If your startup is building a consumer product, your product has to be viral. For consumer products, the average revenue per user tends to be relatively low, so you won’t be able to afford spending much on advertising or sales. It’s hard to slap a viral channel onto an existing product, and you’ll have to think about how to make viral growth happen right at the start.
Consumer product startups have to bake a viral channel into their product from the get-go. They can’t merely glue it on later.
Sam Altman of YCombinator says that your product has to be so good that users spontaneously recommend it to their friends. It’s true that people don’t recommend crappy products. But you can do a lot to nudge users to invite others.
At Area 120, I spend a lot of time with Googlers encouraging them to think about early product growth. It’s hard to think about virality in abstract: Most people are only consciously aware of one or two products to which they’ve invited their friends. I figured it would be useful to provide a guide to 9 common viral mechanisms:
Now let’s break these down:
1. Two-Sided Reward
Create an incentive for a user to invite their friends, and for their friends to accept the invite. Rewarding both sides is crucial — if the incentive is missing for either side, it won’t catch.
In 2008 and 2009, Dropbox grew by leaps and bounds through their referral program. In the beginning the product, looked like this:
For every friend that joined Dropbox, they would give both you and your friend 500 MB of bonus space on your account for free. This created an incentive for the inviter to invite their friends, and for the invitee to accept the invite. This rewarded both sides and created a viral loop.
What made Dropbox’s reward particularly brilliant is that the cost of additional storage space was due to drastically decline. The storage space that they gave you was worth a lot to you at the time, but for the company it cost less and less to provide over time. 500MB of free storage seems quaint these days.
Giving users a monetary incentive to invite their friends is also compelling. For example, here’s Good Eggs, a Bay Area-based grocery delivery service, which offers $20 of store credit for every friend I invite to their platform:
This is a version of a 2-sided reward loop: Good Eggs gives me store credit, and will send my friend a free item in their delivery if they order. In order for this to work, Good Eggs has to have a lifetime value (LTV) of the invited customer greater than $40 (since they are offering $20 to me and items worth about $20 to my friend). In order for my friend to receive the free item, they must also make an order to purchase at least $30 worth of groceries, which makes this work economically for Good Eggs.
The above is also a nice example of the gifting effect. By letting me choose which of the options I send to my friend, my attention is redirected from fears of spamming them, to finding something that they would like. By receiving both the value of a gift and knowing that I put thought into selecting it, my friend is more likely to accept the gift than if it were impersonal.
2. Appeal to Vanity
Appeal to users’ sense of vanity and competitiveness to encourage them to spend more time using your product and invite others into it. Expose metrics for them to drive up.
Remember the 2009 popularity contest on Twitter between Ashton Kutcher and CNN of who would first get to 1M followers first? By showing follower counts prominently on Twitter profiles, both Ashton and CNN’s competitive bent was activated.
This battle created a lot of free publicity for Twitter, as both Ashton and CNN used their media power to blast out requests for followers. Users would sign up for Twitter, and then immediately follow one or both of them, thus populating their feeds with interesting and relevant content. Having your Twitter feed populated with interesting information would then in turn make you more likely to stick around as a Twitter user.
In the early days, LinkedIn had a similar thing going. By showing the number of connections front and center on your profile, they incentivized you to start thinking about inviting more users to connect with you. And by inviting more people to your network, you consequently brought more into their network.
It’s hard to see your profile metrics and not think about how to make them go up. One sure sign that appealing to vanity drives interest is that the Internet is littered with articles about how to make your LinkedIn connections go to 500, and how to get more Twitter followers.
Apps built for collaboration or communication between coworkers are inherently viral. Some have a single user mode, but additional value is realized from the product with multiple people.
Slack is built for instant communication between teams, and in order to unleash its power you first have to invite your whole team. Google Docs has the single-player mode of just editing a document by yourself, but the real power of it gets unlocked once you click “Share” and start collaborating with coworkers. Buttons labeled “Invite people” or “Share” are the viral mechanisms of these products.
My guess is that we’re going to see more specialized tools that take advantage of this type of virality. Think of tools like AirTable, with its part spreadsheet, part database. Once they establish a foothold within a company or organization, these tools tend to spread quite rapidly within each company.
Create the ability for others to embed your product into their website. This will create exposure, and also model to potential users the expected content and behaviors in your product.
If you’re building a product with user generated content (UGC), you should make the content easily embeddable into other sites. You’ll often see tweets from Twitter embedded in news articles, and similarly it’s common to see YouTube videos on a variety of sites:
Embeds also serve as an educational tool for potential users: Seeing examples educates them about the type of content that’s on your platform, the creators that are on it, and models the behavior of your platform to them. Moreover, seeing the content on a site they already trust, gives your platform credibility.
Your hope for this viral channel is that users will eventually be curious enough about the content that’s embedded to click through to your site and sign up. But it may take some time to get there — depending on your content, it may take many impressions of your embedded content before they are familiar enough with your brand to consider signing up.
5. Artifacts Shared on Social
Your product naturally produces artifacts that can be shared manually or automatically on social networks. Clicking through takes potential users back to your product.
For user-generated content (UGC) products, there exists an even more powerful viral channel than embeddable content: social sharing. If your product creates uniquely interesting content, users can be encouraged to share it on their social networks, thus spreading knowledge of your product to their network. The most successful examples in this genre is auto-shared content onto other networks.
Instagram initially existed to create beautiful photos from your crappy smartphone camera. Those photos were often auto-shared to Facebook or Twitter, with a link back to Instagram. If you saw someone else’s beautiful photo, you would become curious as to how they made it, and in turn would have download Instagram.
Pinterest would let you pin items from the web, and then share the pins onto Facebook. This worked really well. When someone discovered the Pin, they would click into Pinterest and be shown a collection of more interesting items, quickly going down a Wikipedia-style browsing black hole.
Another way to draw attention to your product is watermarking: If your product creates images or videos of some kind, add a watermark into the main image to let users know where they were created. However, this is less effective than the social loops above because the watermarks don’t directly link into your product.
Chris Dixon calls this “Come for the tool, stay for the network.” You create a uniquely appealing artifact, and make users want to share it on their networks. As that user’s friends discover the artifact, they click into your experience where you can educate them about your platform. Eventually they’ll consider signing up as well, creating more content, which in turn is then shared out to their users.
6. Artifacts Shared via Messaging
Your product generates artifacts or URLs that are expected to be shared via messaging. Your product provides a rich experience around the information that’s shared.
While you’re on a ride in Lyft, you’ll see a screen like the one below. The “Send ETA” button is actually a viral feature. Pressing it prompts the user to send a tracking URL to another person, which in turn spreads the word about Lyft.
By sharing the real-time tracking page with them that Lyft sends out, they’ll be reminded to use Lyft in their next time-critical scenario as well.
There’s another class of examples where the link being shared is less timely and more permanent. It’s when a URL in a product can serve as the canonical representation of a concept or thing. Here are a few examples:
These are all items that are usually shared with others via some form of messaging. You want to send over the address where you’re going to meet, or the product you’d like to get someone as a birthday present. By creating a URL that richly represents what your user wants to share, the product becomes viral as well.
Your product sends messages of some kind. You attach a promotional signature to each message, linking back to your product.
Appending promotional signatures is a powerful way to make messaging products spread. Folks that are about my age usually think of Hotmail’s signature line at the end of every email, which helped Hotmail grow from 20k users to 1.5M users over the course of 4 months in the early days of the web:
Another example in this genre is MailChimp which is email automation software, and sends emails itself. If you’re in market for an email marketing service yourself, you’ll see something like this appended to each one of the emails they send:
Users may find this annoying. It’s easy to provide an opt-out. For example, the iPhone’s Mail app by default appends “Sent from my iPhone” to every message, this can easily be turned off, yet most people don’t bother to do it.
There’s an obvious limitation to this strategy: It only works for products that send out some form of messages, usually over email. But when it applies, it’s very powerful.
8. First Message sent on Existing Messaging Platform
Your product offers a differentiated type of messaging and encourages users to invite their friends to join them on the new messaging platform.
There’s a category of apps that will send you a message on an existing messaging platform, and try to move you into using their channel as well. A great early example of this was GroupMe, which would send messages to you over SMS at first, and include an upsell to the full experience in those messages.
This also works for reactivation: Booking platforms like AirBnB and VRBO will email the host each time they receive a new message, with a reminder that the experience is better in the app. LinkedIn is similar in that you get notified about inbound messages over email, but you’re guided to reply in the app. That’s the limitation of this strategy: If the messaging experience isn’t actually better in-product than over SMS or email, users will just find it annoying.
9. Highly Visible Hardware
A hardware product set in a highly visible location that is exposed to potential customers.
Hardware products can be viral too. Think about Square: Imagine you are a bakery owner and you walk into the bakery pictured below, you’re going to be exposed to a brand new POS solution — the slick Square terminal experience. That may make you consider getting Square for your own bakery too. This viral channel is the main reason why Square is growing at more than 50% year-over-year, and why 80% of their signups come at no acquisition cost to them.
Another very different example in the high visibility hardware category is the outdoor Nest Cam.
While the industrial design is somewhat muted, people are naturally aware of cameras that point at them, which draws attention to it it. Because the outdoor Nest Cam is usually installed in visible locations, lots more people will see it than would see an indoor Nest Cam. If you later find yourself in the market for security solutions, you’ll consider Nest. This is why Dropcam — the company that Nest acquired which invented this camera — should have started building the outdoor camera first.
Hardware doesn’t let you track the stages in your viral loop very closely, or iterate on your product quickly — that’s the nature of the game. But the boost in virality for slick, highly visible hardware products is worth it.
The one commonality among all different types of virality is that they encourage existing users to advocate for your product, which then generates awareness and consideration with their friends.
Viral loops contain the user spreading awareness and consideration of your product. But conversion and loyalty all have to happen based on the quality of the product:
As a startup building a consumer product, you almost always have to make the product go viral in order to make the economics work. But virality is something that you have to build into the core mechanics of your product from the start — in most cases you can’t just glue it on later. As a startup founder thinking about business ideas, ask yourself: Is there a viral channel that naturally lends itself to my product idea?
Your product has already gone viral? Now you have to retain those users: Check out the companion piece, 11 Ways to Build Retention into Your Product.