Need direction in your business?
Take the Fizzle Quizzle

In less than 2 minutes this quiz will show you where you're stuck and what to do about it.

TAKE THE QUIZ »

An Insider’s Look at a Viral Blog Post: 220,000 Visitors and a TV Appearance

220,000 visitors to one page. That’s what I saw after just three weeks. And thousands more continue to pour in every day.

I’ve had my site mentioned on big sites before and gotten a lovely spike of traffic, and I’ve had certain posts go viral on one particular social networking site, but both of them are dwarfed by recent events. Let me tell you all about it!

On their birthdays, lots of bloggers have the idea to share x things they’ve learned, where x is the number of years old they have become. I was feeling reflective, so I wrote up that post myself, but they weren’t just things I’ve learned, they were life lessons. As well as this I didn’t actually learn them in 29 years, but in the last 8 years that I’ve been on the road as a nomad. So I wrote:

29 life lessons learned in travelling the world for 8 years straight

Some of my posts take days to write, but this one took one afternoon because I had already been reflecting for several days and knew what I wanted to write. I submitted it, got some very positive initial feedback in the comments and didn’t think much more would come of it.

And then this happened:

Not a spike of traffic, not a single social networking or large site, but several major sources of traffic started pouring in (and continue to pour in). These include:

  • Stumbleupon (7k stumbles in the first 12 hours – currently at 63k). Total traffic (according to Google Analytics): 51,108
  • Highlighted on “cool” recommendations in Google Reader; a feed many people subscribe to. Traffic: Unknown, but likely lots of new RSS subscribers (see below)
  • Lots of Facebook likes (currently 18k). Total Facebook traffic over highlighted period: 46,094 – almost as high as S.U.
  • 1.5k Google +‘s and many shares on Google+, leading to 4,796 visitors
  • Over 2,200 RTs: 4,992 visitors from twitter.com
  • On several main sub-pages of Reddit for several days (mostly on Travel, YouShouldKnow). Not the main page, but sent me over 8k visits nonetheless!
  • The iPad app “Zite” featured the post
  • Dozens of forums and other blogs posted links to it – discussions about it spanned over 10 pages on two separate major forums

The big spike you see on July 19th is from when Hacker News linked to me on their front page. The total traffic from Hacker News in the above period is 15,635 visits.

Visits/day before this post: 2,485

Visits/day three weeks after the post: 7,035

Peak Traffic Day (July 19th): 35,070

How to Prepare Your Site for Going Viral

There are several different factors at play here, as I understand them. Let me try to explain them as best as I can.

The entire viral effect was driven almost exclusively by Facebook and Stumbleupon. I got waves of traffic from the other sources, but these two clearly win out and ultimately led to the other sources getting sparked off.

As far as I see it, my site is highly optimized for both networks. I am personally a very active Stumbleupon user and understand how its users think because I’m one of them. So I know the evaluations made when you arrive on a potential page. I’m also an active Facebook user, but so are half a billion people. I actively investigate all the links shared by my network but I know what works to get me to re-share.

Some key factors that contributed to the post going viral are:

  • My site has very clear and prominent share buttons that hover on the left. When reading a particular point that strikes you as relevant, the share button for your favorite network is always there at the crucial time. I asked my web designer Charlie to mimic the buttons as seen on Mashable. A static share button is not as useful; at the beginning of the article you aren’t quite “sold” on it yet, and at the end you may already have lost your buzz from the best part.
  • The site does not come across as spammy.

This was a major decision I made in the redesign: to remove banners for my product, to continue to not add in the blackout screen for e-mail sign-ups. (Yes, it’s effective, but as a reader of blogs, I maintain that it is incredibly annoying. Purely as a consumer, I lose respect for your site when you do it and I’m visiting for the first time.) I also worked to give it a clean and professional look.

There are also no advertisements at all on my site.

Interestingly, since I took banners off the main page, I started making more sales of my product (the Language Hacking Guide) than before because I could earn people’s trust easier than the first impression they may have gotten from seeing my product in the top right that was there before.

People find the link eventually when looking at my resources pages, or through simple links positioned only when relevant. At this stage they would have read enough of my content to know that I’m genuine. When an ad is the first thing they see, then what reason do they have to click it if I haven’t earned their trust yet?

This also makes it a more enjoyable reading experience for everyone, including those who may not buy from me but will share my site with their friends.

When I stumble onto any page (from Facebook, S.U., Twitter, etc.), if I don’t get the impression that the author is trying to pitch me something as soon as I land, then I will let my guard down and enjoy the content. In my opinion too many of my blogger friends pitch too hard, and they come across as spammy.

This is especially true to audiences outside of North America. Over 50% of the traffic in the highlighted period in the Google Analytics graph was from outside of North America. There is a very important cultural difference in accepting advertising and “enthusiastic” pitches. If you don’t take this into account you will only ever be popular in North America where people are used to it and treat it as background noise.

  • While on the topic of international audiences, I have the Google Translate bar installed. If your browser’s language is not set to English then it will ask you if you want the content translated. (Similar to what natively happens in Google Chrome, but it will happen in all browsers on my site). If you can understand the post, you’ll be more likely to share it.
  • This is clearly a personal styled blog.

Even though the topic is languages and travel (this is not a diary styled blog), I make sure to inject myself into every aspect of the site. You see several photos of me on every page (and videos), and the writing style is more casual.

I put great effort into personalizing my blog and brand, and readers can relate to me much better because of this. (It’s so effective, I even get recognized in the street by readers quite a lot!)

A connection made with an author can people make them more likely to share the content.

Too many other websites seem too impersonal to me – as if corporations rather than individuals wrote them. The author’s photo is only on the about page (if at all), too professional a writing style is opted for, and no rapport can be built. People are just as likely to share an article because it’s good content as they are because they think the author is a nice guy/girl.

  • A powerful main photo. Along the same lines as personalizing it, I make sure to always have a relevant photo for each post. I don’t use stock images – only ones I am in myself, and this one (of me on the top of mount Teide – Spain’s highest peak) was particularly effective.
  • The list post formula.

This contribution to the post going viral isn’t one I’m such a big fan of. I took advantage of a formula that definitely works: List posts get shared more. Despite this, I will not be writing list posts again for a while.

When blogs overdo it they lose the ability to write about one subject in greater depth and sacrifice their readership and brand just for some traffic spikes. The reason I went with the list post this time is because it was a fit; I genuinely had dozens of points that I was reflecting on after 4/5 of a decade on the road.

Too many people force the formula, and their readers start to become immune to it. When I broke out a more “sharable” post than normal, then my readership enthusiastically shared it. If I did this every time and desperately tried to viral-ize each post, then I’d wear their patience thin.

The list post got the new readers on the blog, but the vast majority of much more in-depth content about one topic kept almost 20% of the 220,000 visitors on the blog after that (bounce rate for given period was 80.76%, which I’d expect from the sources, especially Stumbleupon traffic).

Over 8 minutes (8:15) average time on the site for all 220,000 visitors meant that the list post approach worked to keep them reading even though it was 3866 words long!

  • At the end of the post, I ask people directly to share it if they liked it. Sometimes, asking nicely really goes a long way.
  • A very unique and eye-catching title.

If you are doing list posts, it’s pretty well established that using a strange number helps a lot. The number 29 really begs the question as to why I didn’t just round it up. This encourages people to click on it even more to satisfy their curiosity. This technique is well known to many people, and there are ton of articles of this format online that don’t go viral.

What really made mine stand out is the qualifier. It’s not just 29 life lessons learned in 29 years, or 29 lessons learned as a blogger etc. It’s in travelling the world for 8 years straight. This implies that it won’t be the same as other lists you’ve read, and it’s such an uncommon thing to hear of someone travelling so long that once again, the title alone will draw people in out of pure curiosity.

As well as this, I made sure that the first words of the article itself were also powerful so people would be encouraged to keep reading.

  • The content is just good. Easily overlooked, but this is actually the reason the post went viral. Read it and you may be inspired to share it yourself!

Really, There is No ‘Formula’, Only a Good Direction to Go

While all of these contributed – the fact is that of course you can have a post go viral without following many of these steps. But in my opinion, these all contributed a lot to my traffic of almost a quarter of a million visitors in the last month.

I have written posts that I’ve been proud of before that didn’t go viral. In fact, I’ve learned the hard way that if I really try then it’s forcing the system too much and it just won’t work. That’s not how social networking operates.

I was sure that my post about how to speak Na’vi for your Avatar would be on the front page of Digg within hours, since the timing was perfect during the buzz of the movie. But in the end, there was little interest in it and all of the traffic to that page came because the one before it went viral [Any phonetic script can be learned in just a few hours – 101k Stumbles so far]. To me that was a “normal” post, and not one I expected to go viral.

Another post I wrote to simply fill-in between “better” posts ended up also taking over, and when I wrote about How to speak English like the Irish, that ended up skyrocketting towards 111k stumbles.

As a side note; whoever stumbles your post initially does not matter. I submitted my Phonetic script post, and my sister submitted my recent 29 life lessons post. Neither of us had anything else go viral from our accounts, but it was Stumbleupon that got the traffic flowing initially on the 29 life lessons post. Make sure it’s easy to submit something, but after that, the network will decide if it’s worthy of further sharing.

You’ll notice that the style of these posts is much more like my typical ones in going into detail without separating into bullet points. And when these posts went viral I was breaking a few of the rules shared above – my site design was pretty bad (a free theme I found somewhere that I paid $20 for to remove the copyright info), I had ugly banners on the right, and I had no floating share-buttons (but did remind the reader to share within the post). However, I was still applying many of the other points.

Results of This Traffic

Apart from the traffic outlined above, there have been some interesting side benefits to this viral effect. Some of them are worth noting because in some cases viral traffic might not actually do anything at all for you other than boost your ego if they aren’t converting into something else.

For the first week of viral traffic, sales of my product remained the same. Even though I was getting ten times my daily traffic, I was not making ten times as many sales. Not even twice as many, and I linked to the product within the article itself, twice. Many of the sources of traffic I was getting are simply not people who are in the mindset of buying something.

Things changed however when I got on the front page of Hacker News, and made ten times as many sales over two days. Since then, sales have been about twice my typical rate. Clearly, that site has had more immediate and direct benefit to me for something to get viral on.

Hacker News visitors are much more likely to be computer savvy than Facebook users, and perhaps more likely to have already purchased something online. If traffic had continued to come just from Facebook and Stumbleupon, I may not have had any financial benefit at all from this traffic.

I did get them to convert in other ways though! And down the line this is always something that I can take advantage of in other ways.

For example, new Facebook likes for my brand’s Facebook page followed a similar pattern to the overall site traffic.

This was possible because the Page widget is prominently featured on the site.

And what about email subscriber numbers? Let’s take a look at the whole month:

(The peak at the start of the month is due to guest posting on Art of Manliness.)

There were clear changes, but not following in the same ratio as the traffic itself (although the sign-ups do somewhat resemble the graph of the traffic). The last spike you see above is one of the most interesting results of all of this experience.

TV Appearance

Yes – all of this viral traffic even led to me appearing on national TV in Australia! Someone who works for Seven Network came across my site and got in touch with me after looking through my site in more depth.

Seeing the video interview you can see that it worked out pretty well – they showed my product sales pitch video to hundreds of thousands of people, and after a quick interview I made sure to say my browser’s URL at the end.

I was very curious to see what kind of traffic I would get from an important TV appearance. Estimates put viewership of Weekend Sunrise as anything up to half a million people, and yet you can definitely see no spike appearing over the weekend in my Google Analytics graph.

I think something must be wrong in how Google tracks IP addresses because the day before the interview it says I got 620 visitors and the day of the interview, 833 from Australia. It also says that my direct traffic (i.e. people just typing my site’s URL into a browser, which you would expect them to do after hearing me say it) was less the day of the interview than the day before it.

It would have been possible to create a special redirect URL to say just on that TV show to “track”, but I wanted to keep it simple. While it seems like I got no traffic from appearing on such a great timeslot on national TV, other data tells another story!

Email sign-ups surged up and I, once again, made ten times the sales on the Saturday of the show than what I normally make. I’ve also been getting dozens of emails from people (after signing up to the email list) who say they saw my interview.

Not Every Metric is Useful

The data from the TV appearance paints a different story to the results, so I will soon be returning to my status quo of only checking Google Analytics every few weeks (and really only to follow up on incoming links at that). Metrics are less useful than you think.

One metric you’ll notice I haven’t brought up is RSS subscriber numbers. The reason for this is because I decided over a year ago to stop tracking it. Feedburner is too moody on good days, and it’s just not relevant enough for my niche.

If you blog about technology or the Internet you are more likely to get useful information from your subscriber numbers, but too many of my readers have told me (via email) that my site is among their five or so only bookmarks in Internet Explorer that they check daily, for me to presume that a useful percentage of them even know what RSS is.

I really feel it’s an irrelevant metric for blogs with audiences that aren’t necessarily going to know what it is, and in my case languages and travel don’t necessarily mix with people who know what Google Reader is. So frankly, I don’t care how much my RSS numbers have changed.  There is an icon on the site, I’m subscribed to my own feed so I know it’s working and that’s all I care about.

Although I suspect that being showcased in Google Reader’s “cool” posts would have increased this number noticeably because they would have much easier one-click action to subscribe and be in a position to appreciate it.

Since I stopped checking my RSS count and focused more on content and more interaction with readers, rather than on faceless numbers, things have improved dramatically for me over the last year.

Conclusions?

So, as you can see I am quite verbose!  Being just shy of 3k words is the norm for me.

I’m sure this is breaking a rule of a few hundred words working better in viral posts. You can easily have a post shared by thousands of people by breaking many of the guidelines I’ve mentioned here, but I hope you realize that aiming to improve on a few of the points I mentioned may indeed increase your “share-ability” factor!

If you have any questions about this whole experience, I’d be happy to answer them in the comments! Thanks for reading along.




Related Articles

Earn a living doing something you love.

Fizzle membership has the training, tools, coaching and community you need to succeed.
Join Fizzle today and get your first month for just $1.

Start your first month for $1


Get your first month for $1. After that it's $39/month. Cancel anytime.