A bit OT - Article: Decision making help

Status
Not open for further replies.

emp

New member
Jun 29, 2006
7,465
211
0
Summary
This article will explain a technique for coming to decisions in complex environments.
This works best in a spreadsheet program, but in a pinch, a notepad and a pencil will do.

Introduction
We often come to face problems where multiple variables come into play. Often deciding which action to take does not only involve these variables, but also our perceived value or priority for each item.

Deciding on a CMS, a blog system, a hosting company, a place to work out are just a few examples where this might come in handy.

To make quick headway in these decisions, you can use a technique called a "weighted matrix".
In this articledeciding on a gym will be used as an example application.

How to do create a weighted matrix
1.Categories
First off, you identify the influencing variables (categories, features, ...)
In case of a Gym these might be something along the lines of:
- distance to home
- distance to work
- price
- equipment sports
- equipment other (sauna, pool, etc..)

2. Priorities
You decide on a priority for each item. This reflects how important it is for you. I recommend a scale between 0 and 3 meaning 0 = not important, 3 very important.
A value of 0 is of special merit when you are working with categories someone else put together, so you can give this a value of "I don't care about this".

So these would look like:
- distance to home 2
- distance to work 3
- price 3
- equipment sports 3
- equipment other (sauna, pool, etc..) 1

3. Values
You give each gym a value depending on it's features.This is given by merit so, 0 = CRAP, 3 = great. Often, binary values work ok, for example software features might be just classified as 1 or 0 for "software has it" or "software does not provide this feature" respectively.

Some examples:
Gold's Gym (Far from home, close to work, great weights, no extras, medium price)
1 - distance to home
3 - distance to work
2 - price
3 - equipment sports
1 - equipment other (sauna, pool, etc..)

Platinum Gym (Close to home, a short commute from work, ok weights, ok extras, high price)
3 - distance to home
2 - distance to work
1 - price
2 - equipment sports
2 - equipment other (sauna, pool, etc..)

4. Scoring
To actually get the values, you now need to multiply each value with the priority assigned and sum it up for each gym.

Gold's Gym (Far from home, great weights, no extras, medium price)
2 - distance to home
9 - distance to work
6 - price
9 - equipment sports
1 - equipment other (sauna, pool, etc..)
Weighted Score 27

Platinum Gym (Close to home, middle from work, ok weights, ok extras, high price)
6 - distance to home
6 - distance to work
3 - price
6 - equipment sports
2 - equipment other (sauna, pool, etc..)
Weighted Score 23

So for the priorities I have given Gold's gym is the better choice, although not by far. (In this example, the maximum Value would be 45)

Summary
Although this technique looks a bit childish at fist, it shows its benefits when applied to large datasets.
When you use a spreadsheet, you can play around with the values, to help in thinking about situation/offer from various standpoints.

"What if I really LIKE having access to a pool?" - ramp up the extra equipment priority value
"Is the commute really going to suck that bad?" - tone down distance prio
"I guess the equipment was not really that bad." - up the equipment score

For your perusal I have attached an excel spreadsheet using this technique on Blog Systems, using a comparison chart found via wikipedia somewhere on the web.

http://www.captainklinge.de/bilder/weightedblogsystems.zip

If you have any questions or comments, let me know.

::emp::
 
  • Like
Reactions: JeremyL


Beastie!

This will definitely help since it takes me weeks to decide on something like a CMS.

Gracias emp
 
Very nice emp! Much better than the old fashioned "T" chart of pros and cons!

+rep /edit/ or not, apparently I've given out too much recently. Oh well, Good Job anyhow!!
 
Status
Not open for further replies.