The more information that is out there on the Internet, the more we need filters to sift through it, to get what is most relevant to us. More and more sites like Google and Facebook try to solve this for us by using information from our social graph to feed us the information that is, supposedly, most relevant to us.

On Facebook, this means that the content that is clicked, liked or commented by friends and family or friends of friends – people in our social graph – is what is delivered to us most. All done on the assumption that what is relevant and interesting to them is interesting and relevant to us.

But it also means that a lot of content is filtered out. Other content, from people, businesses or organizations that are not part of our social graph will not be delivered to us to the same extent.

For a business or organization the first implication is that they need to become part of our social graph to give their content a chance to get through the filter. On Facebook, they need us to “like” them.

But, there is more. The algorithm has other implications too.

Facebook’s Edge-rank decides how important something is to us and it currently consists of three components.

From Techcrunch:

“First, there’s an affinity score between the viewing user and the item’s creator — if you send your friend a lot of Facebook messages and check their profile often, then you’ll have a higher affinity score for that user than you would, say, an old acquaintance you haven’t spoken to in years.
Second, there’s a weight given to each type of Edge. A comment probably has more importance than a Like, for example.
And finally there’s the most obvious factor — time. The older an Edge is, the less important it becomes.”

This means that the more a user interacts with content by a person or business, the higher the Edge-rank they get for that user.

The implication for a business or organization is that they have to try to create content that is highly relevant. They should probably also try to create incentives or suggest interaction, by for example asking questions that really begs a reply or at least makes you want to “like”. However, creating a lot of content doesn’t translate into a higher Edge-rank as the average number of interactions per piece of content probably goes down. It is a constant search for the right content and the right amount of content to maximize interactions and Edge-rank.

Another positive result of trying to achieve a higher Edge-rank is that the interactions between a fan and the business is more visible in the feed of that user.

The Daily Beast has tried to decode further how the Edge-rank is calculated and come up with some conclusions, even though this is in no way scientific facts:

1. Facebook’s Bias Against Newcomers – newcomers rarely show up in the feeds of others.
2. Facebook’s Catch-22 – to be more visibile in friend’s feeds, they need to interact with your posts, but without being visible, they won’t interact.
3. The Velvet Rope: “Top News” – a lot of interaction is no guarantee to show up in Top News.
4. “Most Recent” News Is Censored, Too – the Most Recent feed is also capped to a maximum number of friends shown in the feed. Can be changed though.
5. “Stalking” Your Friends Won’t Get You Noticed – you can’t change your visibility to a friend by constantly visiting their page.
6. Having Friends Who Stalk You WILL Help Your Popularity – you will become more visible if friends stalk you. Even with friends who doesn’t stalk you.
7. Links Trump Status Updates – links are more visible than status updates.
8. Photos and Videos Trump Links – photos and videos are more visible than links.
9. The Power of Comments – when friends comment on your updates, photos and videos, your visibility will increase.
10. Why Facebook Really is Like High School – visibility to people with large networks takes even more time.

For more about Facebook’s and Google’s filters, take a look at this interesting TED-talk by Eli Pariser below.

For more on Edge-rank you could take a look at The Next Web’s “Everything you need to know about Facebook’s EdgeRank” or EdgeRank, den hemliga ingrediensen i Facebook! (in Swedish).

It is finally ready! Download by clicking the image or link below.

Viral Thinking – a “Viral Marketing Toolkit” is my ebook about how to maximize the potential for viral spread. Creating something that is viral is sometimes the goal, but it is not easy. In this ebook I have tried to cover the areas you can work with to improve the odds.

You could call it a “viral marketing toolkit”. However, just increasing the spread improves the ROI, even if it doesn’t go viral. So, my hope and belief is that it is helpful for anyone who works within marketing, PR or communications, whether online or offline.

I would like to thank the people who have generously helped me with feedback on this text and made it a lot better.

In alphabetical order: Valerio Franco (doublebblog.it), Anton Johansson (antonjohansson.com), Henrik Johansson (theviralcompany.com), Daniel Mencák (danielmencak.se), Patrick Möller (flavors.me/patrickmoeller), Cédric Rainotte (bwat.be), Niclas Strandh (digitalpr.se) Pere Tufet (tonterias.com) and Emmanuel Vivier (emmanuelvivier.com).

And of course, if you have feedback to give, do not hesitate to send me an email on kristofermencak[at]gmail[dot]com.

I hope you enjoy it!

And feel free to pass it on!

Have you ever met an old friend, one you haven’t seen for a while, had a little chat, exchanged your new contact details and promised to meet up and catch up on all the latest? Have you also happily forgotten about it all for weeks until you remember that you actually met by accident and promised to catch up? Why didn’t you catch up? You did call your other, newer friends when you said you should, right?

It’s the power of presence, or in this case, the absence of it. The old friend you meet is simply not present in your mind often enough for you to remember that catch up-meeting. While your new friends are present all the time and you sure do remember to call them if you say you shall. They are top of mind.

If in the old golden era of advertising, top of mind meant advertising enough to be remembered when choosing between brands in the aisle of the supermarket, now it also means to be present in social media. Replying to questions, helping out, sorting out complaints and create a positive experience for both current and potential customers. Then your brand might be top of mind during that moment of truth in the aisle in the supermarket. Then you might even become the preferred choice.

That is the power of presence.

The other day I read a post about “influentials” – these people marketers and PR-professionals so desperately want to reach.

From research by Duncan Watts, we have learnt that influentials aren’t that influential.

Influential or not, Watts believes that whether a trend (or idea or similar) takes off depends on how susceptible the society is to it overall.

Of course, the actual context is also important – people are influential within different areas – or as MediaPost Publications concluded:

“A new study from ICOM — a division of direct marketing agency Epsilon — finds that there is no universal influencer, and that consumers are influencers strictly within product categories, rather than across all categories.”

The actual number of people in a persons network is of course also an important factor in deciding whether this someone is influential or not. It is not the factor, but an important one, and the reason it is important might be that a large network makes it possible to take advantage of the big seed marketing approach that Watts suggests works.

However, research made on Twitter says the number of followers doesn’t really matter much and that the three measures of influence – followers, retweets and mentions – has surprisingly little overlap when looking at the top influentials.

The post on ReadWriteWeb also concludes that:

“…the most influential users hold significant influence over a variety of topics, as opposed to being experts in just one area.”

Hmm.. .. wasn’t that in direct contrast to what ICOM concluded?

My own conclusions are that:

• Context matters
• The size of the network matters

But whether or not someone is “influential” or not must have a lot to do with the ability to get retweets or inbound links for example. So, I think it has a lot to do with the stickyness of the message. What gets links and what gets retweeted has more to do with the actual content then who does the tweeting. And then we are back to what Duncan Watts concluded – it depends on how susceptible the society is to it overall.

Oops, I’ve been had again. This is weird, but of course there’s some secret and logic behind it. Try it out yourself! Right now, it spreads virally in a lot of places, like on Facebook and blogs. It’s closing in on 2 million views right now.

Interesting to see how it has spread in the english speaking world (especially UK, where Richard Wiseman if from) and is relatively unseen in other parts of the world, except Scandinavia, where pretty much everybody is fluent in English too. To spread even more to a broad public, it has to be very easy to understand visually, or accessible in more languages.

And yes, I have started reading Richard Wisemans book, Quirkology, though I haven’t finished it yet. Too much to read! =)

Thanks Judith for the tip on Facebook!

A viral campaign – everybody wants one. Few ever gets one.

I am the first to admit that it isn’t easy, but there are things you could to to better the odds of ever creating a campaign that goes viral. I prefer calling it “a campaign that goes viral” rather than “a viral campaign”, because, by definition, it can’t be a viral campaign until it has proven itself to actually go viral.

Then there is of course the distinction between campaigns that go viral on a larger scale and the ones that only go viral on a small scale, in very contextual environments. But, I am drifting away here, to topics that I didn’t intend to write about this time. The intention was to write about using an amplification loop as a way to increase potential for viral spread.

Viral expansion loops were discussed throroughly in a Fast Company article about a year and a half ago, in April 2008 – Ning’s Infinite Ambition.

The article tells us about Ning, a startup in Palo Alto, designed specifically to exploit viral loops. In the article, viral loops are said to have emerged as perhaps the most significant business accelerant to hit Silicon Valley since the search engine. But, we also learn that while viral advertising can’t be replicated; by definition, a viral loop must be.

About viral expansion loops from wikipedia:

“A viral expansion loop is similar to viral marketing with one notable difference: viral marketing can’t be replicated indefinitely, while a viral expansion loop must be in order for it to exist. When properly conceived and implemented, a viral loop almost guarantees self-replicating growth…

And a short quote from the article:

…Ning grows because each new user begets more users. Every time someone sets up a social network, he has no choice but to invite friends, family, colleagues, and like-minded strangers to sign on as well.”

Ning benefits from a “double viral loop,” which spreads two ways, because every network creator is a user, drawing in more users, and any new user can become a network creator.

Campaigns that go viral are usually not created in a way to permit something similar. But, there is something equally interesting – the viral retention loop – incentivizing users to come back to a service. This is very often used by Internet start-ups. One way of doing it is to send updates on what has happened on their account while not there.

I think we can expand this thinking and try to build campaigns that makes interaction with campaign material more than a one-off. For this, we have to build in incentives for users who have already been touched by the campaign to come back to the campaign material, interact again, and spread the material further. This should increase the basic reproductive number. If we can double the basic reproductive number of a campaign from for example 0.8 (by definition not a viral campaign, since it’s below 1) to 1.6, we have managed to set an avalanche in motion. This by simply adding an amplification loop into the campaign. Of course, numbers might differ a lot – these are just examples.

The difference between a retention loop and an amplification loop would be that the latter also incentivizes spread of the material, while the former just incentivizes the user to return to the service.

Through this (these – there can of course be more than one) extra loop we amplify the potential virality of the campaign. The trick is, as always, to come up with something interesting, something worth returning to,  and something that makes the user want to pass the material on again, maybe with a new twist.

Here are some reading tips about viral loops:

5 crucial stages in designing your viral loop

What’s your viral loop? Understanding the engine of adoption

And some for the swedish readers:

Viral loops vs. viral tools

Viral- & retention loops

Virulenta nätverk och loopar

Har din produkt en viral loop?

This really gives an idea of the enormity of content that is created all the time. Gary Hayes created this counter with real time stats on social media usage, updated for the time you have been here.

It took about 14 seconds for 1.000.000 million SMS to be sent, and about 8 seconds for 10.000 videos on Youtube to be watched. Impressive!

Hat tip to Ronnestam. =)

When talking about viral marketing, there is no question about it – content is king. To make videos, games or applications go truly viral, there is no substitute for great content.

However, even if you can have a pretty good idea, you can never know for sure how the content will be received by the audience. So, that’s one reason to do everything possible to optimize the campaign for viral spread. Besides this, there are two more reasons why you should try to optimize your campaign material, no matter what the quality of the content:

1. There is just too much other content out there to compete with, so there is a risk even good content will not be seen enough by the right people to make it go viral.
2. Once content has gotten attention, having it optimized maximizes spread and ROI.

Four questions
To maximize the possibilities for a campaign to go viral, I believe there are four important questions to ask yourself, evaluate the campaign and direct further actions and improvements. All questions are from the users perspective:

1. Why pay attention?
2. Why stay engaged?
3. Why act/interact?
4. Why not act/interact?

Finally, we also have to deal with the fact that there almost always are multiple audiences to a viral campaign. And, you may or may not want to appeal to them all.

Let’s go through each of the questions.

Question 1: Why pay attention?
Attention is the currency of the new media landscape, and to get rewarded with attention, you have to give something in return. That should be your content. But, millions of pieces of content are produced every day and users are only scanning it to decide what to pay attention to. To get their attention, you have to stand out among all other content when users are scanning. The “indicators of good content” have to be right. Otherwise you won’t get attention. Now, here are some bad and some good news – you can’t control all indicators of good content, but you control some of them. What you can’t control is often user-generated, like for example:

• Ratings
• Conversations
• Toplists
• Most viewed

On the positive side, to a certain extent, you can control:

• Features
• Titles
• Descriptions
• Images
• Tags

Once content is out there, it might get better or worse features, it might get new titles and new images. But, when launching the campaign, these should all be optimized.

Features
Features are something you usually have to buy, just like an ad. You can get features without paying, but then it is due to the content itself, when it’s being awarded video of the day or something similar. Paid features are the ones you can control. As always, more exposure, like for example through a feature, the more initial attention you will get, and this will increase possibilities of interaction with your content. Over time, more interaction equals greater possibility to get the content to spread.

Titles
The titles or seeding angles of a video or a game is also something that users scan to get an idea of what is hiding behind it. And, titles can sometimes be almost as important as the content itself. Titles, viral hooks or seeding angles should be appealing, connect to the content, be simple and memorable. Preferably, it should be intuitive, so that if you think of the content, and make a search on the words that come up in your mind, you should find the content. This makes it easy to find, to show to friends, and increases possibilities to be passed on.

Descriptions
While not as important as to create a catchy title, the description gives you yet another possibility to make it easy to find the content through search, and recognize it quickly when found. Descriptions should be accurate and filled with keywords you wish to rank high on.

Images
Images or thumbnails are usually scanned even before titles, and therefore, it is extremely important to optimize them to make it as appealing as possible to engage with the content. Images should be optimized in two ways:

• The first image, the thumbnail, should be optimized to get the user to want to engage with the content.
• The last frame, in for example a video, is often freezed on video sites, and should be optimized to increase potential of pass along.

Make it as appealing as possible, to first get the user to engage, and then at the end to keep the user thinking positively about the content. Using a black screen in the beginning and/or end of a video is the worst you can do. Sometimes video sites auto generate the thumbnail from the first frame and then you get a black thumbnail. Who wants to click on that?

Tags
Tags are also essential in making videos accessible through searches and increase exposure of the content. Tag the content with all relevant words that relate to the campaign material. If you have more than one video for example, tag them with the same words, as long as they are applicable to the video. This will increase potential for additional views of the content, as videos with similar tags will appear in “related videos” on most video sites.

Question 2: Why stay engaged?
With limited time, and the fight for attention, once the user choses to interact with content, you have to give something in return. Normally, a user decides if they will keep engaging with the content within 10 seconds. So, this is the time-frame you have to use as best as you can. Something interesting must happen, something that makes the user stay engaged. In tv-series and other fiction, a “cliffhanger” is often used to keep viewers interested in the next episode.

The same concept can be used in online content, to keep the users engaged. Build “cliffhangers” into the content right from the start. You can even use cliffhangers in the title.

For online games, long load times is a no-no. For videos, the first 10 seconds are essential. Give the users a first payment for investing their attention and time. In general, the shorter the content, the easier it is to keep the users engaged until the call to action. Unless they stay engaged to the end, they have no reason at all to pass it along. Shorter content is usually better.

Usability is also essential. If a user doesn’t understand how to engage with the content, they will stop trying.

Question 3: Why act/interact?
For content to be passed on, there has to be an expected positive experience for the receiver. This means content should be:

• Funny,
• Spectactular,
• Provocative, (but not too much)
• Unexpected,
• Relevant,
• Taboo (but not too much)
• Highly informative,
• Secret (but not too much)
• Controversial
• True? (is it true?)
• New

The one question to ask here is: If I received this clip, would I want pass it on, and if so, why? If you can’t come up with some clear reasons why a user should want to pass it on, they probably won’t either. If someone has deemed the experience worthwhile, there is a chance that they might want to share it. Being  first with the latest creates respect among peers, especially if the content is really good. Passing on information is part of what ties us together with our friends and family, and good content makes the cut. Someone passing good content to their friends will gain in esteem, which is the ultimate reason for viral spread.

But, this will only take place it it is reasonably easy to do so. Which takes us to question number four.

Question 4: Why not act/interact?
The reasons for not interacting with content can be many. Top of the list is of course when there is no real incentive to pass it on – basically, the content isn’t interesting enough and the expected value to the receiver isn’t high enough. We already covered that.

But, even if content is good, there are still many reasons why you might chose to not pass it along. These have to be addressed to maximize the possibilities that a campaign goes viral:

• Is it too commercial relative to the quality of the content? Specifically, think of:
  • logos,
  • urls,
  • the content itself.
• Is there too high an effort to pass it on?
  • Long urls that break
  • Deep-links not possible
• Is there a non user-friendly process?
• Is it too hard to understand?
• Is too much information required to pass it on? Generally, personal info, like:
  • Name
  • Year of birth
  • Phone number
  • City of residence
  • Zip codes
• Are all options to pass it on available and easily accessible?
  • Embed
  • Post to Facebook, twitter etc.
  • Link
  • Download
  • Send to friend

In general, is the effort or investment needed to pass it on bigger than the pay-off? If so, the material has to be re-worked in terms of quality of content, ease of spread, or possibly both.

Multiple audiences
Something to really think about when it comes to launching campaigns on the Internet is that they almost always have multiple audiences. There is of course the target audience of the campaign, often a regular user, who passes good content on. But, there are also the webmasters, bloggers and other power users that can act as “gatekeepers” to the content, often boosting the number of views if they accept the content. These work as springboards and are extremely important to create a successful campaign. Of course, they have to be taken into account when creating campaign material. Some issues they think about are:

The content:

• Quality of the content.
• Technical set-up, like:
  • Plug-ins needed?
  • Software upgrades needed?
  • Java needed?
• Size of game/video in pixels
  • 400-500 pixels maximum width – preferably sizeable
• Difficult to embed games on site?
  • Embed code available?

The user:

• Load times – too long?
• Language of content – does he/she understand?

In general, go for the smallest denominator. The less that is required, the better. Try to make the material flexible, easy to integrate on sites and easy to spread. If a webmaster doesn’t understand the content, either because it is not good enough, it’s too complicated to use or integrate, or language is too complicated, you will loose out on a lot of potential users.

(Yes, this is actually a two year old post I made on my old blog, slightly updated to be more accurate today. Things change fast on the Internet!)

Metaphors are great to make something a lot easier to understand. The constant mantras in social media circles telling businesses in general and customers in particular are pretty easy to understand. Some examples: “listen, learn and interact”, “don’t be afraid of losing control” and “people will talk whether you are there or not. Better be there.”

Still, as easy it is to understand, they lack the kind of power that the metaphor has.

Andy Sernovitz makes it a lot clearer already in his post title: If you were afraid of bad press would you shut down your PR department?

“If you want to increase control in social media, you need to do more social media outreach so people want to talk with you instead of about you.”

Well, it is hardly news anymore that people trust “someone like me” the most. Nielsen’s recent study just confirms the picture. However, it is interesting to know that the numbers are actually going up. In this study, 90 % stated that they “completely” or “somewhat” trust recommendations from people they know. Jonathan Carson, President of Online, International, for the Nielsen Company, says:

“The explosion in Consumer Generated Media over the last couple of years means consumers’ reliance on word of mouth in the decision-making process, either from people they know or online consumers they don’t, has increased significantly.”

Second to “recommendations from people known” comes “consumer opinions posted online”, with the same percentage, 70 %, as “brand websites”. Interesting to know though, is that there are quite large regional differences. For swedish readers it might be extra interesting to know that we are the ones who trust brand websites (40 % compared to 70 % on average) and brand sponsorships (33 % compared to 64 % on average) the least.

As consumer generated media continue to grow, I expect trust in consumer opinions posted online will grow too, especially the aggregated kind in the shape of different average ratings. Quantitative data gives a fuller picture, and in combination with qualitative data in the form of written reviews it might move the percentages upwards even further.

Via: Church of the Customer