Phew… It’s back again.
Really seems though as if the run on waves offers is huge right now.
www.waves.com just went down – right when I was about to check out.
I guess I broke it.
I wonder if it’s possible to assign scores to tracks and give results based on those scores, which are determined by click through to search ratios.
For example, let’s say a buyer searches “Happy”. The search returns 10 songs called “Happy”. For the sake of the example, let’s say all songs have the same title, have happy in the tags and the word happy at least once in the description.
Who should be displayed first? I think it would be cool if the default return was based upon the number of times a word was searched vs how many times the song was purchased when the word was searched.
So if 4 buyers searched “Happy” and ended up buying a track (we’ll call it track A), and 3 buyers searched happy and bought a different track ( track b ), it would display track A first.
If the search to buy ratio became greater for track B or any other track, it would move above track A. So the search relevance is always changing, always dynamic.I don’t even know if this is possible, but it would sure create a very relevant search and very happy buyers I think. Of course they could still sort by all the other things like sales and date too.
Interesting thoughts. I’d really like to know how ‘search term relevancy’ is determined in the new search engine anyway… I read about how important relevancy is supposed to be in the new system – but I couldn’t find any real information about how this is accomplished. Or simply put: How would the new serach engine know which tracks actually are relevant to certain search terms?
I think a search-sell ratio is one way to achieve this. But this would of course again just favour the already selling tracks.
So – if you don’t want to put it all on sales – why not also track how long a potential buyer listens to a preview track?
If a potential buyer skips to the next track after listening to one track just for let’s say about 10 seconds (or any number you might think works here) it could be seen as one indicator (of many) that the track was not relevant to his search.
If he listens to the track for longer this could then be an indicator that he found a track relevant to his search – even though he later on might buy another track because it fits his project even better for whatever reason.
So again my question to the staff is: If relevancy is such an important factor in the new search engine – what kind of data do you track and use to determine this?
Thanks for all this awesome feedback!
For those of you who might be interested – here’s a bonus video showing the sound design of our film, which was done entirely in post production:http://www.youtube.com/watch?v=DVoNcf4j3Uc&noredirect=1
Have a great day!
I wonder what happened to NoizMan’s profile and portfolio…? Seems to be gone completely…
Just a system hiccup?
from my experience ‘nWallet’ is one of the best Android apps for this:https://play.google.com/store/apps/details?id=ba.dysko.envato&hl=de
And it’s free!