I have this massive list of "product features" for an app that I am working on, and I am having trouble just making sense of which to look into first.
I thought it would be cool to have an app that would present all the items successively in "pairwise comparison" format, to enable to determine an "implied priority"
|A 1000Minds presentation of two contenders in a pairwise comparison format;|
The idea being that initially you would start with a rough list and then successively present pairs of these items (e.g. "add X widget feature" vs "improve Y process") to the stakeholders.
The ranking would be adjusted with each iteration over time, to a stable natural order of priorities, and also that the assigned priorities can be dynamically reconfigured by re-presenting the options again, as their needs change or further information becomes available.
I think the tool would have to work with some other todo, feature, task or activity tracking system, such as http://todotxt.com, MS Project, basecamp, or remember the milk etc....
The only example I could find of a live example of pairwise comparison (in a 5 minute googling) was one that allows the comparison of 2 "babes", but you can select either girls or guys - The photo sharing playground for hot babes
Here is an example of such a sort proceeding, with the rankings becoming ordered correctly as time passes with each comparison;
image credit: http://sortvis.org/algorithms/bubblesort.html
I would imagine it something like a double elimination Bracket (tournament), and by initially assigning a random priority to each feature, it would show convergence with similar dynamics to bubble sorting;
Update: It seems that PAPRIKA-potentially all pairwise rankings of all possible alternatives is a implementation of this idea; how 1000Minds works (under the hood)
Also, I see that spigit (Innovation Management Solutions & Idea Management Software) has a product called "icon" which has an implementation for Q&A answer ranking;
1) obviously the intention is to generate a "priority indicator", not to enforce the order of activity planning. (some things technically have to come before others, even if the stakeholders deem them less important)
2) presumably the final priority would be the weighted average of the various contributors
3) an implementation of a gui in excel - dailydoseofexcel.com
Daily Dose of Excel
4) I also asked the question on Quora
5) LL Thurstone paper on comparison of psychological value http://www.brocku.ca/MeadProject/Thurstone/Thurstone_1929a.html