1/26/2014

The distribution graph of website landing page

Every one know there is a famous "Long Tail Theory", that is, if your website is good enough, your site traffic won't be narrow in few pages. But is it true? or how will it looks like? I start a project to "show" the guess and prove it.

This project is based on "Google Analytics API" to read the raw data, and I use the "Landing Page" as X-axis, and visits as "Y-axis", the X-axis is the percentage of the rows of landing page, the Y-axis is the percent of accumulated visits percentage, so it start as (0,0) and reach the end (100,100).


I add the color and size of the line to show the difference, the latest line (latest month) is red line, the nearest line is darker, and the earliest one is lighter, so you can see the "involving" of these lines.

And I try to define a index called "Distribution Score", that is the area of line above, if the visits of every page are the same, it will be like a straight line (red line) following:
If your website are popular in few pages, it will show like this:
You can see the line will reach over nearly 80% visits at first 5% rows, this is far beyond 80/20 rules, that is the first 20% items, accumulation more than 80% amount.

By now, this will be a good way to evaluate the effort of SEO, because if we look through the "Landing Page Distribution Score" and "Keyword Distribution Score", the percent of reach from search engine will be higher. But we can talk about this next time, the following is gallery of these chart which have done.






By now, you can notice that the "distribution score" of some site is getting better, but some are getting worse, and other are keep status quo. Using these images, you can easily identify your work of SEO, if you want to try, you can login in http://seokpi.analysis.tw/ to check it out, both of your landing chart and keyword chart. Have Fun~~

Are you jaming into Global Game Jam 2014 now?

There are 4 sites in Taiwan for Global Game Jam 2014, including Microsoft sponsored site.

This site are about 40 jammers divided into 7 teams, most of jammers  are under 30, I might be the oldest one, sigh~~~

But it's fun to participate this event even I have no ability to create game, all I can do is watching the group finished the game, interpreted the languages, and take photos.

It's 40 mins left before upload, we (not including me) develop a flash game, cause most of crew member familiar to this framework, the game call "Rainbow Braver". The game play snap shot is above:
We create a facebook secret group to co-work, we post near one hundred articles in 48 hours by 5 jammers, all of them is series discuss like this photo, this is a prototyping test using stickers.

In fact, we discuss about more than 10 game plays, although this game play is little far away game concept topic, but it is most easy to developed.

The game description in game jam is http://globalgamejam.org/2014/games/%E5%BD%A9%E8%99%B9%E5%8B%87%E8%80%85-rainbow-braver , you can take a look.


The Game strategy and concept is above:

1. There are 6 different color light of projectors, when you press the button, the cube of this color wiill vanished.
2. When you press the button, you have a limited time to move all the cube.
3. All you have to do is make three cubes with the same color in a line when time up, the cubes will vanished and scoring.
4. The empty slot will filled next turn.
5. Each of 6 different color projectors can be using once, so you have 6 turns to scoring. It's time to make challenge how many scores you can get.

The highest score is how you can using the color to filter what you want and strategy.

The game concept is using your mind to make thing what you want, that is "We don't see things as they are, we see them as we are". If we wearing colored glass, things like the same color will vanished in our sight, so that we have more chance to make things possible we want.

As time I post, the highest score is 4000, how about you?

1/16/2014

Annual Analysis -- Working with Google Analytics API

It has been more than four years to blog here, that is to say, It has been nearly five years since last time I publish new system outside Taiwan. That's too bad, it can all blame on "Friend Connect" ruined me... sigh~~ But I know the I should blame on myself.

In these years, I use API more than ever, including Facebook API and Google API, but most of the systems are only available in Taiwan, in fact, these suit for other users, but I didn't make them multi-language, so I didn't publish here.

"Annual Analysis" is another kind of approach, cause this is a small but useful system, and it "must" be done in January. I don't think some other will use it after February. This is a system helping you review your blog history in 2013.

This system started developing on 7, Jan. It takes nearly 9 days to work out. In these 9 days, we held up two small "hackahop" for three developer in co-working space (in board game cafe actually), cause most of us have daylight regular job himself.

OK, that is preface, not important, let's begin with this system.

Annual Analysis is a tool of Google Analysis, cause we know it is hard to get "real information we want" with basic function of Google Analytics, so I have develop a system call "網智" ( http://web.mas.ter.tw/ ) to helping people using Google Analytics.

If you want to analysis a blog system what are most popular new articles in last year, at first, you must know what are "new articles", that beyond most of the people using Google Analytics, we must group up the articles before last year, the rest are last year, it sounds easy but not easy, that is what "Annual Analysis" can do for you easily.

"Annual Analysis" recording most popular articles each year, so it can easy identify which are new articles for latest year, and "searching keyword" can be done by these.

But most of people want to know how popular my blog articles it is in SNS, including Facebook, Twitter, Google+, Plurk and other, so when the system identify most reference from SNS, using API can fetching data and connecting with articles, but it only work with Facebook now, if you want to know more, let us know freely.

All you need to do is "Entering" this website, login and authorize, wait for less than 30 seconds, you can see the the analysis report, but if you have more than one property of Google Analytics, it will take half minute each property, so please wait in patient, or you can read the report that already generated.

I often forgot to tell the url in my articles, you can check it out in http://annual.analysis.tw/ , have fun, and I appreciate that telling me your needs and problems. Thank you.

Above is screen shot of this blog, but you can check the demo of my chinese blog here.