in

ASP.NET Weblogs

This Blog

Syndication

Windows XP Expert Zone Community WebLog

By the Windows XP experts and enthusiasts at Microsoft.

"Lets just use the 80/20 rule!"

By Bojana Ostojic, Lead Program Manager, Media Center Edition

I work on the Windows Media Center (MC) Team as a Program Manager.  Most of my work is focused on the large scale integrated user interfaces including information architecture, interactive models and navigation, ‘look&feel,’ and the overall user experience (UX). This here is my first blog contribution.

 

 

I’ve been stuck on the 80/20 rule lately.

 

I first learned about this rule long time ago, in school, as the Pareto Principle. One is likely to recognize the gist behind it if I used a couple of its most typically utilized statements:

  • “20% of the employees do 80% of the work” or
  • “20% of the customers generate 80% of the revenue.” 

The abstracted notion is that the minority input in any system can often be responsible for the majority of the output in the same system, or better yet, to quote what I found offered on one of the related sites, “vital few and trivial many.”

 

 

This powerful and handy 80/20 blurb is frequently used around our MC hallways, and I am starting to get curious about its many layers and incarnations and just how true to and consistent with the rule’s actual premise those are.

 

For example: Does the 80/20 rule have anything to do with time-driven schedules? If we didn’t have time-driven schedules, but rather feature-driven schedules, how would the rule apply then (would we go for the full 100)? Or, is the rule agnostic to the type of the product cycle approach and it simply focuses on a good design practice—one that states that no good product should try to be everything to everyone?

 

But, employing the 80/20 rule as a good design practice proves difficult in a system that, for example, covers a broad-ranging audience—say, novices to experts. Because they are so different, one inevitably ends up optimizing for one audience type over the other. This has certainly been my experience working on MC. Unlike many other products I've been a part of, MC ensures to optimize for the novice range (in part a design philosophy and in part due to the added difficulty of remote control operated 10’ UI). This is all good as it is very likely going to encourage simple and easy-to-use product design, but it also inherently limits the scope of offering, focusing on the basic features only. And for this reason, it is also likely going to cause dissatisfaction and annoyance on the experts end as they expect more power features.

 

So, when we end up making an in-out call for a feature and we do so with the 80/20 tagline, are we practicing good design practice or are we applying prioritization due to a time-sensitive schedule? And, in either case, are we in accordance with the actual meaning of our tagline?

 

-Bojana

This posting is provided "AS IS" with no warranties, and confers no rights.

Comments

 

Dean Harding said:

I think the problem is that no matter how you structure your product cycle (whether you say "it must be finished by x/y/z date," or whether you say "it must have x, y, z features") you're still bound by a time limit.

You want to make your product useful to the greatest number of people possible, but over time the number of people who could find your product useful will drop off. For example, if you set out in the late 1800s to build a mimeograph machine (http://www.sil.org/lingualinks/literacy/referencematerials/glossaryofliteracyterms/WhatIsAMimeographMachine.htm) then you spent 100 years making it the most perfect and feature-rich machine possible, your investment would have been all for nought, because someone else would have invented the photo copier already.

That's a pretty stupid example, but it actually happens quite a lot in software. You spend so much time polishing a product that by the time you finally release, someone else has come along and stolen all your potential customers before you ever get to market.

So I guess my point is that whatever rule you use to decide what goes in and what stays out of a product, you need to draw the line *somewhere* and the 80/20 rule is probably not a bad way of doing that.
February 1, 2005 9:30 PM
 

SBC said:

February 1, 2005 10:42 PM
 

Mark L Pearson said:

I have watched the 80/20 rule at work for many years. http://www.claytonchristensen.com/ has spoken about disruptive technology and also when you follow the 20 of the 80/20 it is certain death of the project.

As I look over a 30 year business career I believe he is correct. I also belive that Blogging will be the next disruptive technology.

Media Center is a good direction. The only complaint I have is a good hardware list for best results!!
February 2, 2005 10:06 AM
 

Joel Barciauskas said:

Have you read Joel on Software's (different Joel :) ) article on the subject? (http://www.joelonsoftware.com/articles/fog0000000020.html) His basic premise is that, yes the 80/20 rule is applicable, but it tends to be a DIFFERENT 20% for every user.

Another good article on this subject is the Wired Magazine article entitled, "The Long Tail", which deals with the huge range of niche products that are now available via places like Amazon.com, that would never have been feasible to include in a traditional brick-and-morter store. It's a similar idea: B&M can get 80% of the market stocking just 20% (or less) of the products, but Amazon.com has tapped into a whole new world because of its ability to offer such a wide variety.
February 2, 2005 11:12 AM
 

Troy Allen said:

Bo...

Is this the Bo of UW Tech Comm fame? If so, this is Troy! Please drop me a line at tallensr at gmail.com so we can re-connect!

If not, sorry for the "spam".

Regards,

Troy

June 25, 2007 9:04 AM
 

S P Boyd said:

Albeit a day late, happy birthday.

August 16, 2007 6:27 PM

Leave a Comment

(required)  
(optional)
(required)  
Add