Increasing Download Rate (Proficiency)

[Coyote]

Russell Carroll talked about a "Proficiency" rate in his article in GamaSutra a few months back - the % of visitors to a webpage that results in downloading a game (If I understand the statistic right).

I'm only set up to measure proficiency on three games right now, and the results are pretty varied.

So far it looks like the rate isn't that exciting... 1 in 8 for Outpost Kaloki (http://www.rampantgames.com/outpost.html), and 2 in 3 (and falling) for Aveyond (http://www.rampantgames.com/aveyond.html). Aveyond may have a higher rate because I've pushed it more in my blog, linking interested readers to the page.

I'm fixating somewhat on the screenshots - I'm not extraordinarily impressed with the quality or positioning of the Outpost Kaloki screenshot thumbnails, for example. But I want to decide on a clear strategy before I start re-arranging everything.

Anyone have suggestions that might improve the proficiency rate?

[Nexic]

Aveyond is at the top of the page, Outpost Kaloki is much further down below the 1024x768 fold - no wonder Aveyond gets more downloads.

[amaranth]

The screenshot is very important. When I changed mine (remember my old one?) it made a difference. Out of curiosity, can you tell us more about the other games click-through rate? For example, on screenshots alone, I would click on: Mystic Blades, Virtual Villagers, Cute Knight, Void War, and Styrateg.

From an artists perspective, I would say that the Outpost Kaloki cover has too many focal points. One of them should be very distinct. I think that the Mystic Blades image is a good example of this.

Here are more thoughts:
When I see the Outpost Kaloki cover, it doesn't inspire me to learn more. However, because of this post, I did click to learn more and was blown away by the beautiful game graphics.

Good luck! :D

[Coyote]

Nexic - I'm talking about hits on the Outpost Kaloki (http://www.rampantgames.com/outpost.html) page versus hits on the Aveyond (http://www.rampantgames.com/aveyond.html) page. I'm fully expecting fewer visitors to the Kaloki because it is below the fold, and an older game.

What I'm curious about is if someone goes to the Aveyond or Outpost page, and after looking at that page opts NOT to hit the "Free Demo" button. I'd like to decrease that. What is making people decide not to try the download after looking at the game's page?

Thanks!

[amaranth]

Ahh! Okay, I think I got confused too. So this is only when someone visits the game page...

In that case, maybe the user was expecting to see a Sierra adventure-style game? That's what came to my mind. For me, I love games with old-school 2D graphics. Maybe it's the 3D that scared them away?

[Coyote]

The screenshot is very important. When I changed mine (remember my old one?) it made a difference.
You mean the old title screen - the screenshot as opposed to the artwork now used?

Out of curiosity, can you tell us more about the other games click-through rate?
I can't without adding an extra bit of re-direction. Kaloki and Aveyond are hosted directly on my servers, so I can look at the number of downloads directly. The Void War download is pointed to by many different places all over the Web, so I'd have to create a customized report out of the logs to see where the downloads are really coming from (just comparing the page hits to downloads makes it look like it has an AWESOME proficiency rate. But that would be erroneous).

I can add an intermediate download page (or automated redirect) to get a better handle on all of those stats - and I have been considering it - but I worry about the heisenburg principle playing into it here. The extra step used to measure the number of clicks on the "Free Demo" button could negatively influence the number of real downloads.... :)

The biggest change I'm considering across all of the games is moving the screenshot thumbnails up above the copy.

And I do believe you are right about the quality of the thumbnails for Kaloki. I have learned not to just shrink the whole screenshot since then.

Any other suggestions?

[soniCron]

If the game is visually impressive, try a gameplay video with Flash Video (http://sonicron.solaristudios.com/2006/10/08/jeweltopia-gameplay-video/). It really makes the game come alive and beats the pants out of screenshots. (Unless, of course, the screenshots are more impressive than the video! ;))

Speaking of the Heisenberg Principle, regardless of how many downloads you count, it doesn't tell you much if you don't know how many complete. Try feeding the downloads with a script and take note of how many are canceled and how many are successful. You can do this with about 40 lines in PHP. It has been entirely valuable for me - I strongly recommend you give it a shot.

[cyrus_zuo]

Can't help but love this discussion... ;)

I've done a bit of playing with screenshots and videos on Reflexive pages as well to see if they directly impacted completed downloads (completed vs non-completed downloads is another issue as noted).

With changing screenshots I've seen an increase of 10% on the game I targeted (Ricochet Lost Worlds), though I made the changes with a specific thought in mind..."people like thumbnails that are recognizable objects." I then put a castle and a train into the screenshots. With SO many changes happening to Reflexive, it is difficult to determine the result of any one change...unfortunately, but I do believe a lot can be done with screenshots and images, especially since the audience is so visually attentive.

On videos we added a video to the Mosaic page this summer. Since adding the video, the proficiency decreased. However, conversion has increased. This may be to the video helping to pre-select people, but again it is hard to determine a single reason for the change. Mosaic is a fairly new game and the numbers for it may have still been riding the wave of newness when the video was added.

Anyway...lots of interesting things to consider...for those who want to read the article from a few months ago (notably rather long...but with bullets!) it can be found here:
http://www.gamasutra.com/features/20060711/carroll_02.shtml

[Coyote]

Hey, heyas Russell. Been missin' you at the Utah Indie nights - been wanting to pick your brain about this kinda thing :)

Thanks for the suggestions!