Hi, I’m Jaryd. 👋 I write in-depth analyses on the growth of popular companies, including their early strategies, current tactics, and actionable business-building lessons we can learn from them.

Plus, every Friday I bring you summarized insights, frameworks, and ideas from the best entrepreneurs, writers, investors, product/growth experts, and operators.

Subscribe now

Happy Friday, friends! 🍻

Welcome to the 293 new folks who joined us this week. 🙏

Like always, there are an infinite number of ways I could start todays post. And, while waiting for my tea to brew last night, this article from CNBC made it clear…

Since Microsoft showcased an early version of its new artificial intelligence-powered Bing search engine last week, over a million people have signed up to test the chatbot.

But beta testers have quickly discovered issues with the bot. It threatened some, provided weird and unhelpful advice to others, insisted it was right when it was wrong and even declared love for its users. Testers have discovered an "alternative personality" within the chatbot called Sydney.
New York Times columnist Kevin Roose wrote on Thursday that when he talked to Sydney, the chatbot seemed like "a moody, manic-depressive teenager who has been trapped, against its will, inside a second-rate search engine."
Sydney later tried to convince Roose that he should leave his wife for Bing, and told him that it loved him, according to a transcript published by the paper.

And that NYT article is titled: Bing’s A.I. Chat: “I Want to Be Alive. 😈”

Lol…things are weird in the world right now. 🫠

white christmas black mirror gif | WiffleGif — “White Christmas”, Black Mirror

Anyways…

Here’s what we’ve got this this week:

Let’s dive in.

Context, conspiracy, and the bullshit asymmetry principle

Brandolini’s law (AKA the bullshit asymmetry principle) says that “The amount of energy needed to refute bullshit is an order of magnitude larger than to produce it.” 👇

If you’ve ever tried to rebuttal some baseless statement with logic — you know how true that is. Brandolini suggests that “an ounce of prevention may really be worth a pound of cure.”

While Andrew Bosworth (Boz) is busy running all things technology across Meta, he also occasionally shares his thoughts on business, life, and the intersection of the two. In his essay Context and Conspiracy, he shares 3 suggestions on how we can make sure what we say/write/do isn’t taken out of context — saving us the effort of having to explaining ourselves.

I’ll let Boz tell you…

Write Defensively
After a post of mine leaked well after the context around its authorship had collapsed I wrote an internal post encouraging my colleagues not to write less, but rather to write more. If you think something is important enough to contribute as a comment or a post or a slide deck you have to really commit to it. Include all the context necessary to survive the passage of time. As the old adage goes, don’t write anything you would be ashamed to see on the front page of the NYT. I think we can all agree that has never been more relevant.
If you are skeptical that writing more can work, I will note that I put it to the greatest test available. I decided to rewrite that shameful post that made me the main character on the internet for a day. I posted it internally and then publicly on my blog. In the three years it has been public it has not generated any meaningful concern because it is a more complete thought. It is what I wish I had written the first time: https://boz.com/articles/growth.
If this seems like a tremendous effort I can assure you it becomes natural much faster than you might expect. I also think my writing stands as evidence that it can be done even for those who are busy and without coming off as overly corporate. Few would accuse me of being watered down! It is just about adding more context to the writing rather than relying on the reader.
Read and Comment with Humility
At a large enough scale, even people who are working full time on something can’t likely see the full picture. I know I often can’t. Large organizations have not only their own weather but their own microclimates and sometimes they form on opposite sides of the same problem. Leadership uses the various points of view to triangulate on the entirety of a problem and understand it. But for any individual who is working one side of a problem they can understandably feel like they are more informed than they are.
Instead of making comments, focus instead on asking questions. Misinformed comments aren’t just liable to be taken out of context they may also lead others to follow. We have learned on the internet that it takes relatively few people to mistakenly believe they know the answer to assume a conspiracy, when the much greater likelihood is that they just don’t know enough.
I admit this is a hard trap to avoid for me personally. I think of myself as a smart person who can do my own research and understand something. But I suspect the savvy reader can re-read the last sentence and spot the error. I can be smart and do my own research but I must also maintain humility about the veracity of my own conclusions.
Manage your Audience
Another way to avoid context collapse is to ensure it is going to an audience who has context. Sure you can slap “draft” on your findings and send it wide but that has turned out to be a meaningful risk nonetheless. In addition to being a meaningful risk, it has also turned out to be of marginal value.
At any meaningful scale, even well intentioned people internally rarely share full context. The time it would take to get to fully shared context at scale would leave no time to build things. So we need to find ways to divide the work up and focus our communication on those whose experience and expertise allows them to understand it.
By ensuring not only that your message includes all necessary context but also that it is targeted at an audience capable of understanding it in perspective is critical.

For more of Boz’s thoughts, you can check out this blog here.

Creating products that stick

As Andrew Chen says — it’s never been easier, cheaper, and faster to build a product and get something off the ground. In a previous 5-Bit, we unpacked this in more detail.

But, while it’s easier to build something, Greg Isenberg notes that “it’s never been harder to create a product that sticks with people”. Adding that, “you’ve built a product that sticks when your product stands out and is overwhelmingly valuable”.

And with all the noise in the world and the multitude of solutions for almost any problem, I agree.

Fortunately for us, Greg, who’s an expert on community-based products (having advised Reddit, TikTok) has a framework for how he build products to stick.

It’s a framework I covered when analyzing Notion’s growth, but let’s give it some more attention:

The single-player vs multiplayer framework:

A single-player product is one where you can get value all by yourself — no other network of friends or users needed. The more sticky your single-player experience is, the more trust you build with your users, and the magic moment for these products is when you see yourself making it a part of your daily or weekly habits. Think to-do lists.
A multi-player product is pointless if you’re by yourself. You need a group of friends for Whatsapp to work. That’s why Discord (or Slack) have intricate onboarding flows for new admins to invite their friends/communities into the product. The aha! moment happens when people find an active community there. Think Reddit.

Now, to create a sticky product — you want both.

The single-player mode helps you activate a user and gets them to stick around (retention). And the multi-player mode unlocks a new type of value, drives growth via invites, and brings you network effects. This is where you eventually want to get to.

Examples of companies nailing SP and MP are Notion, Figma, and Trello. Which makes sense, because productivity/creation tools work well alone and with collaboration.

Okay, which one do you start with?

Greg… 👀

First you start with the community. You find an underserved niche that you feel you understand them and speak to them in a way that resonates.
Then, realize that the holy grail is combining single-player and multi-player. For example, I first heard about Figma from a designer that had switched from Adobe XD. He had built significant trust with Figma as a single-player, recently moved to multi-player and the guy was literally screaming from the top of his lungs on rooftops about how amazing Figma was. That’s the power of single + multi in a nutshell.
However, it is usually easier to begin with single-player. Building multiplayer is more complex. You need to have liquidity (ie: friends needs to be online for the ah-ha moment to be most fertile).
I believe multiplayer mode is the most misunderstood part of the internet today. Way more upside. That’s why you’ll see a multiplayer-first business like Figma sell for $20b.
TLDR;
If you’re building a new product, ask yourself, what part of this is single-player, what part of this is multiplayer? Why does it make sense to start single versus multi?
It’s an important question because if you nail it, finding product/market fit becomes easier and customers find you instead of you finding customers.
— Greg Isenberg

Definitely one to add to your product framework toolkit.

If you’re into building internet communities, checkout Greg’s newsletter.

Unbundling, and how to verticalize Reddit

Every great empire has eventually disintegrated and fragmented into smaller states. Some of those states became empires of their own, sometimes larger than the predecessors.

In business, this is equally true…

One of the holy grails in the newco world is to build out a digital platform that successfully serves the needs of a broad number of adjacent verticals, and become the definitive platform in its space. We know the now-canonical early examples of this: Amazon, eBay, Craigslist. And we also know that once that holy grail of a new digital platform is attained, competitors quickly come a’ callin’.
One of the most effective forms of that competition often comes in the form of newcos who aspire to take chunks out of that emergent platform by better addressing the needs of a specific vertical within that platform—by creating a user experience or business model that’s much more tailored to the unique attributes of that vertical.
— via a16z

In 2010, Andrew Parker wrote a now classic post of this dynamic in action. It was about the unbundling of Craigslist.

Craigslist was the empire — it was everything to everyone — and was a diamond mine of startup ideas. In a wonderful visual, he illustrates the opportunity in carving out niche products from broad horizontal networks — and the value in specializing for a certain group.

If you add up the value creation of the vertical companies (ie: Airbnb, Zillow), it’s far greater than the value of Craigslist. Airbnb alone is almost 30X bigger.

Verticalizing is clearly very good business.

In another post by Greg Isenberg (who as you can probably tell, I’ve just discovered), he talks about the unbundling opportunity of Reddit.

Reddit is one of my favorite places on the internet. It's like an early version of a metaverse, where people come together to create interconnected worlds, each with its own culture.
That interconnectedness is what makes Reddit great, but for the system to work, each community has to conform to the one-size-fits-all mold of a subreddit. One size fits all, but it doesn't fit anyone particularly well. That creates unbundling opportunities.
There are hundreds of wonderful startups waiting to be built using this simple strategy: create a product that serves the unmet needs of an individual subreddit.

In an older essay of his, he lays out a step-by-step process for carving out your own vertical business from Reddit.

Reddit is not only extremely horizontal (i.e communities for everything), but each vertical also comes with insight into the audience, what they think about and like, and things they’re complaining about.

This makes it a goldmine for startup ideas, with Greg’s framework as “your pickaxe”. I’ll summarize. ⛏️

Step 1: Find a Subreddit

First, go and find a subreddit. For non-Redditers, that’s means a community like r/gaming.

You can use a tool like redditlist.com to see rankings based on activity, subscribers, and growth rates. The goal is to find one that checks these boxes:

It's growing fast. Look for 40%+ in past 6 months.
It has a critical mass of subscribers. Look for communities 50K+.
You have a competitive advantage there. This means leaning into your strengths (see “how to make luck” above) and choosing a subreddit were you have experience/knowledge.
You're passionate about it. Find a vertical you actual have an interested in, otherwise you will have a hard time connecting with people and caring to solve a problem for them.

Step 2: Join the Subreddit

Immerse yourself like an anthropologist in an isolated jungle tribe. Learn their language and their rituals. Their memes are your memes now.
Aim to spend 1-2 hours everyday on the subreddit.
Take notice of which posts resonate with the community (pro-tip: some of the best posts have lots of comments but not a lot of upvotes). Ask questions you're interested in and answer questions when you have helpful knowledge. Share useful links.
The most important thing is to be authentic. Engage like a human - don't spam or use marketing speak. At this stage, you aren't looking to establish a brand or sell anything.

Step 3: See What They Want

Once you’re inside the community — your only goal is to be curious and learn.

What problems do people have, what do they care about?

At this stage, some questions to orient yourself around could be:

What recommendations do they ask for?
Do they keep complaining about something specific?
If you were trying to achieve their goals, what would make it easier?

Step 4: Create a Closer Space for Communication

This step of Greg’s framework I find the trickiest. In short, he suggests making your own space for this community (i.e on Slack or Discord) so you can have more personal and direct interactions.

But on Reddit, this is very hard to do. Moderators are extremely strict about any self-promotion and anything too abrasive and/or sales-y will get you banned.

So any steps to try bring a Reddit community into your own space is a delicate art of gradually adding value and not selling.

For example, start a conversation about a theme you’ve identified and make sure the point of the post is to keep the discussion on Reddit. If you have something substantial posted and a soft link at the bottom, you should be okay.

Step 5: Build Something for Them

Now that you've gotten to know the community and can talk to members directly, you'll be in a good place to build something for them.

Figure out what to build
Figuring out a community's needs is the hardest part of knowing what to build, but you've already done that. Next, brainstorm ways you can help with the needs you uncovered. Try to think of the smallest thing you can build that will solve a problem for them.
You're still exploring, so think of this as setting up camp. You can start developing the city after you find out if the land is farmable.
Build it
Once an idea has come into focus, get building. (I won't get into how to build product in this guide, but here's an awesome overview by Michael Seibel if you're interested. Late Checkout also occasionally partners with leading brands to do this)
This is where the space you created becomes really valuable. Share what you're building and ask for feedback, then incorporate that feedback into the product. Repeat!
Launch it
You can use the same strategy you did in Step 4 for spreading the word.

Step 6: Build on Top of What You Built

At this point, you have two very valuable resources: your space and your product. Those are the twin boosters that will get your spaceship to escape velocity.
You now have everything you need to build a big business.
To do that, find opportunities to build on top of your product. There should be no shortage. Ask yourself how it could evolve into a community or marketplace. Build in social components. Add on features to cover more of the community's problems.
Iterate, iterate, iterate.
It's also helpful to use your predecessors for inspiration. Try to reverse engineer how Discord evolved out of the League of Legends community, or how IndieHackers evolved out of the bootstrapping community.
If you pay close attention to the community, they'll show you the way.

For further reading on unbundling, check out:

Matching Price to Value—3 Lessons in Monetization

According to Naomi Pilosof Ionita (partner at Menlo Ventures and expert on early-stage B2B SaaS), monetization is the most underutilized lever for growth in SaaS.

Typically, when you hear the word growth, it’s referring to user acquisition.
While acquisition is a critical part of the flywheel, what many companies don’t realize is that monetization can actually have the biggest impact on growth.
A study from Price Intelligently, a software engine for pricing, looked at more than 500 SaaS companies and analyzed what a one percent improvement in acquisition, retention, and monetization did to their bottom line. Their results were astounding—the impact of improving monetization was 4x that of acquisition in building business value.
— Naomi Ionita, Matching Price to Value—3 Lessons in Monetization

In her essay, she shares three lessons on how to create value and maintain trust with your customer while driving revenue, AKA, to better monetize. 💸

#1: Know who you’re building for and segment users by what they value and what they are willing to pay

Knowing who you’re building for means understanding who your customers are, why they pay you, and how much they’re actually willing to pay.

From here you can segment your pricing and do all sorts of interesting things, like upsell them in different ways and price based on different value points.

A great example is Uber: “Everyday, everywhere, people need to get from point A to point B. What Uber has done is segment those offerings, differentiating how one gets from A to B. If you’re on a budget and have time to spare, choose Uber Pool. If you’re in a rush, pay more and opt for Uber X. If you want to arrive somewhere in style and aren’t price-sensitive, choose Uber Select or Uber Black. Uber maximizes revenue by effectively matching price to value.”

A key point she raises is that what your charge needs to be based on “perceived value”, i.e the value the buyer sees based on their needs, and that “by better matching price to (perceived) value, you’re less likely to leave money on the table.”

In other words, do your customers care about all of your 20 features, or do they really just care about 3 of them? The argument (which has been proven in Madhaven’s book) is that you can be a lot more successful if you focus your pricing on emphasizing just the 3.

Perceived value > value.

#2: Balancing sticks and carrots

Rifting off that example of 20 features above, where 3 of them are core.

Said differently, some are sticks and others are carrots.

Sticks represents the main reason people use your product. It’s your value metric, like storage for Dropbox, or seats for Slack.

Carrots are premium features that boost the core value, but are not the value in-and-of themself. With Slack, a carrot would be custom sidebar sections, as it’s something a user doesn’t have in a free or lower-priced plan, and must upgrade to get.

Sticks are carrots work together to drive both conversion and expansion.

Sticks get users to upgrade beyond their designated quota limits by showing contextual upsells (“Your storage is full. Upgrade for more space”)—these are highly effective since they catch users in a high-intent state (“but I need to store this file!”) and no additional education is required since users have already been trained on the core product offering. Carrots typically require additional work around awareness and are dangled in front of users as a way to drive upgrades.

#3: Improve monetization with predictive data

#1 and #2 above are the easy and medium difficulty takeaways here. This one is pretty sophisticated, but data is deeply powerful and if you can get this right — you’re likely winning the monetization game.

Simply put, this is smart pricing driven by data. Naomi can explain:

We started building propensity models to understand which variables were not just correlated, but actually causal to downstream monetization actions. We then used the most predictive variables to create a scoring mechanism for our customers in order to segment them based on propensity (i.e., very likely, likely, not likely, and very unlikely to renew).
The next step was to operationalize these segments. For customers who had a low propensity to renew, we increased perceived value by reinforcing the benefits of core features and building awareness of premium features that were likely relevant to them. Beyond increasing value, we also reduced the price by offering them targeted discounts and promotions.
On the other end of the spectrum, we looked at our customers who had a high propensity to renew. By deepening our understanding of how they were using the product, we unlocked an opportunity to not just drive renewal, but to drive an upgrade in the process. When we predicted that subscribers would benefit from the features packaged in a higher-priced plan, we offered contextual upsells in their renewal experiences.
By understanding a customer segment’s needs and willingness to pay, we could effectively tweak different knobs to drive perceived value and revenue, all in an automated way across the product experience and related marketing channels.

That’s very cool. 🤓

With these three tactics — hopefully you can find the best balance between price and value.

And, to see an excellent example of a company that nailed pricing, checkout our deep dive on Superhuman if you missed it. Or you can just jump right to the section on pricing here.

For further reading:

Find anything valuable so far? ❤️

Share

And our last one for today…

Increasing your surface area of luck

Samuel Goldwyn famously said, “The harder I work, the luckier I get.”

A simple example which I think illustrates that well is a tweet going “viral”. Somebody who joined Twitter yesterday is unlikely to have their first tweet get a ton of traction, but the person who has been putting stuff out there consistently for months may well get “lucky”.

In other words, luck is a function of effort. But saying “work harder” is extremely vague, broad, and useless. So to help answer the question of “how to create luck” in a more tactical way, I’ll summarize a video by Dalton Caldwell, partner at Y Combinator.

Note: This is in the context of early stage companies, but can be abstracted to other things too.

Move much faster than everyone else. Startups moving fast have more opportunities to speak to customers, ship tests sooner, make iterations, and even cycle through different ideas. Fast movers create the most surface area to get “lucky”.
Expand your network of “interesting people” and “learn from as many interesting people as you can”. I personally don’t like the term “interesting” to describe people, as I think everyone is interesting. My uncle once said to me (although I don’t know if he came up with this), “the smartest man in the room is the man that can learn from anybody”. So, rephrased, build a network. Having people you can ask for advice, ask to look at a prototype, or give you a tweet down the line are all things that help you get lucky.
Do things that make you uncomfortable. Go and talk to people about your idea, share loose prototypes knowing you might get bad feedback, and work outside your comfort zone. It’s where good things tend to happen.
Double down on your strengths. Usually, leaning into what you’re good at and what you know gives you a bigger surface area to find success. Take this newsletter, I chose to focus on something related to product/growth and play to my strengths. The flip side would have been exploring something like healthcare. Maybe I’d get lucky, but I’d rather bet on (1) what I know, and (2) what people in my network know me for — 2 good chips.
Bonus: Know how to make things yourself. If you’re able to design and put together a prototype without relying on anyone else, you’re putting yourself in a position to move faster and show people more things. If you can do some basic code, even better.

You can watch the video here.

And that’s a wrap for this week. As always, I really appreciate you reading and spending your time here. Every tweet, DM, share, and kind word sent my way means a lot. 🙏

If you’re enjoying what I’m putting down (which if you got to this point, I guess you are), you can keep on supporting HTG by liking, subscribing, sharing, and all that good stuff that I feel very awkward asking about. 🙃

Subscribe now

Share How They Grow

Have a lovely weekend, and I’ll see you on Wednesday for our next deep dive. I can guarantee you won’t guess what company we’ll be looking at.

Clues: A duck. Ouch! An airport. Nostalgia, and Smørrebrød 🤔

See you then!

— Jaryd ✌️

+ Other interesting things if you have time

[Read] Why Building Your Startup's Brand From Day 1 Matters
[Read] 10 Lessons From My Journey as a Creator and Product Leader
[Listen] The 1000x Developer (via a16z)— does it make sense to learn to code?

I was inspired by Lenny Rachitsky to start writing, and it’s been the best thing ever. If you’re feeling inspired, it’s very easy to…

Start a Substack

How They Grow

5-Bit Friday’s (#14): Snackable insights…