I often read that's it is important to analyze our stats, referrers, track the clicks etc...
Ok but then once you've read them what do you do ?

If you know that on 100 people 80 clicked from a link inside the game, 20 clicked on the link outside the game, how does that help you ?

How to you convert the data into something profitable. I'd like to hear concrete examples if possible.

Thank you
pat.

I think the usefulness varies depending on the stat. For example, looking at the keywords that allowed people to find your site on search engines can help you figure out which keywords are working, which ones should get more exposure in your meta tags, which ones aren't getting used much and so maybe ought not to be the focus of your efforts, etc.

For the specific stat you mention -- how people are finding your site -- I think that helps identify where to put your marketing energy. In the example you give, where the majority of hits are coming from people who have already found a game and downloaded it, that might suggest to you that you focus more effort on bringing in new visitors -- e.g. by targeting advertising on sites where your games are not available for download. (You may decide that's not a good strategy specifically, but I think that's an example of how one might react to the statistics you describe).

Originally posted by patrox

If you know that on 100 people 80 clicked from a link inside the game, 20 clicked on the link outside the game, how does that help you ?


Thank you
pat.

Whatever stat it is, it's all about interpretation and knowing want you want to achieve.

In your example (assuming those 80 are going to your buy page) then you are doing a good job in-game. You may want to look at trying to improve the in-game text etc and try to figure out why 20% click the external link. It maybe that those 20 didnt feel comfortable/confident clicking an in-game link. The 20 external clicks may only be a part of the bigger picture here, because what the stat is not telling you is how many people did not click on that external link (maybe another 80 people did not feel comfortable).

Bear in mind that stats are a science but interpretation is an art.

Alfie

Originally posted by alfie
Whatever stat it is, it's all about interpretation and knowing want you want to achieve.

I just gave dummy stat in my example.
Obviously what we all want to achieve is sell more :).

To simplify I'm looking for real life examples of
"stat xyz - do this - get an increase"

Mark example is insteresting, discarding bad keywords and replacing them with new ones. ( of course there will always be keywords with smaller hits, but it can probably be optimized to a certain extent )

pat.

I think its more of a logic-finding game.

Like Jack Norton's thread about the randomness of the Indie scene, i think that your stats are the key to finding the logic of it.

Think more of questions like "why do costumers leave my site?" and find the answer in the page they left (stats show exit page), "what do costumers look for when they are in my site?" and make that part of the website bigger/invested.

Originally posted by patrox
How to you convert the data into something profitable. I'd like to hear concrete examples if possible.

This is more abstract than it is concrete:

Statistics are measurements of something. These measurements can be used to improve your business by prompting you to ask the right kind of questions.

For example, you might find that you have a conversion ratio of 1%. From that statistic, you can then ask yourself, "Self, how can I double my conversion rating to 2%?"

Statistics also help you to focus on where you can make the best improvement in your business. If your conversion rating is a stunning 50%, you might be better off asking: "Self, how can I double the number of downloads of my game?"

The answers to those questions are the concrete part. What statistic would you like to improve?

Cheers!
Carl

The stats help you decide where your problems are and what direction you should head. They should/could be telling you what to do.

A concrete example is... You release a new game. The game only sells 2 copies per month.

If that's all the date you have how do you increase sales? Personally I don't know. I could guess, but I might be wrong. Contrast that with...

You release a new game. The game only sells 2 copies per month. The demo files receives 2 downloads per month.

The answer in the 2nd case is find methods or try to increase the number of demo downloads.

But the stats might be...

You release a new game. The game only sells 2 copies per month. The demo receives 100,000 downloads per month.

In this 2nd case increasing your download numbers will still help but it's pretty clear that the main problem is that people don't like the program for some reason, so your effort would be best spent developing a new game or trying to fix the one that isn't selling. What you might actually do in this case is organize to get more stats. Like try to track how many people are able to actually run the game. Maybe there's a bug that prevents it from running. Again without stats you don't know how to fix it. You're just guessing.

Referrer stats reveal who's sending traffic so can contact for better / more partnership plus those who aren't producing and not worth the effort. Search stats show how people are finding you and potential keywords to work on optimizing. Pageview stats show what is popular. I'd die w/o my morning stats and cup of water (you we're thinking coffee huh).

A small tip on a related note: When dealing with external web sites try using alexa.com to ascertain how much traffic they actually get. It's free to use. While it's not 100% accurate, it will give you a rough gauge of their actual traffic

Every site in the world is ranked according to its position, eg download.com is 550th, tucows.com is 1,090th. It's fun tracking where your own site ranks, and you can also do simple competitor comparisons.