The Social Media Channel


App.net launches Backer, a service to help startups and developers crowdfund new features

Posted: 25 Jan 2014 12:48 PM PST

tnw 645x250 520x245 App.net launches Backer, a service to help startups and developers crowdfund new features

App.net, the social platform by Dalten Caldwell has launched a new crowdfunding service called Backer that allows companies to ask users for money before committing to build software features.

The idea behind Backer is that when a company wants to add a new feature to its software, it should be able to test just how much users desire a hypothetical feature by asking them to commit money to the functionality being built before development even begins.

App.net believes that this method works better than traditional methods of gauging user interest because "running a poll or doing informal research can easily lead to false positives" whereas asking users to commit money leads to more reliable results.

Backer comes after App.net launched a trial run that asked its users to commit Bitcoins before the service would add subscriptions to be paid with Bitcoin. Backer is now a platform of its own and App.net fully hosts the service for anyone who wants to use it. As with other crowdfunding services, if funding isn't reached for any campaign on Backer before the end date, all money is refunded.

Applications are now open for those who want to use Backer to crowdfund a new feature they have in mind.

This posting includes an audio/video/photo media file: Download Now

Tumblr drafts new terms of service that encourage attribution, ban impersonation and more

Posted: 25 Jan 2014 12:12 PM PST

shutterstock 171257741 520x245 Tumblr drafts new terms of service that encourage attribution, ban impersonation and more

Yesterday, Tumblr published a draft revision to its terms and policies to Github for reading by the public before it updates them on the site. The company said in a blog post that the updates mark the first time in two years that the service will be updating its terms of service, as well as the first time they've been tweaked since the company was acquired by Yahoo last year.

The updated terms are mostly just tweaks to the legal wording but the company has reworked their recommendations on how to be a good Tumblr user. Notable additions include the company asking that users attribute each other when reposting content, rules against impersonating companies, personalities or other users and encourages users to discuss hateful content rather than have the company remove it.

The company obviously hasn't had its fun side squashed by coming into the Yahoo corporate fold, since the documents still contain the occasional expletive and even calls those that torture animals 'dicks'.

These 'guidelines' are more like rules. Tumblr says that if users are caught repetitively breaking the guidelines, they may have their accounts suspended or be blocked entirely. It's interesting that Tumblr has added them now, as it appears that the company is attempting to clean up user habits across the site going forward.

You can see the changes for yourself compared with the current version of Tumblr's terms of service here on Github.

Feature Image: Gil C/Shutterstock

This posting includes an audio/video/photo media file: Download Now

More than messaging: Why you should stop comparing WeChat to WhatsApp

Posted: 24 Jan 2014 01:46 PM PST

wechat whatsapp 520x245 More than messaging: Why you should stop comparing WeChat to WhatsApp

Rui Ma is a Venture Partner at 500 Startups, a leading accelerator and seed investment fund in technology companies.


Tencent's trailblazing WeChat app is often referred to as the "Chinese WhatsApp." But WeChat is not just another messaging app. It is a messenger plus social network plus social media, all on top of an extendable transactional platform.

To see why WeChat can potentially be a much bigger business than the traditional chat programs, it's only important to understand how WeChat users interact with it.

I am only going to focus on China, because that's where I live and the market with which I am familiar, and since some of the features that make WeChat completely kickass in China may not necessarily work well overseas.

And also because 300mm users in a smartphone market that grew 79% last year to over 400mm is probably sufficiently interesting to warrant its very own analysis.

Before you discount my assertions that WeChat is innovative, in useful, logical, and thoughtful ways, I would like to give you some background on my experience with social media and interactive apps/sites — my credentials as a user, if you will.

In the US, I am a long-term user (although not necessarily lover) of Facebook, Twitter, LinkedIn, Instagram, and WhatsApp. In China, I have used, with varying regularity, Renren, Kaixin, Weibo and Momo before WeChat.

I also use Viber and Skype frequently on my mobile – Line as well, but far less frequently. I'm not one of those investors who download every app I hear/read about or am pitched — but I have been on WeChat since it was way under 20mm users, and on average, I'm guessing I open it more than 50 times a day.

Wow! 50 times a day is a lot right? Yes and no. As this recent NYT article notes, it seems that I am in good company — plenty of Chinese people are on WeChat everyday … some for hours.

So here we go: the reasons why WeChat is much more than just another WhatsApp-type messaging app.

WeChat is optimized for Chinese consumers

Hold to Talk

WeChat began with a hold to talk feature that some consider a gimmick.  This feature isn't quite as relevant for Western users, but for Chinese people, it's actually quite useful.

Why? Simply put, the QWERTY keyboard is not optimized for the Chinese language. Of course, there are input methods that are faster than typing in pinyin (which is based on the 26 letters of the English alphabet), and modern software has made significant leaps in predictive and fuzzy logic.

Nevertheless, even for those that are fluent in both Chinese and English, it is more difficult to type in Chinese. Ease of use can convey a huge competitive advantage, a fact that Sogou leveraged with its superior input method, which enabled it to subsequently launch a billion-dollar search engine business.

Without going into too much detail, the core differences are:

  1. Chinese typists must type the pronunciation then select the appropriate character, which is not sped up by that much even with a lot of the aforementioned logic upgrades.
  2. Pinyin is based on Mandarin, and many in China often speak a dialect that has different pronunciations which don't translate well to pinyin. Additionally, many have never even learned pinyin.
  3. The English language is conducive to abbreviations (i.e. ASAP, OMG). This advantage doesn't really exist in Chinese.
  4. Users who need to mix languages need to toggle between different input methods. I have to do this frequently, and it is slow and painful.

So aside from the fact that there are obvious cases where voice can deliver more convenience to the speaker and greater sense of intimacy to the listener, WeChat, just by putting in push-to-talk, had already provided more utility to users than SMS and even its own QQ Instant Messenger.

While I don't use voice as much as I used to (my contact list was only close friends as opposed to nearly 1,400 strong), I notice that it is significantly easier for my computer engineering mother and my computer literate grandmother to use voice over text.

QR codes

WeChat incorporates QR codes into many elements, such as adding a new contact, or entering into a group chat. Prior to WeChat, I had never scanned a QR code in my life. But in China, I began to see how it is actually useful.

The Chinese language uses characters, not Roman alphabets. This makes it not only difficult for the Chinese to pronounce words, but also individual letters. That explains why a lot of Chinese URLs have numerals, such as this famous finance site.

Even my WeChat ID, "missruima," which really just contains my name in pinyin, is difficult for people to spell correctly.

Solution? Scan the following code and do away with all manner of typing.  In fact, I've begun to include it at the end of every PPT presentation I make, and it's a funny sight to see at least a dozen or so hold up their phones and send me a WeChat request right then and there, via QR code.

pngbase64bf7da9e64ab88637 More than messaging: Why you should stop comparing WeChat to WhatsApp

WeChat is a social network

Despite many attempts, a dominant social network in the sense of what Facebook is to the U.S., for example, hasn't really emerged in China. It hasn't helped that neither Facebook nor Twitter are accessible in China.

But that means there was a vacuum in the market, and so … enter WeChat Moments. It is, hands down, the simplest photo-sharing experience I can think of for a non-photo-sharing app. It lets you put in text (there might be a limit but so far I haven't hit it) and if you want, up to 9 photos, mention up to 10 friends, and specify your location if you prefer. Your friends can Like or Comment like any other network out there.

wechat moments More than messaging: Why you should stop comparing WeChat to WhatsApp

That's all Moments is for, and that's all it does. You can't tag faces, and your photos don't get sorted into unwieldy albums. You don't have to put in descriptions for each photo (in fact, you can't).  It's aptly named "moment" because it's just that, a stream of posts (photos or other content) from your contact list.

Sure, Facebook status updates and photo sharing accomplish a lot of the same, but I actually think Moments allows for a better social networking experience. Here are the 2 main reasons why:

1. Your contact list is not visible. Neither total number nor content.

Why's this important? Because it keeps a sense of privacy, intimacy, and authenticity. It doesn't degenerate into a popularity contest of who has more friends, which friends do they have, or how many friends do we have in common. You get the point.

Also, this destroys zombie accounts, which plagues Weibo. There's no point in buying fake accounts because no one can see how many friends or fans or likes or anything really, that you have, unless they are already on your contact list. Everything is shaped by that initial authentic and private one-on-one interaction.  And it really is quite private.  Because different from Facebook …

2. You can only see the interactions of those on your contact list (i.e. your "friends").  That's it.

This matters so much. Let me explain, because it confused the heck out of me initially.

Suppose I put up a photo, which receives a comment from my good friend Jack. Jack doesn't have any other friends in common with me and makes a totally obscene inside joke. No one else but me sees the comment, not unless they are also friends with Jack.

My friend Jacqueline writes a comment, as does my friend Jeff, and they are friends with each other. There are 30 other likes on the photo but none of them are friends with Jacqueline or Jeff. What do Jacqueline and Jeff see? Just each others' likes/comments, and my responses (unless directed at others not on their contact list). They do not see the 30 other likes or 10 other comments from people they do not know.

What does this create? In the case of Jack, it creates the opportunity for me to have an intimate conversation without having to especially go initiate it (post to Jack, then post to "public"). In the case of Jacqueline and Jeff and myself, it reinforces a mini circle of trust, while cutting out the noise of strangers.

This creates more authentic interactions all around. It alleviates pressure on the poster (are people going to engage with what I post? who cares?) as well as the commenters (I can be myself, and I'm not competing to be wittier than the next person).

Basically, what this very simple mechanism does is it is a sort of proxy for "relationship closeness" via interactions amongst shared friends. The primary implication, in my opinion, is that I feel much more in control of my network, and will add even those who are much more loosely affiliated with me (i.e. "Twitter followers") without feeling like my inner circle of more intimate relationships is subject to voyeurism from those I do not know (as) well.

In other words, it allows me to fold my Facebook and Twitter networks into one without feeling overexposed (hence my willingness to put my WeChat account on my business card and all my presentations, but not my phone number or Facebook ID).

Are you beginning to see what I mean? WeChat is far more than a messenger, and honestly, if it had stopped at just being an awesome social network, I would've already gawked in awe, but yes, it's even more than that …

WeChat is a social media

And here is where I think it really gets interesting. WeChat is actually Facebook and a lot of Twitter rolled into one.

WeChat started off with your real-life phone book. How did it do that? Partly, I think, by chance.

But let's break this down — why is Twitter valuable, to the user? Well, to me, it allows me to see what people who share my interests are doing, reading, saying. Occasionally, we interact in direct conversation and end up sharing contact information for deeper dialogue. But mostly, we circle around a pretty defined set of topics, which may arise because of shared geography, hobby, or any marker, really.

In WeChat, these are themed group chats and official accounts.

1. Group chats

WeChat 5.1 defaults each user to be able to create groups of 100 people (it used to be 40). The current maximum is 500. Sounds unmanageable? Not really.

Firsthand experience tells me that most of my groups below 200 members are quite productive and manageable, especially if there is a motivated moderator and clear theme. I have about a dozen organized by different angel investor groups and entrepreneurs, a few for the schools I've attended, as well as interests and entertainment.

These roughly reflect the topics I'm interested in keeping up-to-date about, and are not that far away from my Twitter follows. Except it's already curated for me. Am I going to Shanghai soon? Maybe I'll keep an eye out more on that room. Or am I really nervous about my Bitcoin holdings? Then better see what the other bitcoiners are saying.

And what do people share? Links to good reads, photos and videos, just like what I mostly click on via Twitter. WeChat has done an amazing job of making the groupchat and sharing experience pretty damned easy.

And if I don't have anything immediate to say? I can just stay in the room and "listen."  I wrote in an earlier article that I often heard news first in WeChat before I read it elsewhere. It's true. That's about half the time.

The other half of the time I notice I'm reading a lot of articles from a particularly awesome media source (or individual writer/curator) and I add their …

2. Official accounts

Two types of "official accounts" exist on WeChat – service and subscription (RSS feed). I mostly interact with the subscription accounts for now, which is exactly what it sounds like – subscribed RSS feeds. Anyone with a PRC ID can register for one, although it is not an automatic approval process.

I am, for example, registered with all the major Chinese tech media, who usually push out content once or twice a day of their top 2-5 headlines.  I can then read the (mobile-optimized!) article directly in the app. I can share this easily with my contacts, group chats, or straight to my Moments and include some witty remark.

WeChat is a platform

[Note: I'm on the Chinese version of WeChat, which seems to be something like two product releases ahead, so I apologize if you open up your English version and do not see any of what I'm talking about.]

China's version of Yelp has been integrated into WeChat forever now, it seems, as are music and video players, and my personal favorite, the air quality index app. There is a social gaming platform, which shows you how many of your friends are playing each game, compete against them as a matter of course, and also allows you to send invites to them via WeChat.

As of this writing, the top 3 games on the iOS store in China are all WeChat-"enabled" games.

WeChat has also launched all sorts of mobile wallet functions, allowing you to pay for more mobile minutes, movie tickets, split bills, or buy select products.

I don't personally use any of them (yet), but initial traction seems strong. Taxi-booking app Dididache made $170K in bookings in the first seven days, travel meta-search site Qunar achieved 10 times that amount in its first ten days, and smartphone maker Xiaomi sold 150K phones in 10 minutes.

Anything that allows people to plan out and pay for (transact) what they will do when and with whom (communicate!) on-the-go (mobile!) is probably up for grabs. No wonder fellow Internet giant Alibaba feels so threatened by WeChat, and has repeatedly taken measures to defend against it.

Conclusion? WeChat is not just a WhatsApp competitor

WeChat is much more than a messaging app, it is social network +  social media + interest/news feed + rolled into one on top of a platform that was (I believe, given Tencent's DNA in gaming) always meant to monetize off of real transactions – whether they be micro such as in-game virtual goods or more substantial such as traditional e-commerce.

And for those of you still with me… I just have one last thing to say.

Some users think WeChat is starting to look bloated, and yes I agree that danger exists. However, as long as it stays focused on giving users that authentic communication experience, and focus on connection discovery, maintenance and growth, I am incredibly bullish on the WeChat team to deliver a reimagined and much streamlined mobile communications and transaction experience.

WeChat vending More than messaging: Why you should stop comparing WeChat to WhatsAppDetail-oriented readers will note that I spent an inordinate amount of time trying to explain why WeChat excels as a social network and media, and only briskly went through its platform capabilities. That was intentional. Distribution for mobile apps is difficult no matter which market you go to, and China is no exception.

I wanted to make sure I understood why WeChat is here to stay, its central role in the app ecosystem, and how it's justifying its existence on my phone as by far, the number one used app (aside from native e-mail or browser client) for over two and a half years.

What were the thoughtful details that made it such a good experience? What were the design features that made it incredibly sticky and useful to a Silicon Valley-bred brat who readily had access to more "mature" social and communications products (WhatsApp being one)?

Once I worked those out in my head, I could see that a lot of possibilities existed to exploit the new crop of activities and behaviors that WeChat was enabling and training its hundreds of millions users to do; things I could actually bet on, as an early-stage investor.

We'll see how open the WeChat platform ultimately becomes, but the current prognosis is good, and if I had to give a forecast, I'd say it's sunny and clear.

Embarrassing Disclaimer: I was not hired by Tencent to write this op-ed piece. I have never been paid by Tencent, directly or indirectly, to do anything (unfortunately.) My chance to have had that happen was about 3 years ago, when I interviewed for Tencent's corporate development group.

At that time, WeChat was about 20mm users, and Weibo was worth something like $1.5Bn, or so I read somewhere in some lousy Wall Street report. After some trivialities, my interview with the M&A Director consisted solely of the questions – how much is WeChat worth and what would you do with it?  I don't even remember if I got to the second question because I answer "$300mm" to the first and knew I had immediately lost the interviewer's interest.

What was I thinking?! Well, it is clear I wasn't. If you think I have grown some brain cells since then, please add me on WeChat — my handle is "missruima" (or you can use a WeChat guru tip and scan the QR code mentioned before).

Featured image credit: Shutterstock/dolphfyn

This posting includes an audio/video/photo media file: Download Now

“How we grew our daily sign-ups from 5 to 25 in six months”

Posted: 24 Jan 2014 09:29 AM PST

164665100 520x245 How we grew our daily sign ups from 5 to 25 in six months

David Arnoux is Head of Growth and co-founder of Twoodo. Twoodo helps you organize everything and anything using #hashtags, giving you headspace to focus on what's important.



Six months ago, we went from five daily sign-ups to 25. As a SaaS startup building a team collaboration tool we developed a cost-effective content marketing strategy, going through the mill of guest posting, social media marketing, content creation and applying various growth hacks.

After an intense three months at Europe's premiere accelerator program, Startupbootcamp Amsterdam, our product was finally taking shape and we had reached product-market fit. The next step was to start acquiring more traffic.

It was important to increase sign-ups to please our investors by starting to scale but also very importantly to keep testing and fine-tuning our landing page, sign-up process, on-boarding and activation rates. We also wanted to start generating buzz by earning more early loyal users that would convert to evangelists in the future. Nothing new here.

After six months we had increased our unique visitors by 295 percent but most importantly, we increased the number of sign-ups by 400 percent.

The growth curve is steadily rising per month – not a once-off peak and then high drop-off rate. This is a healthy state to be in.

Part 1: Assume nothing – do your marketing research

It is popular knowledge that early-stage SaaS startups are not the sought-after fodder of top tech journalists. There's a hell of a lot out there. So we didn't start off naively assuming that we'd be appearing on Inc.com any time soon.

This is how we chose guest blogging as our primary strategy. Via a link in the author bio, plus offering good tips and advice, we hoped to draw people to Twoodo. It became, and remains, our most effective growth strategy. It wasn't easy getting started, though.

Finding niche blogs with a decent engaged readership that matched our tested personas took a long time to research. We used Technorati, top ten blog lists and Google for hours on end trying to discover the right places.

technorati 730x103 How we grew our daily sign ups from 5 to 25 in six months

This involved:

  • checking the last 20 or so blog posts to see if the topics were relevant to readers we wanted to target
  • checking how much engagement there was by looking at the level of commenting for at least five recent articles
  • checking their social media following on Twitter, Facebook and LinkedIn for activity (our targets were mainly Twitter and LinkedIn for a B2C2B strategy)
  • checking that they accepted articles
  • checking that if they did accept articles, could we have a backlink (if not, there was no point for us to write for them)

We wanted 30 to 50 blogs where we could pitch articles to, since different blogs require different topics and styles. The team was new to this, so we weren't sure where our knowledge would fit in best. We also wanted to have a broad reach to many communities.

What we learned: 

  1. Few blogs have an engaged readership
  2. Many of the best blogs are small company blogs, not general news blogs
  3. There are a lot of liars out there (seriously, not all of you can be the "top," "best" tech blog)
  4. SEO tricks diverted us away from the good material to the mediocre in our searching
  5. Places that beg for guest blog posts or rely completely on guest posts are not where you want to spend your time
  6. People often don't take down dead websites
  7. In the end, getting on the big publications of the world really is the best for a return on your writing time since they are reputable, have a huge following and have plenty sub-categories of topics to fit into

Part 2: Distilling the data into useful stuff

Quicksprout is one of the greatest blogs on growth, and this article by Neil Patel gave us the information we needed to act upon the data we had gathered.

From the extensive research on blogs, we not only had a list of blog domains, but also bloggers, their contact details, their topics of interest and a sense of their preferred writing style. We set up a series of spreadsheets to manage all the information.

  • First, we ranked the blogs from 1 – 4 (1 being the most desirable to feature on, 4 being the least)
  • Second, we collected the names and emails of targeted journalists and editors on each one, noting which ones were assigned to deal with the guest posts
  • Third, we collected their Twitter handles and linked up with them
  • Fourth, we separated those that certainly accepted guest blog posts to those that were ambiguous (and of course, deleted those that never accept guest posts)

What we learned:

  1. You'll know you've figured out the niche when you notice the same names appearing over and over. The people in these circles will often reference each other and use each other's services.
  2. You should hit them up on Twitter when you have something interesting to share, and ask their opinion.
  3. Support their organizations, events and so on by liking them on FB, [re]tweeting their content and commenting on their posts.
  4. If they use Pinterest, you use Pinterest. If they use Quora, you use Quora. Follow them and learn more about the relevant places for you to be.
  5. Find out what they DON'T like and avoid it.

We created templates for everyone to use here.

template content mar 730x172 How we grew our daily sign ups from 5 to 25 in six months

Part 3: Content experiments

There are multiple forms of content marketing that you can test, but ultimately you will have to choose just a handful so that you can execute them well.

A scattered and inconsistent presence over too many channels is a waste of your time. Thankfully, there are tools to help.

It's hard to dismiss any single channel until you have tested it thoroughly, but if your team is of a limited size you may have to do that in the early days. For example, based on our personas and early users, we assessed that Pinterest was not a useful channel for us (80 percent women, brick-and-mortar sales).

Preferably, you would run data-driven experiments with the right tools (We used Mixpanel and Google Experiments etc.) but you can also tell from social engagement if a channel is successful or not (the same with a guest blog post). In the end, we have found Twitter the best platform for sharing content, engaging with people and growing our community.

This is not so surprising for a SaaS company! This is based on the speed of how our follower base grew and the level of interaction.

Guy Kawasaki has one of the largest social media followings in the world, and based on a webinar I attended, he attributes it all down to sharing excellent content. For keeping your channels on Facebook, LinkedIn and Twitter busy, try tools such as Buffer or Hootsuite.

Naturally, we used the list of blogs we had researched to choose our content and occasionally notified authors when we shared their content via including them in a tweet. We also use AllTop, ContentGems and Prismatic to watch out for new and interesting blogs, posts and bloggers (though I did not know about them at the beginning!) as well as more content to share.

blogs content market How we grew our daily sign ups from 5 to 25 in six months

What we learned: 

  1. Sharing content without reading it first is going to catch you out. If you tweet 10+ times per day then you are going to have to read a hell of a lot. Some people say you shouldn't be on Twitter and only tweet a few times a day, but since finding and reading lots of good quality articles/infographics/slideshares etc. takes time, you have to make that decision.
  2. Learn how to hashtag – use this guide from us on The Next Web
  3. The same content rarely works across all channels
  4. Focus on a few good communities
  5. Keep a minimal presence on the big channels (FB, G+, LinkedIn, Twitter, Flickr/Instagram, Slideshare) so as to show an effort – it makes a difference to some people, and as your company grows you may find that a channel activates that didn't work before
  6. For SEO, be on G+
  7. Make an effort to contribute to the most important communities to your business. When the time comes to ask for a favour

Part 4: Take advantage of everything

This is one of my personal growth hacking definitions. We tried out many different tricks depending on the situation – in some respects, you have to be shameless and "do what it takes" to get noticed! There are many "free" tools out there for the content marketer to use (I say "free" because you do need to spend precious time).

On Twitter, we grew the social media accounts of the CEO, CMO and growth guru. The CEO account curated people interested in productivity and work-life balance (the core philosophy of our company).

The CMO account focused on marketing and social media enthusiasts (since our tools is about social collaboration). The growth guru targeted growth hackers (we are going through a lot of growth experiences now and write about it frequently).

This meant three times the audience because we could appeal to three different communities. Remember not to share the same content to all these different groups – a productivity person is not going to care about user-testing tools.

We recycled old content (usually old blog posts). This maximizes the ROI of the time spent preparing great content. As mentioned previously, other channels may open up and this is a great way of getting on board without having to prepare completely new content.

  • The best of the images used (that we have copyright to) are put up on Flickr or Instagram.
  • Infographics could be made from some blog post material.
  • We use the text and images to create slideshares.
  • We make comic strips out of our landing page copy to explain the pain our product solves and use them in email campaigns.
  • Once a week we check our Buffer statistics and publish a Flipboard magazine with the most popular shared content.
  • We have plans to use the images to create Vines in the future if it looks like a valuable market.

What we learned:

  1. When converting old content, it is really useful to have some designing chops, or at least to have a good designer at hand.
  2. Every crazy idea is worth testing.
  3. Be alert to new and better tools to help you automate your marketing efforts (without damaging them).
  4. Decide if riding pop culture waves fits your company and audience. This is especially important for visual marketing.
  5. Personally thank as many people as possible when they share your content (use TweetDeck or Hootsuite for this endeavour)

Finally, the results!

  • Over the past 24 weeks our average daily sign-ups rate has increased from an average of 2 per day to around 25 per day. That's almost a 400 percent increase and we don't plan on letting the curb stop there.

Daily average sign ups 730x415 How we grew our daily sign ups from 5 to 25 in six months

  • Close to the top social media influencers according to Klout (a score of 63 and over puts you in the top 5 percent):

klout score 730x295 How we grew our daily sign ups from 5 to 25 in six months

  • Our YouTube video is just about to hit 3,000 views. This might seem low but people who watch the pitch are extremely qualified leads as they have heard what the product is about and really want to start using it.

youtube 730x541 How we grew our daily sign ups from 5 to 25 in six months

  • Reached 1,000 Twitter followers within two months, all over the world:

twitter followers 730x404 How we grew our daily sign ups from 5 to 25 in six months

  • Over 2,000 views on Slideshare

slideshare 730x404 How we grew our daily sign ups from 5 to 25 in six months

  • Access to bigger and better blogs due to our portfolio of articles on past smaller blogs

You might think that the numbers are small, but you're not looking at the bigger picture. These are targeted, high quality visitors with a high lifetime value. Our conversion rates are all above industry average.

Our future Content Marketing goals:

  • Increase our Twitter follows to thousands
  • Explore if visual marketing is worth spending time on
  • Create an improved testing process for content marketing strategies
  • Re-test Facebook and LinkedIn with a new approach
  • Refine the work we are already doing (faster and better!)
  • Release our own blog with a theme unique to our experience and knowledge
  • Increase guest blogging on major websites
  • Discover new growth hacks from fellow startups

What content marketing techniques worked or didn't work for you? Please share!

This posting includes an audio/video/photo media file: Download Now

Facebook’s referral traffic share grew over 48% in Q4 2013, Pinterest was up 30%, but Twitter dipped 4%

Posted: 24 Jan 2014 08:00 AM PST

144134368 520x245 Facebooks referral traffic share grew over 48% in Q4 2013, Pinterest was up 30%, but Twitter dipped 4%

Social discovery and sharing platform Shareaholic today released a report covering referral traffic data from the eight largest social media sites in Q4 2013. In short, Facebook, Pinterest, StumbleUpon, and Google+ all gained, while Twitter, Reddit, YouTube, and LinkedIn lost.

Taking a closer look at the data, however, shows that only the top three are really worth considering. Here's the breakdown from Shareholic, which tracks 250 million users visiting its network of 200,000 publishers:

social media traffic report Jan 14 stats Facebooks referral traffic share grew over 48% in Q4 2013, Pinterest was up 30%, but Twitter dipped 4%

The numbers in the month columns are represented as "share of visits," a percentage of overall traffic (direct, social referrals, organic search, paid search, and so on) that sites received. The change column shows how much each social network's share of visits grew or declined.

Facebook's referral traffic share grew from 10.37 percent to 15.44 percent, Pinterest was up from 3.68 percent to 4.79 percent, while Twitter dipped from 1.17 percent to 1.12 percent. In other words, Facebook and Pinterest showed phenomenal growth given their positions, while Twitter remained essentially flat.

The other five sites all had less than 1 percent share. This is important to take into consideration when you look how much StumbleUpon or Google+ jumped, or the declines of Reddit, YouTube, and LinkedIn. While their changes were all in double digits, their shares are too low in this market pie for those figures to be significant, especially in a single quarter.

social media traffic report Jan 14 graph Facebooks referral traffic share grew over 48% in Q4 2013, Pinterest was up 30%, but Twitter dipped 4%

As is typical in these studies, the real winner here turns out to be Facebook, as you can see more clearly in the line graph above. Last month, the social networking giant drove more than twice the referrals all seven of the other social media platforms sent to sites combined.

See also – Facebook's referral traffic has exploded over the last year, and mobile now accounts for over a quarter and Report: Over the past year, search traffic has dropped while social traffic more than doubled

Top Image Credit: Brendan Smialowski/Getty Images

This posting includes an audio/video/photo media file: Download Now

 
Tips and Tricks Blogging © 2013. All Rights Reserved. Powered by Blogger
Top