K-Factor is not an Accurate Viral Measure

Posted 2 years ago by Tom Maiaroto.


In the course of my research, I've seen quite a few claims about how k-factor can be applied to measure how viral something is. It's a great value to understand and use, but some people apply the formula to measure the wrong things. What does it measure? It measures adoption rate or conversion rate. It does not measure "virality" or how viral something is. Though you should see a greater k-factor if something is viral, this is not always a true statement nor is it something that holds any accuracy.

K-factor is simply the number of invites multiplied by the number of new users. That's it. Adoption or conversion rate. However, people have erronously taken that as number of "shares" multiplied by number of "views." They've also complicated the formula with a bunch of other values that don't really help. 

The problem here is that k-factor is something you need to watch over time and is something that is relative to just your own application or product. Assuming we're talking about a web application; an increase in value could mean that you need to bring on more servers to handle the traffic. It's great to show that you've had more users and if you watch the values over time you might even be able to forecast how many new users you will receive. It could also tell you when to shut off some servers or to increase or decrease your advertising budget.

However, don't confuse this with how viral something is. Virality is really the measure of the spread of information and not the adoption rate or new subscribers that you get for your product. Again, like views, that's a byproduct of being viral.

It is said, in a few places, if your k-factor is over 1.0 then you've gone viral. This is a completely unfounded claim. A k-factor can be much, much greater than one! It's a very arbitrary value and it's unfair to compare it to another application because there's too many variables. 

Let's just start to go through the process here...If 50% of your users are sharing your application which gives you X number of new users...Per what? Day? Month? After you've answered that (and understood why you chose the time period), what's 50% of your users? 1,000 users? 500 users? Is that good? What's the average number of users for any given application on the internet? Do all applications have the same number of users? What if a user shares more than once? Can you determine that? Where does it fit in the formula? Can a user only share your application once?

I hope a few of those common sense questions just made you realize the problem.

One application's k-factor of "1.0" or "0.5" compared to another's will mean different things. Your application can be doing well for itself without question...But by comparison, no one may really know that it exists. You could derive a k-factor value higher than say Twitter or Facebook for example. Now, would you say your application has the same adoption rate? Or that it is as viral? I hope not...The most fair statement you could make would be that at the time Facebook or Twitter had as many users as you have now, you were growing faster. Cool beans! That still doesn't mean your product is more viral. It certainly doesn't mean your growth will stay that way either, but you know that. Honestly, deep down, you know that.

So where's the value? The value is relative and is only measurable against your own product and its history. Only you will know what a k-factor of 0.5 or 0.25 or 1 or 5 or 500 will mean to you and your product. It is most definitely useful in determing when you might need to expand on your operation, but it is not a reasonable measure of how viral your product is.

Again, this only measures adoption rate. It doesn't measure the spread of information. It won't measure how many different social networks your content appears on...It won't measure how many potential eye-balls are on your content or have heard the name of your application, product, or brand.

However, there is a relationship between the k-factor and your virality score. You should expect a higher adoption or conversion rate (k-factor value) if you are more viral. This is common sense and is known to be quite true. That's why everyone puts share buttons all over the place. Even spammers have been known to increase adoption rate or downloads simply by spamming links across Twitter. Though you may want to take a look at your retention rate if that's your modus operandi.

This relationship between adoption rate and virality is why some people misuse the k-factor measurement. It's not bad to get a sense for how viral your product is, but it's not really accurate either. For example, it won't tell you if your video or application was as viral as Gangnam Style or not. You likely can't even get the numbers that you need from other things to be able to compare k-factors.

That's what you need an index for. That's why you use ViralityIndex.com.


Filed under
comments powered by Disqus

Search for Posts

Popular Labels

social media virality score general internet web development reviews and opinions php web tools hosting lithium web design

Recent Posts