BRAINDUMP ON VIRAL LOOPS

Everything we learned (and forgot) about viral loops from Web 2.0

Hey all — recently I wrote a three-part brain dump on everything I remember from building viral loops in the Web 2.0 days and some of the ways in which all these ideas have evolved in the mobile era and beyond. I posted it as three parts on X and brought it together as one essay with some minor formatting changes here and there but otherwise preserving it in its entirety. Hope you enjoy! -A

BRAINDUMP ON VIRAL LOOPS #1

The golden age of Web 2.0 (~2005-2010) was a special time for viral products, which were systematically engineered to reach millions of people. Back then, people were building the first versions of things we take for granted: Social networks, user-generated platforms, collaborative workplace products, messaging apps, etc.

During this time, the industry developed a comprehensive and systematic understanding of creating viral loops. There was measurement, A/B testing, and equations that needed to be satisfied. Products optimized and optimized, engineering virality. And some of the most successful of these products eventually grew to billions of users and became recognizable names.

Then all of a sudden, it ended.

And funny enough, the people who were successful at creating viral products all became billionaires, or if they didn’t create successful companies, many became FAANG executives or VC investors. Eventually, all the knowledge around building virality was basically lost. This happened as Web 2.0 ended and the mobile era began.

I thought it might be nice to write down a full brain dump of everything that I learned about viral loops during this period, because it does continue to be relevant, although the mechanics and tactics around it keep shifting and changing. Yet the overall theory is still the same, and I think it can be extended into future technology platforms and marketing channels. It also has a high degree of applicability to this era’s Product-Led Growth, all the sharing flows of gen AI apps, and referral programs that still exist within every marketplace/ecommerce product.

With these notes, I plan to cover a bunch of different ideas and topics:

• Simple viral loops that work, but then quickly disintegrate.

• Viral factor and how to calculate it.

• How to actually break down a viral loop step by step and optimize

• What increases the viral factor, and what decreases it?

• How to improve viral factor

• Case studies like content sharing, invite flows, referral loops, collaboration loops, etc.

• How retention drives viral growth.

• How do you incorporate retention into your viral factor calculation

• Why new users invite more people than existing ones.

• How to think about word-of-mouth vs. engineered virality.

• How to think about paid marketing, SEO, social media, and the effects of other channels on viral factor.

• Why mobile eventually killed the classic viral loops.

• What kinds of viral loops work in the contemporary era

• ... and much more

If you have thoughts and questions please ask. Hopefully I’ll eventually turn this into a big PDF or something that people can read if they’re interested. Or maybe a deck if we have some kind of long-form text to PowerPoint deck gen AI tool one day that’s actually good :)

Viral growth as an equation
Usually when most people talk about viral growth, they mean something silly (and non-durable), like tweeting a really cool video that then gets shared a bunch of times and drives traffic to your product. That’s not what I’m talking about. Instead, the viral loop I’m talking about is designed into a product, with invite features, tagging, referral links, and so on.

These types of viral loops have interesting properties:

• it is measurable and can be tracked

• it can be improved with product decisions

• the math applies to any form of product-driven virality (invites, shared content, referral programs, etc)

The first thing is trying to measure viral factor. The viral factor is a simple ratio -- the usual thought experiment is that if you bring a 100 users into your product, and those users invite and eventually sign up 150 users, then those will sign up 225 users and so on. This is a viral invite flow, and your viral factor in this case would be 1.5 (rarely seen in the real world). On the other hand, if your 100 users end up signing up 50 of their friends, who then sign up 25, then your viral factor is 0.5. When it’s <1 then your viral loop eventually stops working.

Very precisely, the viral factor is a RATIO. The denominator is the # of users within a time-bound cohort (like everyone who signed up from X date to Y date), and the numerator is how many users they eventually signed up via the viral loops from that cohort.

Let’s use an example to make it concrete.

Case study: Content sharing
A very common form of viral loop has an app that lets a user create something really cool (maybe that’s with AI or photo filters or otherwise). After the user creates something, they want to share it, so naturally a link is provided. Some new users receive this link, view the content, and some smaller % of these folks sign up to make their own content. At its core, this is what the new generative video and photo apps do (like Sora). But this is also the same viral loop that made Instagram successful with photo filters or what’s made blogging popular as well. After all, when you create some type of content, you naturally want to share it with people, but those people naturally may want to participate as well.

So conceptually, you might understand the smile loop, but the question is, what can you do with it to optimize?

To make it systematic, you want to track it. When someone makes a generative video, you want to track things via the URL structure:

productdotcom/vid/[video ID]?sharer_id=[sharer]

(of course, your userID is encoded as the sharer id whenever you share) Now when you share it out via email, or messaging or whatever, if an invitee gets the link, watches the video, and then signs up, you should then store the sharer’s ID alongside the user’s signup row. So imagine you’d store the usual stuff:

• id

• email

• password

• name

• sharer_id <-- this is who caused this invite to happen

Now when you look at your rows and rows of users in this table, sometimes the share ID will be blank if that user just showed up and signed up on their own. However, if they are part of a viral loop, then their share ID will point to some other user that already exists in the table.

To calculate viral factor, you then take some cohort of users, for example, the ones that joined 3 months ago. Let’s say that there are 100 of these guys. Then, what you do is you grab that list of IDs and you ask the database how many times do those IDs show up as share IDs in new sign-ups that happen afterwards? Let’s say that’s 50. Then your viral factor at that moment in time is 50/100=0.5.

What about all those users with blank sharer IDs? I sometimes think of the users who just show up as “Gen 1 users” or “onramp users” and you should just discard these guys as part of your viral factor calculation. Instead, what’s interesting is to compare the ratio of Gen 3 and Gen 2 users. Or really, Gen N+1 and N, as long as N isn’t 1. It turns out this ratio actually tends to stabilize pretty nicely over time.

The big questions
Do this with your product, then you’re bound to ask, “Wow, this is cool, but how do I make this number go up?” and in particular, “How do I make this number go over 1?”

This is the right question to ask. The minute you realize these dynamics, you’ll realize that you can make product changes to increase that viral factor so that your product becomes more viral. For example, maybe you should ask new users to invite when they first come in, or you should make it easier to invite people because there’s a link that you can copy and paste or some other mechanical change that increases viral factor. This is one of the core insights that you can measure viral factor and that you can actually make product changes to increase it.

Because you can track and calculate this ratio, this means you can stick it onto a dashboard as well. As I described, it’s easy to do track, as long as you encode the user id into the URL that’s shared. Of course, as soon as you’re able to calculate it, the next thing is then to be able optimize it because then you can run A/B tests to see if more people share, and do they do more of them? Do they share with more people? And for the people who receive a growth link, what percentage of them convert, and how you make that conversion higher. This is sort of exists in the cookbook of all the different ways that you might be able to optimize your viral factor.

The final aspect of the viral loop idea that’s powerful is that it’s actually very generalizable. Which is why this is all so powerful for the modern era of Product-Led Growth. As long as you’re talking about being able for a group of users to somehow generate another group of users, then of course that growth process might be because they’re inviting other people. It might be because they are sharing a content link that then causes someone else to sign up. Or it might be a referral program or many other tactics that cause one user to invite another. This means that at the core, there ends up becoming a lot of applicability across marketing channels and strategies.

Alternative ways to calculate viral factor
When you ask the Internet how to calculate viral factor, it gives you something slightly different, which looks more like this:

VIRAL FACTOR = # INVITES x % CONVERSION RATE

This definition has been floating out there for a long time. However, it’s flawed because not all viral loops exist as a type of invite. Instead, we see sharing flows, collaboration flows, referrals, and many other variations. I also think that although it’s mechanically correct, because increasing the number of invites and increasing the conversion rate naturally does help your viral factor. It does not capture the fact that really what you want to know is this ratio between two cohorts of users.

Viral loops and spam loops
There is also a dark side of simply trying to increase these variables as well:

The more simplistic version of this equation pushes you towards thinking about how to be as spamming as possible. How do we get users to invite as many of their friends as possible? This sort of incentivizes pushing users to invite pushing them to invite a lot of people, and then for the invites to be spamming in such a way that there is a high conversion rate.

Going back to the Web 2.0 era, this actually was the way that people tended to think about virality. If you’ve built the bare bones of a social network as folks at Bebo, Tagged, Bebo, Hi5, MySpace did, you were generally growing your network based on getting people to invite their friends via email. In the early days, you’d actually ask people to type in their friends’ email addresses, and usually they’d type in like 5-10 emails since it was tedious. It turns out that if you could get them to import their address book from Hotmail or Yahoo Mail, by asking them for an email/password combo, then crawling their inbox, you could push the number of invites to 200+. Of course whatever viral factor you had, a 20x increase would often put you over the top.

The downside is that you would generally drive down your conversion rate because you were emailing dead emails. Eventually, all of the email providers would start labeling you as spam. But for ~10 years, the glorious email invite loop worked, and many large products were built -- including ultimately the big winners, Facebook and Linkedin.

Chain letters and why spam loops fail
You might know that over a hundred years ago, there was a concept of a chain letter where people would write actual snail mail letters to each other. Each chain letter would have a list of other addresses, and it would tell you that if you mailed a nickel over to them, and then also added yourself to the list, then eventually everyone would get rich because you would be receiving an exponentially higher amount of nickels from people over time. This was actually such a big deal that eventually the post office had to make it illegal since it swamped all of their delivery. But of course, mathematically, chain letters are going to fail.

The reason is simple, which is saturation. If you have a viral loop and you invite 200 people on each go, eventually after a bunch of generations, you’re going to completely saturate your target market. If you do the math, once you pass 5 rounds of 200+ invites, then you’ve already hit every human on the planet and more. So eventually, naturally, you’re just hitting the same people over and over again. Naturally, the response rate is going to go down. Why? 1) You’ve already signed them up, in which case that invite can’t convert. 2) Or, alternatively, they got invited before, but they’re not interested and they don’t interact. Either way, the response rate tends to decrease over time.

If the response rate goes down and your product is not retentive, then mathematically all you’ve done is created a huge spike of new users which looks good for a while, but then that lack of retention means that your total number of active users ends up being a tiny fraction of new users.

This is why some highly viable products end up jumping the shark. In the past, I’ve listed out some of these metrics to indicate that a product will be sticky:

1) cohort retention curves that flatten (stickiness)

2) actives/reg > 25% (validates TAM)

3) power user curve showing a smile -- with a big concentration of engaged users (you grow out from this strong core)

3) viral factor >0.5 (enough to amplify other channels)

4) dau/mau > 50% (it’s part of a daily habit)

5) market-by-market (or logo-by-logo, if SaaS) comparison where denser/older networks have higher engagement over time (network effects)

6) D1/D7/D30 that exceeds 60/30/15 (daily frequency)

7) revenue or activity expansion on a per user basis over time -- indicates deeper engagement / habit formation

8) >60% organic acquisition with real scale (better to have zero CAC)

9) For subscription, >65% annual retention (paying users are sticking)

10) >4x annual growth rate across topline metrics

These aren’t meant to be exhaustive, but they are all strong indicators that if a product can generate a lot of viral users, there’s enough there that it’ll stick. If not, then eventually the spike will go away.

The fact that you need product-market fit ends up being one of the reasons why sometimes you hear about highly viral apps that then disappear from public consciousness. You can actually have a very high viral factor or high word of mouth and sometimes you can even engineer it. But it’s almost like having a big bang launch or having a Super Bowl commercial or something like that. You get a huge spike and it’s great for a while, but inevitably, unless your ratio of active users to sign-ups is very high, it means that you’ll eventually lose them.

This was ultimately the outcome of many of the apps that were created during Web 2.0 or during the Facebook platform era. You heard about a lot of virality and potentially millions of users or even hundreds of millions of users, and sometimes this happened quickly. But a lot of this were people engineering viral loops. And as those users left the app because of a lack of stickiness, these viral loops could be engineered to try and reacquire them over and over again. The vast majority of these viral apps did not actually become successful businesses, which is why- Although I remain very interested in the ability for teams to create growth hacks. Of course, I also care a lot about their ability to actually retain users.

---

BRAINDUMP ON VIRAL LOOPS #2

You’ll often notice that very viral products tend to fall into one of two buckets:

Category 1:

very simple apps that do one thing, and that one thing is highly shareable or pushes out a lot of invites. Overnight successes. (think gen AI creative tools, but also Instagram/YouTube back in the day)

Category 2:

highly sticky products with deep functionality, and some sharing capability -- tends to grow slowly, but consistently. (think Figma, Slack, early Facebook, etc)

These two types of loops are very different.

Apps in category 1 have tight, simple viral loops with high conversions at every step, so that there’s almost nothing to do in the app other than creating viral content. The successful version of this is something like Instagram in its initial form with photo filters and sharing to Facebook, and not much else. The spammy bad version of this includes the litany of quiz apps and anonymous/secret apps that failed over the years. However, they have a huge advantage of being easy to create, and often in the early days of a viral platform, these are often the most successful ones.

The second category of apps ends up being complex products to build with high retention and a little bit of sharing functionality. The plus with these is that every time they acquire a user, they’re very sticky, and so, if you put the sharing functionality in front of users over a long period of time, eventually the viral factor will add up. However the growth doesn’t look quite as explosive as category one.

This type of loop is interesting to examine in the current era of AI apps because they days have viral loops very similar to Instagram, YouTube -- category 1. They have some of the same strengths (simple, high growth) but also some of the same weaknesses (spiky, low aggregate user retention). Maybe this is history repeating itself?

Viral content creation loops, step by step
These types of simple content creation loops work like this:

• you see something cool online

• then you watch

• click a link back to the creation tool

• then you try out the tool yourself. Cool!

• you share it on social media (or chat or whatever)

• more people see it

• ... and the loop repeats

I’m describing this very generically because you can imagine this being a video hosting service like YouTube, or it could be a cool photo with photo filters, or it could be an anime-style profile photo, or an AI-generated video with your friend’s face in it. In any of these cases, ultimately that takes advantage of the psychology of being creative. That is, after you create something, you want to share it with other people.

This type of loop is easy to understand, but it’s not just qualitative. You can quantify it too. Just break down each of the steps into percentages. For example, after someone views a video, there will be a clickthrough rate on the percentage of people end up clicking to your website. Then, after they go through the website, you’ll see what percentage end up creating content themselves. In fact, there are drop-offs at every single stage of this except for one stage: once you share it on social media or messaging, how many people end up actually seeing the whole thing. You might visualize the whole thing like this:

• you see something cool online

• then you watch (50%)

• click a link back to the creation tool (10%)

• then you try out the tool yourself. Cool! (20%)

• you share it (50%)

• more people see it (X people see it)

• ... and the loop repeats

The basic math on this viral loop is that if 50% * 50% * 20% * X > 1, then the loop will go viral. Of course this is hard because the first few terms together multiply to 0.005, so you’d need at least 200+ people to see it to go viral. But all of these numbers are highly sensitive.

When these loops work, they are magic. But funny enough, when a product is only slightly viral, you can barely tell. If you are attracting 100 users/day based on organic word of mouth, and you have a viral factor of 0.1, then the multiplier effect is only 1.11x. So your 100 users end up inviting 10 friends who invite 1 more, so that’s only a small bump. Can you really tell a small bump like that if your organic signups go up and down 10% each day as well? Not really. You really need high viral factors, like >0.5 to feel the difference. At 0.5, then 100 users will sign up 50 that will sign up 25, and so on, and that gets you a 2x boost. Now we’re cooking! This ends up being valuable because it means whatever you are spending on ads, you get 1 user free for each one that you buy. So even if you’re not viral, it’s helpful.

You can see what the multiplier effect is at different viral factors:

• viral factor of 0.1 is a multiplier of 1.11x

• 0.25: 1.33x

• 0.5: 2x

• 0.75: 4x

• 0.9: 10x

For the nerds, the math here is that the infinite expansion of gen 1 inviting gen 2 inviting gen 3 etc is 1/(1-v) where v is the viral factor. So if you stick 0.5 into that, you get 2x, which tells you that if you sign up 100 users with a viral factor of 0.5, you’ll end up with 200 total at the end -- aka 100 are free viral users.

As a result of the above, you are very much incentivized to get your viral factor above 0.5. Everything underneath that number almost isn’t worth it. Thus, building your app to be highly shareable, very simple, such that there’s almost nothing to do besides creating+sharing, is a clear way to go about it.

And it can be glorious when it works. But we’ve seen that there’s often spikes. One day it’s trending on social, the next day your traffic has crashed. (Let’s hope you pushed enough people through a subscription flow in the meantime, so that at least your ARR is sticky!)

Why the numbers go down over time
The natural direction for these forms of viral loops is that they degrade in performance over time. There’s a bunch of reasons:

• novelty effect creates higher metrics early, worse metrics late

• market saturation is inevitable

• virality runs on pre-existing platforms, and the platforms might get mad

There’s just a novelty effect. When there’s a new class of products like content creation tools with generative AI, users tend to respond to them at very high rates of adoption. A higher percentage want to watch them, a higher percentage want to try them more. And when they’re shared, a higher number of people want or are receptive to seeing the content. We saw this in the early days of 2D image generative AI where people would share amazing AI photos (even when people had six fingers). But when was the last time you shared a 2D image generated by AI? The novelty is worn off, and so you need a much higher bar before you’ll engage in that same behavior. (I won’t rehash The Law of Shitty Clickthroughs here, but it’s all the same idea)

And of course, as the viral loop becomes more successful, it may burn through the entire market of target users and the saturation rates will put pressure on all of these metrics again decreasing viral factor. At the beginning of a loop’s life, the entire market is addressable, so if you are inviting 10 people into the app all 10 are valid targets. But if you’ve saturated your network, maybe out of the top 10 people you’d invite, 5 people have already joined, so that only the 5 are valid, which drives down your viral factor by half. And these final users tend to be lower quality as well, since they are late adopters. Back in the email invite loop days, you would find that if people imported their email addressbooks, you might initially get ~200 contacts out of it. If you emailed them, you might have like a 10-15% open rate, and a 5% clickthrough rate. But as your product grew to many tens of millions of users, then peoples’ contacts typically got smaller -- eventually 150 then 100 -- simply because you were reaching into the edges of the global social graph. Popular, well-connected people tended to be onboarded first and then by the end you were bringing on the introverts, late adopters, and luddites :) Not only are they less viral, but their usage/engagement tended to be that much worse.

Every viral loop runs on a pre-existing platform. Email invites run on email. Instagram’s early sharing ran on the Facebook platform. Today’s generative AI video apps publish content to YouTube/Tiktok/social. So if it turns out that people are sharing a lot of these kinds of videos with certain watermarks and link backs to the creation tool, and eventually the video platform that you’re publishing on doesn’t like this, one of these conversion numbers might drop really, really low. And then your viral factor might go from being over 1 to being under 1, and the whole thing will decelerate. The platforms often regulate this behavior tightly because they are themselves sometimes building competitive solutions, or they don’t want their platform taken over. (Just read about Zynga/Facebook platform wars back during the Web 2.0 days)

The weaknesses of hypersimple apps
Even given these factors, the hypersimple hyperviral Category 1 products can win. I think of YouTube and Instagram as examples, where at the end of the day the entire app can be described with 3-4 screens of UI. Of course in subsequent years a lot has been built out, but the simple core is still there. It turns out the minimum product to generate a ton of retention is just a few bits of UI after all, and the network effects of having zillions of pieces of content means that a small app with a deep content base can remain endlessly engaging. This is sort of the magic of networked products like messaging apps, UGC, social networks, and so on.

---

BRAINDUMP ON VIRAL LOOPS #3
ok -- so here’s the case against all the fun/dumb viral techniques you’re seeing on social media right now. When people talk about “going viral” these days the state of the art -- if you can call it that -- is a hodgepodge of shitpoasting and social media videos. This includes:

• ragebaiting and shitpoasting by startups

• beautifully shot cinematic launch trailers

• tiktok video clipping

• billboards for social media coverage

• influencers astroturfing your app

• founders turning themselves into influencers

• etc etc

These are fun, but I predict they will not stand the test of the time -- not the way that classic viral techniques like referral programs, sharing links, invites, and so on have proven themselves over multiple eras.

One quantitative way to talk about this is -- do these techniques scale signups as a function of active users?

That is, if you tracked this as a ratio:

new users / daily active users

... would you be able to scale new users as you scaled daily active users? If not, naturally, new users will stay constant and even with strong retention, your DAUs will grow linearly but not exponentially.

One core reason is that these techniques generate one-time traffic. Good for creating a single spike of new users in your app. However as you grow and need a repeatable source of user acquisition, you simply can’t repeat these over and over again. If you drop in an amazing cinematic launch once a year, people may pay attention but if you start doing it monthly and then weekly, there’s just diminishing returns. Similarly there’s no defensibility in shitposting. If you shitpost and other founders shitpost, eventually you just won’t stand out as much anymore. As your customers acclimate and become less responsive to the novelty, these techniques will fade over time.

But! But! I admit they are still useful

Here’s why. (But allow me to diverge to a short history lesson here from the Web 2.0 days)

The end of Web 2.0’s virality
In my previous brain dumps on the Viral Loop topic, I talked about how during the Web 2.0 days (~2005-2010) there was insane innovation in email invite loops, content sharing, virality, Facebook apps that went from zero to millions of users overnight, etc. Then it ended.

What killed it? Mobile did.

The “golden age” of viral loops was possible because for the first time during the Web 2.0 era, there was so much novelty in getting an email or notification from a friend who is inviting you to something or sharing a piece of content to you via an app, etc. This made response rates very high, viral factors were able to go >1, and you saw a ton of products grow overnight -- including Facebook, Linkedin, YouTube, Spotify, Pinterest, and many other products that got their start with clever viral loops.

In the golden era this is where you saw hyper-simple hyper-viral products succeed. In the earlier part of the era, you had apps like BirthdayAlarm in 2001 (h/t Michael Birch, later the creator of Bebo) which would email you about a friend wanting to keep track of your birthday (how kind!), driving you to sign up, and then flipping that over to asking if you wanted to keep track of your friends’ birthdays. Put in a bunch of emails or eventually import your email address book and you can now ask hundreds of friends. Repeat that over and over and you have an app that might be used by millions of people.

Another example -- there were loops like Plaxo in 2002 (h/t Sean Parker, later the president of Facebook among many other adventures) that started with a friend asking if you wanted to maintain an up-to-date contact about yourself. Update the contact and then it asks you to add your friends and maybe a bunch of your other friends’ contacts kicking off the loop again. Many of these were hyper simple but when they were paired with social profiles and eventually a feed, these were the underlying mechanics that eventually birthed the category of social networking apps. Importantly the Social Network category had real utility and retention and wasn’t just a viral app on its own. We’ll talk more about these dynamics in a second.

But the golden era of Web 2.0 virality eventually ended. Consumers got used to these techniques, response rates went down, spam filters kicked in, and importantly the whole world moved to mobile. When email virality was dominant, you had the ability to ask users to import their email contacts and often you’d get 200+ emails that could then be invited to the service. Before the Law of Shitty Clickthrough kicked in, you might get something like the following:

• 50% of people would connect their contacts

• 50% of people would invite friends

• 200+ invite emails would be sent

• 10% would open/click through

• 50% of people would sign up

• ... then repeat

This would often generate viral factors of >2 or more (just multiple all the numbers above).

But mobile was a whole other thing. Apple made mobile contacts widely available, but you have to invite numbers one by one. Who’s going to do that? Response rates were high, but the # of invites was small. Some folks tried to build around this, using services like Twilio to deliver invites from the server. But that led to SMS spam, and the CAN-SPAM Act and resulting millions of dollars of fines convinced people to avoid all this. Between the platform change from email virality to SMS virality, and the declining novelty of invites, the response rates (and viral factors cratered)

The era of hypersimple hyperviral apps ended here

And to this day, it’s basically impossible to create loops where the viral factor is >1 in the first session

Retention is king for virality
The modern app isn’t obnoxiously hyperviral -- prompting to invite and invite, as prior generations of products did. Instead it’s driven on a few components:

1) multiple top of funnel channels

2) great retention that drives virality

Here’s what I mean:

First you need multiple acquisition channels that are diversified and can be as spiky as you want. This is where social media, video launches, press mentions, SEO, and even paid marketing techniques can all feed into your signups. It’s fine if this doesn’t scale or grow as long as you get some kind of consistent top-of-funnel dripping users into your product at all times.

For a product like the Uber app -- when I was there -- it was something like 50% of first trips might be from paid marketing. Then 10-20% from referral programs, and the remainder from word of mouth, SEO, and so on. We’d buy activated mobile users, with broad base and untargeted advertising, for $10-20 and the math all worked. For other products, you might have a different acquisition mix depending on how aggressively you pushed paid versus SEO and so on. A high-intent product category like travel might have a lot more from SEO and referrals, since you need to be near transactions. Either way, you just need some set of top of funnel sources that work.

Second you need the product to generate a bunch of user sessions, so that you can grow virality over time -- that’s another way to say, you need strong retention. Oftentimes the viral factor is described simplistically, as:

invites x conversion rate

... but that implies that all the virality is generated in one session. In a highly retentive product instead you get a whole bunch of sessions and you might be able to ask users to share, or invite, or refer every single time. So instead you can think of it as the following:

the sum of:

session 1’s viral factor +

session 2’s viral factor +

...

And I’ll leave it up to an exercise to the reader but you can think of this as the infinite sum of each of the points on the retention curve of which each session can generate a little bit of viral factor. Each session’s viral factor is then based on the % of users that interact with the viral features, multiplied by the resulting shares or invites, conversion rate, etc.

I like to use a rule of thumb where about half the viral factor is generated in the first session and then the rest is generated in all of the sessions after. The reason for this is that in the first session the user is in the mindset of “setting up” their account. This is where you can ask them to set up their workspace and invite their colleagues or to invite their friends, etc. They’re often coming in with a high degree of intent and those set up steps can generate a ton of reality. In the second or third sessions, the problem is not only has have many of the users dropped off but also they are sort of in a different mindset where they’re expecting value from the product at that point. So while you can take them into viral flows, you don’t have the excuse that it’s for setting up.

The other thing that you’re seeing is that in reality apps will have multiple loops that are all generating different types of viral factor. A product like Dropbox might have several, such as:

• sharing a folder with a coworker

• inviting people to share

• referrals program

• using other Dropbox apps that have their own viral loops

Each one might work at varying levels of performance but the important thing is that you might be able to convince a user to engage in all three at varying levels across multiple sessions

This was the case at Uber, which I saw first-hand. People were introduced to Uber in many ways. You had loops like the referral program or you could earn dollars by inviting friends and they would receive dollars too. But you also recruited users naturally simply because you would often invite friends IRL to come take a ride in your Uber. Or engage in functionality like “Share ETA” that would expose friends to Uber, driving them to eventually download it.

A product might have three or four major loops that all work in concert and are ideally all completely useful based on the context of the usage. That way you’re not asking people to invite over and over again but instead there’s a loop for inviting on your first session, then maybe a referral program that you introduce to them in session number two, then you might have some embedded functionality where they share some content or otherwise with friends in a later session and so on.

To nerd out a bit: The viral factor is not generated by how many invites you send in one session. The viral factor is generated by the sum of all the viral features that you engage with over all of your sessions. The more loops you have, the more useful across more sessions because of great retention, the more viral your product is over time. And the more novel and exciting your product is to users (as AI products are now), the easier it is for the entire system to work together.

Low retention apps need to be spammy. Sticky apps do not.
Naturally if your product is highly retentive then you get many sessions to be able to ask the user to share or invite. You can have a small and unobtrusive viral sharing feature. As long as you get a large number of sessions, you’ll eventually have a viral loop that’s >1. On the other hand if you have a low retention product where you only get two or three sessions on average, then you’ll need to be very prominent and spammy to push the user towards viral features. This is why sticky, high retention apps can be more viral over time.

I remember in Facebook’s early days it was amazing how less spammy they were compared to other social networking tools. While they had email invite capabilities, they existed just on the right railing of the website. Small and unobtrusive. If you didn’t have any friends, it would be quite prominent but for most users it really was not a big thing to push people to invite as many friends as possible. I think this is because the product was very sticky and well made from day one and so the viral factor was gained over a really long period of time rather than some of the spammier social network competitors that had to use invites to grow but ultimately a lower retention rate and user frustration given the spamminess caused Facebook to eventually win.

Is viral factor useful when it’s <1?
The funny thing about viral factors is that although people often love the idea of exceeding 1.0, as far as I can tell those unique windows only last in very short periods of time when there’s a new platform or some new novel mechanic that ultimately allows hyper simple viral apps to succeed. Most of the time, you see viral factors that are 0.2 or 0.3 or below.

This is still valuable! A viral factor of 0.2 means that when you have 1000 users sign up, you get 200 users “for free” and that discounts CAC in a meaningful way. In this way viral loops have ended up playing more of a support role because they help stretch the dollars that you’re putting into marketing. A highly-retentive product will over time grow signups in proportion to the active user base even if it’s not doing it in a super quick way.

The “speed” of a viral loop is an important concept here. Virality may not manifest itself quickly, particularly when it’s a high-retention/low-spamminess product. When viral factor is generated slowly over many sessions, you can understand why viral loops have a speed of how fast they generate invites. If you have a product that users engage with every day like a social network app, then they may send off a bunch of invites every day. This leads to an accumulating viral factor that is going to generate more sign-ups quickly. Contrast that to the referral program of a product like Dropbox, which might be high utility and you often use it in the background but you may not mess with the referral functionality more than once a month. This means that while the product is super sticky (by the way, it is -- Dropbox has signed up hundreds of millions of users), it might take years to generate viral signups simply because the loop speed is slow.

In that way, viral loops can be slow, powerful, and make big userbases even bigger. And this is important particularly in later stages when you really don’t want to use paid marketing extensively.

In markets like consumer or prosumer where the eventual goal of an app is to sign up hundreds of millions of users, you’re not really wanting to use paid marketing to buy an audience of that size. After all, 200 million users * $10/install = crazy money. Instead you’d like to get there because you spend millions on marketing but you also get a lot of organic spread. You get a discount that comes from all of your viral loops, you have SEO and ASO, etc. (And GEO, or whatever comes next in the AI era)

Long live ragebaiting and shitpoasting
OK so let’s wrap this up -- what’s going on in today’s viral marketing landscape?

If you believe my framework so far, it means that shitpoasting and ragebaiting and cinematic videos and so on, just serve to help generate random spikes of signups. They’re not repeatable, but that’s OK. But those signups are then amplified due to the nature of many of the AI tools out there.

The current generation of AI tools often have a “create and share” viral loops that can amplify whatever top of funnel comes in. If you give your users a novel way to generate music, or video, or anything else using AI generative models, then naturally a high percent of users will do that. And then a high percent of users that interact with the generated result will want to share it with their friends and so that percentage will also be high.

I think that’s why we’re seeing so many amazing highly visual AI tools becoming super viral because they not only use this “create and share” of our loop. Further, we’re in the age of highly visual social media -- short form videos, embedded clips on posts, etc. And so the nature of gen AI output and what’s working on social platforms is highly resonant. This means it’ll spread far and wide.