The last pattern a successful web 2.0 company should use can be shortened down to ‘do more – with less’. A great example of this is the fact that Wikipedia is the 7th largest site on the internet but it still only has 22 employees. The internet’s cheap distribution methods come from the fact that hardware, software and bandwidth required to start a website are constantly decreasing. Also there is open source software that you can use for your website. Some call it the LAMP stack. Linux operating system on the server, Apache web server, MySQL for the database,  and PHP to code the webpages in. Using free software can drastically reduce the cost of your website. There are also a multitude of libararies and pre-built components that you can use in your website. Using perpetual beta and fast agile development models allows you to scale the project alot easier as well.

All these things help your website to hit the market faster, which helps in getting a ROI(return on investment), along with reduced costs, and greater adaptability. To take advantage of all of these things you need to be able to scale with the demand on your website. Save money by starting out small on a feature and as it becomes more popular than increase its capabilities. This allows you to reduce the upfront costs of a feature that might not get used. This should also allow you to scale your pricing models as more capabilities and features are introduced. A very successful strategy is to produce revenue from multiple sources. Some examples include syndicated advertising (e.g. Google AdSense), sponsered advertising (e.g. TechChrunch), subscriptions, transactions (e.g. eBay), premium services (e.g. Flikr and Flikr Pro), etc.

The example I am going to use this week is Evernote. Evernote’s slogan is “Remeber Everything”. Evernote is basically a note storage website. Evernote can take anything from text to a web page to an image or a screenshot, and it automatically organize the notes for you. You are then able to access it basically through anything that can access the internet. It can also read the text of an image and store that, so for example you could love a bottle of wine, take a photo of it and upload it to evernote. Evernote will scan the image and take the text and remember what wine it was. To find out a bit about Evernote listen to this interview with Phil Libin (CEO of Evernote).

The really amazing thing about Evernote is it’s growth. It opened in Feburary 2008 and it’s user base has grown exponentially since then.

It took Evernote only 15 months to reach one million users. It then only took 7 months to get another million. To reach their three millionth user, it only took 4 months. So how has Evernote scaled to three million users in a little over two years?

The total of their variable expenses in January was only $68,641. Which is astounding to note that they served 2,335,676 total users and 41,598 premium users. Evernote now has 60 servers to host their product and only four system administrators making sure the system is running 24/7. And Phil Libin said that “When we get to 600 servers, we won’t need 30 ops guys, probably 10 will be enough.” So Phil Libin sees the number of users growning exponentially, but the variable expenses to grown in a more linear fashion. This means that the variable expenses per user will constantly decrease as time goes on. This is definitely a great example of lightweight models and cost effective scalability.



Leveraging the long tail as used in Web 2.0 sites is basically  draw people in with your popular products, then sell to them your niche products. This is more easily understood with the diagram:

Long Tail Image

The diagram shows that on the left, the green area,  is the “big head”, which are the few products that make a lot of sales and draw in your customers. The right side is the long tail, which is all of your niche products that do not sell as many as the popular products. The long tail is just as important as the big head becuase it can make up the same, if not more sales. See how the yellow side takes up the same amount of ‘area’ as the green side.

The long tail was previously unavailable becuase it was financially unfeasible to market to such a specific niche. It is now possible becuase of the low cost of production and maintainability of the internet. Such a market could only be reached by a global technology. If for example an online music shop sold a song that only a handful of people in the world would want, they could still make money from it because they do not have the normal cost of distribution.

This allows web 2.0 companies to capture new micro-markets, producing companies to reach a wider audience and the consumers have an increased oppurtinity for choice. The problem then arises where becuase there are so many products that weren’t available before, how do people find what they want? A great example is which has the feature “People who bought this also bought X”. By harnessing the collective intelligence, you can find how to expand a niche or what new niches could be possible.

The example for this week is eBay. eBay leverages the long tail becuase it allows users to sell whatever they desire. eBay sells many popular items, however there are numerous amounts of niche markets withing eBay. Just take a look at their categories page. There are 35 seperate categories, and over 400 subcategories. A traditional physical store  could never offer that many categories to its customers. eBay helps its consumers find what they want with their suggested searches, seller location, sorting by etc.

eBay Pulse shows the current trends of products. The trends can be refined by category and can show popular searches, stores, products and watched items. eBay definitely leverages the long tail by allowing the users to create the niche markets.

What is perpetual beta, and is it good for a web 2.0 company? Beta is a term used for traditional desktop software that is still in production (for example video game developers release a “demo” or a “beta” of their game to the market). So perpetual beta would be software that is constantly still being produced or improved. This was a lot harder when software was distributed through cds. Now with the advent of web 2.0 the software is your website, and O’Rielly stated “release early and release often”. A good example of this is Google. They constantly release improved versions of their websites (e.g. Gmail, Maps etc), sometimes without the customers not even knowing. The possible benefits of perpetual beta are: a faster time to hit the market, reduced risk, and increased responsiveness.

The best ways to benefit from perpetual beta are to release early and release often. A great example of this is Flikr which released hundreds of updates in an 18 month period, and eBay increments every two weeks on average. Another way is to release the update to a small portion of people so that they can test and give feedback on the update. However beware of to much updating or a major update. Facebook recently did a major update to the interface, which improved the service. However it left many people annoyed with having to learn a new interface.

The example I am going to use this week is Zamzar. Zamzar is a free online file conversion website, that is still in beta. The website has been around since 2006 when it only had two conversion types. Now Zamzar can convert between 75 different file types. Zamzar even has a blog on their website that blogs how the development of their website is going.

I can imagine that the future of Zamzar, is that will only be out of beta when they can convert between hundreds of file types. Zamzar has a bright future because they are developing a customer base, while the product is still being developed. There is no need to wait for the final product to reach your market, or to make a profit.

Pattern Five is a design that is greatly increased by the expanding capabilities of the web. Many websites are becoming available on devices other than the desktop. Examples include Google Maps and Google Maps Mobile, Facebook and Facebook Mobile Applications, or even WordPress that has Desktop Browsing, Andriod, iPhone, Blackberry and iPad applications. This software needs to be developed and changed depending on the type of device that access’s it. But this allows users to access your application with device and location independance. Just look at the success with the iPhone and its GPS

An example of installed software is MobileMe by Apple. “MobileMe keeps your mail, contacts, and calendar information in the “cloud” and uses push technology to keep everything in sync across your iPhone, iPad, Mac, PC, and the web automatically. So no matter where you go or what device you use, all your information is up to date — no docking required.” MobileMe’s whole purpose is keeping multiple devices synced, using the power of the web. I would se MobileMe if I used any Apple’s products. Most people are happy to pay the yearly subscription fees for the services provided.

The shortcoming’s of software above the level of a single device are that the software has to be developed for the different devices (e.g. WordPress’s multiple applications). But the power comes in for users to access the software/website from multiple devices at multiple locations.

For many years the world wide web was a series of static  web pages stored on a server somewhere in the world. As technologies grow,  internet speeds get faster and faster and download limits get higher and higher, more is capable from websites. Websites are starting to act and look like desktop applications. The website no longer has to be refreshed for every click you make (through AJAX).

Adobe has some examples  of rich user experiences (some call it rich internet applications, or RIA). The one i am going to look at is a Mini website, and mainly the function of building your own Mini. You can customize  just about everything you would like in a Mini, from colour to suspension to upholstery. With a static website, every change made to the Mini would result in the website having to be reloaded, but after the initial 5 second load of the application, you can change your Mini in every way without having to load anything new. This is acting very much like a desktop application because you initially load the program and do not have to load anything new. Also once the website has loaded, it never has to be refreshed. Separate sections have to be loaded but the whole website does not have to be reloaded.

The Mini website is a great example of rich user experiences. I even had fun while using their website, which can’t be said for a lot of other websites. It is a simple design that fits onto one screen and looks great.

The main idea behind innovation in assembly is providing open APIs for your website. This allows third party companies to use your website. This is a win-win situation for both companies. An example of this is Amazon. Amazon provides open APIs for third party companies wishing to sell products. The third party wins by gaining the selling power of Amazon, and Amazon gains more products to sell, therefore more sales, and therefore more revenue.

The example for this week is Multi Map. Multi Map is a Microsoft web page, that aggregates some of their web services together. It contains a web search engine, a navgiotion, business searching (like yellow pages). The main thing is the extra information you can view on the map. It has things like hotels, restuarants, petrol stations, train stations, parking and many more. Another one is Wikipedia information. Any location that has a page on Wikipedia, will be shown on Multi Map, with links to the Wikipedia page.

If you view the link on the side (Intel Inside Ad) to view the advertisement by Intel, you can see Intel trying to get their name out there. They were the driving force behind the computer (the processing power). And having an Intel chip inside your computer was what everyone wanted. But now, with the coming of web 2.0, data is becoming the thing that everyone is after. This data mainly comes from last weeks pattern: harnessing the collective intelligence.

The example for this week is Waze. Waze is a GPS navigation website, where they track the user’s currently using their GPS to create their data. This allows for real-time updates to the maps and traffic.

Whereas most (if not all) of the other navigation website’s get the data from a geospacial data company (e.g. NavTeq), which can cost alot of money. However Waze has creatively harnessed the collective intelligence to create their geospacial data. And the navigation websites would not work without this data.

Waze has a massive advantage over the other websites becuase they have to wait for new data from another company before a new road is on their website. However Waze can have a road on their website the day it is opened. And maybe oneday other websites like Google Maps will use Waze’s geospacial data.