Lightweight IFTTT Makes a BIG hit!

Being as one of Tim O’Reilly’s web2.0 patterns, Lightweight and Scalability can be applied to both business model and technology. Beginning with a scalable and cost-effective approach has the following benefits:

  • Shorten the time from development to the market
  • Reduce time and cost = Faster ROI + Less risk
  • Greater adaptability

However, in order to achieve the success in web2.0, effectively scale and pivot the model to cope with ever changing environmental conditions are the key. The sooner we can recognise potential risks and failures and make change accordingly, the better we can survive.

Let’s take a look at the IFTTT, the lightweight and emerging website which offers amazing approach for mashing web services together and creating convenient outcomes. See how IFTTT managed to achieve great success just in 2 years (one of 50 Best Websites from Time)and what are their best practices:

Started up with only 2 friends and an awesome idea, IFTTT is a new website established  in 2010 offers simple but effectively approaches for users to connect web services, and even electronic appliance(Belkin’s WeMo) together, and perform specific tasks. It applies the simplest concept of programming:

IFTTT: how it works

Want to be informed when it’s gonna rain? Want to save all the photos that people tag you on facebook to your dropbox? Simply create your own “Recipe” like this:

Example of recipe: Send me a message when it's going to rain

It is so intuitive that even non-programmer can easily adapt and create their own recipes in no time. With the integration of third party APIs and their native “Channels”, indeed IFTTT provides the way to link all our favorite web services together with innovation and great convenience!

But, how they did it?

Start Small and Scale with demand: 

Small technology, revenue, and human resources model

IFTTT started with 2 friends and a beta version in the end of 2010, soon the beta attracted huge popularity with rapidly growing “recipe” amounts, eventually led IFTTT reach more than 400,000 recipes created with 62 channels, but with only 8 employee.

Syndicate business models, not coordinate:

IFTTT can be regarded as a giant web2.0 plant fed by third party APIs. The more web services that IFTTT can integrate, the more traffic and users. It has successfully created an online ecosystem for users to interact with their own recipes, their social connections and the popular web services. Now they are aiming to launch a platform for users or developers to turn their APIs into “Channels” to boost creativity and opportunity.

The cooperation with ESPN, Philips HUE bulb and Belkin WeMo: multi-device Control further facilitate IFTTT’s syndication with different business models to create more possibilities for cross-field collaboration.


However, Twitter decided to change their policy on how third party applications can access and interact with Twitter’s data. That is , users can longer use Twitter as a valid “Trigger” to perform something like: “IF anyone tweets about Web2.0, THEN send me an email”. Though it is still okay to have “IF I check in on Foursquare, THEN tweet”, where Twitter acts as “Action”.

Market virally without spending money on advertising! 

Since the launch in December, 2010 till now, IFTTT grows bigger with increasing Channels and viral popularity among internet users. Their huge success should attribute to the nature of social media: recommendations from popular bloggers and websites such as Forbes, The New York Times, Wired, Time and TechCrunch

Not to mention that how easy we can create a recipe and share it with our friends on Google, Facebook, Twitter, LinkedIn and many other social media platforms.

My Suggestions for IFTTT:

Though IFTTT is still in its early stages, it is better to start to further improve their current business model, for how to make profit. Possible approaches will be:

Premium and Customised Recipes: users can easily have their customised recipes done with extra fees. It would be awesome if there exist one platform where I can integrate all my recipes together and perform more complex task such as: IF the temperature is below 15 °C, THEN turn on the heating 10 mins AFTER I check out from campus. 

Cross-sell and collaborate with other business:  Strategic collaborate with other online service and encourage users to sign up for a new service through IFTTT’s Channels. Possible partners include public transportation API, Cinemas and even restaurants.

Reference List:

IFTTT, A Glue Gun For Sticking The Web Together, Raises $1.5M Seed Round From Top Investors, Eric Eldon, Jan 04, 2012
How to Never Miss a Deal with IFTTT, Thorin Klosowski, Mar 15, 2013
Twitter Screws Up For The Rest Of Us, Erik Deckers, Sep 21, 2012
10 IFTTT Recipes to Optimise Your Business, Stephanie Buck, MAR 29, 2013. 
IFTTT: San Francisco Startup Lets Anyone Control The Internet of Things, Elise Ackerman, Sep 23, 2012

What Is Web 2.0, Tim O’Reilly, Sep 30, 2005


tumblr and its Long Tail


The term “Long Tail” got popularised by Chris Anderson mentioning Amazon, Yahoo! and Apple are businesses which apply this strategy in his article. To put it shortly, Long Tail theory is about:

As the cost of production and distribution fall, the focus on mainstream products and markets at the head of demand curve has shifted toward a huge number of niches in the tails

The benefits of applying long tail theory include: it gives organisation the ability to target potential micro-market while reduces the cost of distribution, inventory and sales, and provides infinite shelf space. It also helps contents producer to reach wider audience and increase customer choice.

Being as a rapidly growing microblogging and social networking platform, Tumblr has many competitive advantages over other similar services such as: it allows users to write longer posts than Twitter and delicious, and most importantly, in an intuitive, easy and quick way.

tumblr001Established in 2007, Tumblr has experienced steadily growth and the dramatic jump happened between 2010 and 2011, led Tumblr to achieve over 110 million blogs with 72.9 million posts were made everyday.

How Tumblr applies Long Tail Theory?

Tumblr was built on the driving force of Long Tail:

The ecosystem of Tumblr is different from other Facebook, Twitter, or LinkedIn, where users interact with others by their real identities. Most of tumblr’s users regard the platform as a stage for expressing themselves publicly. With the help of tumblr’s content policy, miscellaneous categories made Tumblr a place for discovering one’s interests without many limitations. According the the statistic data measure by Quantcast, the major audience group is young adult (aged 18-34), and the most common interests are: humor, adult content, fashion and news. While the tail part of interests which are not in popular demand still makes massive contribution to tumblr’s daily traffic, the key is how to make the tail fat and thick.


Customer self service + Participation architecture = Reduce cost + High Customer Satisfaction

The success of tumblr is greatly based on user generated contents and its diversity to meet different customer expectations.  Users receive a lot of freedom on the creation of contents and controlling their accounts without much support, while tumblr can enjoy better customer satisfaction with lower cost of account management.

My suggestions for tumblr:
1. Optimise yourself with tons of key words for Search Engine (SEO)

Though tumblr has more than 40% direct traffic, in order to reach mote potential users, it is better to have the related posts or blogs shown on the results of major search engines. Compare to other blogging platforms like WordPress or Blogger, tumblr’s blogs indeed receive less favor from Google, Bing and Yahoo!. Tumblr’s blogs and posts are only visible when “tumblr” is included as keyword.

2. Dig into adaptive personalisation and customisation

In order to boost the interaction among users, deepen existing connections and expose active users to more potential followers, tumblr should launch the adaptive personalisation approach by adding customised suggestions like:  “you may be interested in this post/blog” or “users who follow this blog also like…”

3. Be open to encompass new contents, and stay open (Update!)

Tumblr was acquired by Yahoo! for approximately $1.1 billion on 20 May 2013. Though Yahoo! pledged not to screw up tumblr deal, the acquisition still received negative responses from tumblr users. “In order to keep the richness and breadth of the contents on tumblr and allow Yahoo! to reach more users”, CEO Marissa Mayer suggested that “tumblr’s porn can stay“. It is no doubt the best decision for coping with angry users at the early stage of merging tumblr into Yahoo!’s culture, however, as a nature of user generated contents, it seems unavoidably every social media platform will get dirty with time. How to keep the niche market growing within ethical and legal constraints but without losing specific user groups is definitely one of the biggest challenges for all the social media platforms.

Reference list:

* Usage Analysis: Tumblr, Brendan Smith, Apr 16, 2013
* Oh, The Places Tumblr Can Go, Ingrid Lunden, Jan 02, 2013
* Tumblr: David Karp’s $800 Million Art Project, Jeff Bercovici, Feb 01 2013
* Tenth Grade Tech Trends, Josh Miller, Dec 29, 2012
* Web 2.0 and the Long Tail, Chris Anderson, Oct 01, 2005
* Microblogging update: Tumblr feed, Chris Anderson, Sep 04, 2008
* 6 Ways to Leverage the Long Tail in Your Marketing, Prashant Kaw, May 5, 2009

Digg into Perpetual Beta


The concept of Perpetual beta, coined by Tim O’reilly within the explanation of one Web2.0 pattern, end of software release cycle , describes as software becomes ongoing services, “the product is developed in the open, with new features slipstreamed in on a monthly, weekly or even daily basis”, and “engage your users to be real-time testers, and structure the service to reveal how people use your product”.

Being as the once wildly popular social news website, Digg had experienced both the high time with numerous copycat websites featuring its story submitting and voting system, and also the disastrous user revolt toward its notorious release of Digg v4, and followed by  dramatic decline in web traffic and fierce competition from Facebook.

After the acquisition by Betawork, Digg had its second act successfully regain popularity among social news community by rebuilding the website from scratch. Through the launch of, digg invited users to participate in the decision of user interface design and user experience by answering the survey. Based on user’s idea, the new digg v1 came back with newspaper-like front page and more images, new story ranking system integrating Tweets and Facebook likes, to be counted with its traditional thumb-ups, “diggings”.

Critical Success Factor comes from frequent and early release

Through its nine-year long run, Digg has its new version released almost every year between 2004 to 2007. however, despite the “extensive overhaul in 2010“, it took nearly 3 years for Digg to introduce version4, which was, eventually proved to be an death knell for Digg. Luckily enough, Digg got a second life and came back to the market after 6 weeks sprint transformation by the new team. The new team plays well with the concept of Perpetual Beta by frequently introducing new features and fixing bugs almost every week, and so do its mobile apps.

Actively engage users in the development and test cycles

The new Digg team is very keen on extracting the “Digg experience” from its users. Even for the new Digg Reader, users are invited to have their says and participates in the decision of core features and UI. Since only users know what kind of service and features can meet their need, the developer can rarely go wrong if following closely with user requirements and expectations. Plus users can perform the task to test the platform, in real time, and feedback straight away. Thus the process is shorten and the efficiency is improved.

My suggestions for Digg:

Capture the implicit facts that users interact with Digg

I got to admit that the new Digg team really has done a good job by getting Digg back to the old good times. And indeed, surveys and questionnaires do offer a lot of essential information about user experience and provide directions for improvement. However, what users do often tells more than what they say: according to Digg’s survey about what topics that users enjoy to read most, the result is as follows:


However, the statistics measured by SimilarWeb, suggest another thing:



Interestingly, part of the users who visit Digg also like adult websites. But the option is not even on the list of hot topics voted by the users. Apparently observing user behaviour and capturing the implicit usage data can reveal more facts and possible approach to boost website traffic.

Right before the new Digg team started to launch the overhaul, they conducted a survey and which indicated a stark statistic that: 92% of Digg users would NOT recommend Digg to friends. And straight after the big triumph they made, they happily reported that according to the survey done after the relaunch, now 81% respondents said they would recommend Digg to a friend. But, why can’t they just add a feature like “invite your friends to Digg together” and see how many users actually do it? Maybe the survey is simply an indicate of user satisfaction, but I believe there exist more user behaviour facts awaiting Digg to “dig out” and act on it. In order to compete with Reddit, SlashdotStumbleUpon and other social news websites, Digg still have a long way to go.

Reference List:

SXSW: Digg’s Big Redesign Taps Into Social Web
* Digg – Yes, That Digg – Is Building A Google Reader Replacement, Complete With API
* New Digg Vs. Reddit Vs. BuzzFeed: Your Mind On Viral Content
* Digg’s replacement for Google Reader due in June; might cost money
* What is Web2.0? – Page 4,  End of Software Release Cycle