that takes in requests from the client and routes them to different backend servers to balance network traffic (Figure 2). This can take a significant amount of time to get running because it involves building out an entire network of cloned servers that are running the same code and have access to the same data. There are several different types of load balancers that we could use to direct traffic. For example, if we moved all orders from in-person to online, we could track the I.P. address of where all the orders are coming from. We could then redirect customers to the restaurants closest to them to pick up their orders. This is called a Networking Load Balancer (Layer 4). Additionally, we could redirect customers based on what types of cookies they ordered. For example, we could send all the people who ordered chocolate chip to one location, and snickerdoodle to another. This is called an Application Load Balancer (Layer 7).

Of course, Amazon Web Services could take care of all of this for you.

Caching

Ultimately, you decide that building a new restaurant, or horizontal scaling, is the best approach for the future. You’ve also decided to move all orders to an online website that uses a simple load balancer to direct traffic based on I.P. addresses. To further increase the customer experience, you’ve decided that you’re going to keep tabs on the last order that each customer made so that you’ll remember for the next time they return. This is the idea behind caching. Caching data is an extremely important and persistent concept in systems design. This is typically a key-value pairing that sits in between the server and the database. Here’s an example of a simple cache for recent orders:

{
	"Dave": "1 snickerdoodle", 
	"Lisa": "4 chocolate chip",
	"Michael": "3 oatmeal raisin"
}

The purpose of a cache is to store temporary data, which increases the speed at which it is read (reading information from a large database can be a slow operation).

Other Design Considerations

As you continue to expand, there are several other systems design decisions that you’ll need to make that will be crucial to your success. Each of these decisions carry unique tradeoffs. For example, as the amount of menu items increase, you may start to think about only offering certain flavors of cookies in specific locations. This is known as database partitioning, or sharding, and could significantly reduce the amount of ingredients you’d need to hold at any given store. In order to implement this, your load balancer would need to have a lookup table to be aware of exactly which cookies each of your restaurants was selling. Another important decision could be how to efficiently setup your food storage cabinet, or database, to prioritize either organization or speed. To prioritize speed, you could use a strategy called database denormalization: grouping ingredients in the cabinet by recipe as opposed to by category (e.g. all the ingredients to make macadamia nut cookies could be found next to each other). However, as with all other systems design decisions, this would have its own tradeoffs: eggs, milk, and sugar would appear in 5 different places within the cabinet.

Do you ever find yourself aimlessly scrolling through Instagram or Twitter for hours on end? Sure you may spend 3+ hours per day scrolling through social media, but you’d never say you were “addicted” right? Addiction is for people with serious problems, and you can stop whenever you want. Right?

So you decide to take a break. You delete Instagram, Twitter, and all social media apps on your phone so that you can focus on being present. You type up a message to all your followers to tell them that you’ll be taking a leave of absence from social media. You feel proud of yourself.

4 hours pass.

You pick up your phone, scroll to where Instagram used to live on your home screen, and then quickly remember that you deleted it. This happens several more times throughout the course of the day. You start to feel moderate angst. You wonder if people have been commenting on your most recent post about taking a break. Nobody has texted you yet; what’s going on? You decide to quickly re-download Instagram to check your messages.

You’re back on all social media platforms within the week.

Welcome to the life of a teenager. According to the Addiction Center, psychologists estimate that nearly 5–10% of Americans today are addicted to social media. That’s somewhere in between 16.5–33 million people (for reference, the population of Texas is 29 million). So why are so many adolescents addicted to this new era of technology? Is there something fundamentally wrong with our generation? How have these social media platforms been able to completely command over our attention like this?

“So why are so many adolescents addicted to this new era of technology?”

It all starts with the business model.

Social media apps are free to use and so they make their money by selling advertisements. This means that if I have something to promote, I can pay Facebook to advertise it for me. Since they have over 2.6 billion active users, not only will they guarantee me that people will see it, but they also allow me to indicate the exact type of person to whom I’d like to show it (college kids, elderly aged, etc.). This is called targeted advertising, which allows millions of companies to optimize their marketing spend. You wouldn’t waste time trying to sell a pacifier to a college student in real life right?

In 2019, Facebook’s global ad revenue was ~$70 billion, which made up ~98.5% of their total revenue. Put simply, these companies have a vested interest in keeping your eyes on their platforms. But how do they actually do this?

Painstakingly intense product design.

Every social media company invests millions of dollars into hiring teams (engineers, designers, marketing, etc.) that specifically focus on making their platforms highly engaging. These products aren’t addictive by chance; they’re designed that way. Nir Eyal, a famous author and investor, calls these “Habit-Forming Products” and takes a deep dive into the methods these companies use to capture so much of our attention through his Hook Canvas.

Trigger

To hook users into their platforms, social media companies use two types of triggers: External and Internal.

External Triggers are the direct methods that they use to get you onto the platform. For example, continuous notifications telling us that we’re missing messages, and constant emails telling us what we missed.

Internal Triggers are the subconscious associations that prompt you to open the platforms. These are far more important and are where these companies aim to influence you. Internal triggers are most effective when there’s a recurring negative emotion, such as boredom, and opening the app serves as a temporary relief. This is what leads to addiction.

Action

After social media companies succeed in getting you to open the app, their next target is to make sure that you’re immediately captivated. They accomplish this through a process called variable rewards. B.F. Skinner, a Harvard psychologist, explored this topic in his famous experiment with mice in 1948. In this experiment, one group of mice would press a lever and randomly receive a small treat, a large treat, or nothing. The other group would receive the same treat every time. Overall he found that the group of mice receiving variable rewards pressed the lever voraciously while the other group pressed it at very low rates. Humans, similar to the mice, respond compulsively to variable rewards. The endless scroll through your Twitter feed all seems worth it if you can find just one tweet that’ll make you fall on the floor laughing.

Reward and Investment

Now that they have your undivided attention, they need to plant the seeds to make sure that you’ll come back every day. They want to be asked out again after the first date. They accomplish this through offering fleeting rewards and asking you for an investment. The rewards that social media platforms use are fairly similar across the board: likes, favorites, retweets, reposts, etc. From an outside perspective, these appear to be exciting and useful metrics that tell you how much your friends and other users are engaging with your content. In reality, these are designed to be ephemeral rewards that trigger the same neural circuitry as drugs such as cocaine, placing you in a constant feedback loop always wanting more. Finally, to make sure that you never dump them, they get you to invest time and energy into the platform. After months of building followers and curating your page, you’ll feel that you’ve devoted too much time to quit.

“These products aren’t addictive by chance; they’re designed that way.”

What’s really happening under the hood and what are the ramifications of these design strategies?

Instagram, Snapchat, Twitter, etc. are designed to make users addicted to the platform. The more time that you spend on it, the more money these companies make from advertisements. These platforms’ reward mechanisms (likes, favorites, retweets) activate neurons in the hypothalamus causing an influx of dopamine, a neurotransmitter associated with pleasure, to be released into your synapses. This is the exact same chemical reaction that your body undergoes while gambling or through the use of drugs. The brain receives a “reward” in the form of a temporary feeling of pleasure, and subconsciously associates the gambling, drug, or social media activity with positive reinforcement. Just as drug activity is most dangerous when used to alleviate pain, social media use is most pernicious when used to assuage negative emotions such as boredom, loneliness, or fear of missing out.

Social media addiction leads to common symptoms that we see in severe drug addicts: a preoccupation of the mind when not using it, mood modification during use, tolerance to reward mechanisms, conflicts with others because of frequent use, and apparent withdrawal symptoms when attempting to quit (often leading to a relapse). Most importantly, this can lead to prolonged affects on the human brain. Dar Meshi, a renowned neuroscience researcher at Michigan State, found in a 2019 study that excessive social media users face similar deficiencies in making value-based decisions to people struggling with substance abuse. On top of that, several studies have shown an undeniable correlation between frequent social media use, negative mental health, and low self esteem.

In the end, what does this all mean?

Should we stop using social media all together? I don’t think so. Social media can have immense benefits when used correctly: growing entire businesses, marketing, connecting with the world, staying up-to-date on news, etc. However, at a time in our society where depression, poor mental health, and suicide are rampant, I do believe that it’d be injudicious for us to not have awareness over the negative aspects of these habit-forming products. It’s imperative that we advocate for ways that our technology will be used for good.

How social media companies have captivated the attention of an entire generation. was originally published in Digital Diplomacy on Medium, where people are continuing the conversation by highlighting and responding to this story.