Will Marlow

A lot of people think (or hope) that one of their ideas or marketing materials will somehow "go viral," but most people don't put much time into thinking about what actually causes something to spread virally.  Without doing too much math, here's the basics of how something spreads virally: You start with your initial userbase (let's say it's 100 users).  A percentage of your userbase (for easy math, let's say 10%) will invite 10 people to check out your idea or product. Of those 10 invites they send out, let's assume 10% accept the invitation and become new users.  You add these new users to your original userbase.  But you also should assume (for easy math) that your attrition rate is 10%, so you subtract 10% from your initial userbase.  This represents one wave of viral growth that will then repeat itself to continue the trend.  (In this example, of course, you're losing the same number of users that you're gaining.)

But here's how it gets tricky: you can assume that your viral growth will get worse over time, because in the second, third, and fourth waves of growth, you can expect more and more invitations to go out to the same people, which means your conversion rate (which started out at 10%) may plummet as invitations reach people who are already users, or who have already rejected invitations.  So people begin adopting it at a lower rate, while your attrition rate is likely to stay the same.  This means that each wave of growth is smaller, until eventually it stops.  As Andrew Chen explains it, each "new batch of users needs to exceed the previous batch in order to "go viral."

What can you do with this information?  Well, the most important thing you can do is set measurable goals for viral activity to guide your decisions and planning to avoid wasting time and resources.  If you know what you're looking for (adoption rate; conversion rate of new users; attrition rate of your base over time; the size of the universe of users), you can actually measure the factors that have the greatest impact on your success and you can learn how to improve results over time.  Like most useful things, viral activity isn't guided by one enormous X-factor that either is present or absent.  If you know the rules, you can make it happen.

PS - If you want to go deeper into viral models, follow this link to read Andrew Chen's stuff.  I will return to this topic periodically, and I hope my posts are helpful on this subject, but if there's a better place than Chen's blog to go for an in depth, technical analysis of viral products and marketing, I'm unaware of it.

Will Marlow co-founded AlumniFidelity to help his clients reposition their fundraising to benefit from Web 2.0 technology and marketing techniques. He’s working with clients such as UVA, the College of William & Mary, the University of Oklahoma, Bowling Green State University, Randolph Macon College, and he loves nothing better than a thorny marketing challenge.  Email him at will@alumnifidelity.com.  

Loading mentions Retweet

I once heard a great story about someone who accidentally sent an email containing embarrassing personal information to the wrong recipient.  The information in the email was potentially embarrassing to the sender, but it wasn't John Edwards-scale traumatizing, and the sender decided to use an interesting tactic to deal with the error.  Rather than contact the recipient and explain the error, or ask that the recipient to delete the email without reading it, she simply sent the recipient 20 more identical emails - with the embarrassing sections omitted, of course.  This way, the recipient read the most recent email and assumed they were all identical, and deleted the rest without ever noticing that the first email contained a unique section.

Loading mentions Retweet

(Click here for the second part of this post.)

Before I get to the problem with my Twitter strategy (and what I intend to do about it), I want to explain a few things first.

As of today, I have 12,730 followers on Twitter.

The way I've built my following is fairly simple.  I post updates that are focused on my expertise (education technology, charitable fundraising, and web marketing), and I make myself easy to find for people who are interested in the same topics, I update at regular intervals daily, and I follow back individuals who follow me unless they are a spammer or look weird. 

In addition to this, I have been on Twitter for awhile, and in my early days I proactively sought out other like-minded and interesting Twitterers and I introduced myself to them by following them.  Many times they followed me back, @replied me, or interacted with me offline.  I also talk to people via Twitter, respond to DMs, and maintain public profiles on Flickr and at this blog at www.willmarlow.com to give my followers more insight into my personality and my work.  In addition to all of this, as the cofounder of AlumniFidelity.com, I am regularly interacting with customers, investors, and end-users who are tech-savvy.  This has led to lots of my followers.

I also never engage in obnoxious spam activity or use obnoxious Twitter “clients” that alienate the people who I hope to interact with.

One problem with my Twitter strategy is that my liberal “follow back” approach (like Guy Kawasaki, I follow people back as part of Twitter courtesy), I have accumulated a large number of spammers and faux-followers in my following column, which I have recently gotten sick of.  It occurred to me that I’m currently following a large number of feeds that I don’t read, and the fact that I’m following them increases their own credibility (spammers try to build up followes for that reason, I’m compiling a list of users who I interact with regularly, or whose insights I appreciate, or who I happen to know are legitimate users, and I’m going to unfollow all 12,400 people, with the exception of this relatively small list.

If you are NOT a spammer, and I’m currently following you, please @reply me or DM me if you want me to add you to my private list of people who I will follow back after I hit zero.

I’m not sure how long it will take me, but starting in 6 days, I’m going to unfollow all 12,400 people, until I reach zero, and then I will slowly begin adding followers who I’m truly interested in.

To summarize: problem = I’m following too many spammers and bots on Twitter.  Solution: unfollow everyone, and re-follow only a select number of people who I am actually interested in. 

As always, send me an email if you want to chat about any of this.

Loading mentions Retweet

Last night my car was towed from a restaurant's parking lot while I was eating dinner inside.  The owner of the restaurant stopped preparing sushi to personally drive me and my party to the towing lot to retrieve the car.  The owner could have handled this problem in any number of acceptable ways, and most of those acceptable ways would have resulted in me repeatedly telling a story to my friends about how my car was towed while I ate dinner at his restaurant, and how inconvenient it was.  Now, I will instead tell the story about how he personally drove me across town to retrieve my car after it was towed. 

Loading mentions Retweet