Request: Please share your prioritization methods here

[Paul Keith]

I don't know how one can, in honesty, tout a prioritization scheme that says starting a blog is more important than making friends.

Source: http://www.reddit.co....com/comments/1sxv8/

I think it's a valid criticism and since I don't know how to use SpreadSheets, I've never bothered with complicated or over-simplified grid comparing techniques.

I think this holds true for Project Management too.

For the most part, the prioritization methods I've found are either too focus-centric on tasks to the point that they are more self-reflective projections on what you know or intend on doing first anyway or they are as rigid as GTD contexts and it ends up serving those who can already achieve their tasks rather than help those who have problems with prioritizing...or simply put, it's an optimization hack rather than a "help me!" technique and that's no good for those actually searching for a how to.

[mouser]

That's actually a nice link -- while it doesn't always transfer, it's often helpful to see these descriptions of how one might apply military (or another domain) organization principles to other domains.

From that page:

CARVER is an acronym for a military method of target selection.  CARVER stands for Criticality, Accessibility, Return (or Recuperability), Vulnerability, Effect, and Recognizability.  I’ll explain what these are in a moment.

For every potential target, we assign a value of 1 (lowest) to 5 (highest) for each CARVER factor, thereby creating a CARVER matrix.  Then by summing the six CARVER values, we can calculate a total score for each target, and those scores represent the targets’ relative prioritization.  The higher the CARVER score, the more “important” a target becomes.

In truth, i've not made any headway on a system of prioritizing my list of tasks.  The only breakthrough i've had is in moving to a system that let's me easily choose a few items to work on at any given time, and easily lay those out in front of me, using an index card for each one.  For me this is a huge improvement over keeping todo lists -- at least it lets me select a couple high priority items and easily set them in front of me for focus.

[raybeere]

In fairness, if you read the entire article, the system didn't say that starting a blog was more important than making friends. The point was that starting a blog is relatively quick and easy - and might even help you make new friends - whereas "make new friends" is an enormous, fuzzy task which won't get you anywhere until you develop a more specific strategy for doing so.

That said, I have a system which works well - for me, at least. The underlying idea is that you can often reliably decide which is the more important of any pair of tasks. So you pair up every possible pairing of the tasks you need to prioritise, list the "winner" of each pairing, then calculate the relative priorities based on the total score of each item.

I built a very clunky WordPerfect macro, one which is only good for (if I recall correctly) a list of fifty or fewer tasks - because I couldn't be bothered expanding the routine any further. First, it asks you to enter each task, and it stores each of these and counts the total. Then, it pairs up each possible set of tasks, and asks you which of the two is most important. It keeps track of these votes, and generates a final list with the item having the most total votes at the top, and the item with the fewest votes at the bottom - also indicating the percentage of votes each item received.

Yes, I've still found that there are some instances where it is hard to make up my mind. But in those cases, it is close anyway - and over a list of five or more items, it seems to average out the uncertainty and arrive at a fairly useful result. The one flaw, which should be obvious, is that it relies on the user's judgment. If I stubbornly insist on regarding some huge, unmanageable project as highly important, this system won't save me. But, of course, any time you assign any values in any system, your own prejudices are the most unvarying weakness.

[Paul Keith]

whereas "make new friends" is an enormous, fuzzy task which won't get you anywhere until you develop a more specific strategy for doing so.

I find talking to random strangers work although I've heard you can get some too by joining a forum.  

of course, any time you assign any values in any system, your own prejudices are the most unvarying weakness.

True. That's why I opted to make this a survey-ish topic instead of sharing my own prioritization categories. I felt what's more at stake is to have everything out first to contrast and compare rather than stress-testing any prioritizer.

That said, I would love a lightweight cross-platform NANY that allows for auto-sorting for people like me who doesn't know how to use macros.

[IainB]

The most powerful tool that I ever came across for getting things done was the "ABC" method. This was long before the term "GTD" was coined, and was taught to me by a project manager (I was working in a Business Analyst role) in about 1977.
I have used the ABC method over many years and to good effect, and modified/augmented it only slightly. I have coached people in its use, and they have then been able to quite literally transform and take control of their busy lives.

The ABC method of prioritisation for tasks:

• A = Urgent AND Important
• B = Important BUT NOT Urgent
• C = Neither Urgent NOR Important

(There is arguably a logical 4th category: Urgent BUT NOT Important. However, this makes little sense as it is ambiguous, so it gets left out.)

The ABC method of prioritisation for user requirements in a system development:
A = Mandatory
B = Highly desirable
C = Nice-to-have

The ABC method for tasks focusses the mind wonderfully, and helps you to PLAN and to make decisions:

• You only need to worry about dates for the "A's" - as they are urgent. Some "A's" might need to be done before others - dependencies.
• You therefore address the "A's" first.
• The "B's" can be picked up and worked on as and when you have some slack time whilst addressing the "A's".
• The "C's" can be forgotten - because they are largely irrelevant (by definition) in the overall scheme of things.
• Events outside of your control may cause the priorities to be up- or down-graded.

The ABC method for requirements also focusses the mind wonderfully, and helps you to PLAN and to make decisions:

• You MUST address the "A's" - all of them, and you need to identify any potential/actual interdependencies.
• You negotiate with the users as to which "B's" are going to be included.
• The "B's" usually take 2nd place to the "A's" in the queue for resource allocation.
• The "C's" can be forgotten - because they are largely irrelevant (by definition) in the overall scheme of things.
• Events outside of your control may cause the priorities to be up- or down-graded.

How I applied the method:
* Media: I tried using index cards, but it was too fiddly and so I moved to having 3 clear plastic folders - one for each of the 3 priorities.
In each folder I could have n sheets of paper forms. Each form was photocopied from a template. Each sheet had ruled lines down the page, with columns reading from left to right, as follows. You could only write on the form in pencil.
   *   Priority: value could be A or B or C, but you only needed to write the priority in if/when it changed from that of the sheet.
   *   Details/References: value is free text.
   *   Action: value is typically one or more of these:
                  Status flag: ToDo; WIP; Done; ARR (Awaiting Response/Reply);
                  Who: the initials of the person who was assigned to carry out this task;
                  Activity required: Do it; Call; Email; Meeting;
                  Date: due (usually for "A's" only); done.
                (You need to use pencil and eraser to update these.)

* Priority changes: Rarely - and usually as a result of a mistake somewhere - an "A" entry would be downgraded to "B", and similarly a "B" would be downgraded to a "C".
The usual change was a "B" entry being upgraded to an "A", as and when it became time-critical.
To effect the change, you transcribe that entry onto the other appropriate form and rub it off that line where it had been on the original form (where it now no longer belonged).

ABC computer-based method: In 1989/1990, I started using Lotus Agenda, a PIM which was ideal for automating the recording and dynamic updating of the ToDo list and details of any associated data. Lotus Agenda is obsolete now, and I have not found any software that can perform as well as this since, so it is back to the paper-based method. However, I was surprised to see that something similar was starting to be achieved with a Firefox/Chrome add-on called "GTD for Gmail" (still in ß), until they changed ("improved") it a week ago and apparently ruined the emerging potential. I think the developers probably failed to see/understand the significance or potential of what it was that they were building - you sometimes get a lot of that in IT.

Hope this helps or is of use.

[Paul Keith]

Not my own, just dropping it here:



http://en.wikipedia.org/wiki/Kaizen

P.S. Thanks IanB. Yeah, it does. I actually follow/plan to follow the ABC method in a certain subject manner but I haven't experimented on it yet.

[IainB]

@Paul Keith: Ah yes, I first met this at a seminar by Deming, as the "PDSA cycle" (Plan, Do, Study, Act), which provides a rational method for making changes (improvements) to a process, when attempting to effect an improvement in quality of output of the process. I read somewhere that Deming said he borrowed it from the PDCA (Plan, Do, Check, Act) of his teacher, Shewart.


Interestingly, Deming apparently disassociated himself from the over-hyped TQM (Total Quality Management) "movement" (maybe "gravy train" would be a more suitable term) - which included Kaizen - that was largely built upon his work by third parties and who generally seemed to have some difficulty in comprehending his 14-point philosophy.

[tomos]

Thanks for the posts 

ABC computer-based method: In 1989/1990, I started using Lotus Agenda, a PIM which was ideal for automating the recording and dynamic updating of the ToDo list and details of any associated data. Lotus Agenda is obsolete now, and I have not found any software that can perform as well as this since [...]

-IainB (June 12, 2010, 02:08 PM)

~
could you go into more detail about how this worked with Lotus Notes ?

I know you discussed Lotus Notes a little in an Info Select thread where you also mentioned a previous thread (which I couldnt find) -

I did post an "IDEA" for new Agenda-like software on the DC forum a while back, but no takers. I suspect that more modern generations would not have a clue what it is, and therefore would not want to dig down and develop it, and older one-time assembler programmers like me - who knew Agenda well, but also know our limitations - haven't started to try and develop it, even though we might like to.
I am playing with InfoQube (was SQLNotes) at present, to see what that offers.

-IainB (September 23, 2009, 11:33 PM)

~
How did you get on with IQ ? - I could imagine it being very adaptable to the ABC approach.

[Armando]

To prioritize, I use a mixture of :

1- the Covey ABCD system,

The ABCD system is basically the same as the ABC code described above, but with the "D" being Not important and not urgent... Thoses tasks and projects that can basically tossed away or archived for future reference.

To implement that system properly it's "compulsory" to understand what urgent and important truly mean...


2- together with Franklin number priorities (1,2,3...).

These were first used in the paper based Franklin planner created by Hyrum W. Smith.


3. Then, I use a special formula to calculate the urgency of tasks (processed in IQ) depending on their :

• duedates
• Planned Start and End date
• % Done
• time needed to complete them
• numeric priority (1-9)

Of course the formula takes into account whether the task-project is late or not, etc.


4. I use a followup field so that I can defer tasks without touching their real start or due date.

I use other fields (like context, etc.) but these have nothing to do with priorities.



In general, I use the rather standard way of dividing my life activities in a few categories : values, needs, goals, projects, dreams, tasks. I used to have 2 more categories : whishes and dreams, but I don't use them as much now... Maybe I should as my life is not so exciting these days!

Activities in these can be grouped in different (I forgot the English name...) "Life domains" (family, job, health, personal care, finance, etc.) to help see where there are some lacks, etc.

I also try to use the :

- 80/20 principle.

- The Smart Method (to set my goals and projects):

S     Specific & Significant
M   Measurable, motivational, methodical & meaningful
A   Action-oriented & achievable
R   Realistic & relevant
T   Time-bound & tangible


I use other things (like the GTD principles : inbox + todo lists...) but I'm running out of time now...


In any case, the thing to remember is that any of these methods, if not incorporated into a very concrete and integrated system, won't work. Whatever makes sense to someone must be integrated into concrete steps to follow (in lists, computer software...) each day/week. I'm using IQ for that, but I'm not completely satisfied yet -- it's hard to build you own system. That said, I know there's no perfect way and, in the end, what matters most is to do something that makes you feel satisfied or proud. That's why I use 3 simple criterias (my own) to select every project I decide to work on :

                  1- Richness and quality of experience (bringing good memories, happiness, pride, etc.)
                  2- Advantages (returns, positive impact on my future) (convergence)
(En accord avec certaines valeurs -- voir ci-dessous)
                  3- Relaxation (divergence from day to day routine, just... plain replenishment)


They have to fulfill at least one of these 3 characteristic to an acceptable degree (subjective) otherwise I (try to) dump them.

=====

I had a look at Paul Keith's link in the first post and found it very interesting. THere are some similarities with the 3 points I just mentionned, so I'll try to implement some it too as it makes perfect sense in general (maybe not for artistic creation though -- but...). Sometimes it's hard to decide which project to tackle first, and this could help -- setting that up in IQ will be a breeze...


[Edit : I  also liked what I read about the PDCA system. I'm going to integrate that in my Weekly Review of projects. Handy. I also forgot to comment on Mouser's remark about being able to isolate a few tasks to focus on during the day or week -- this is vital for me and I absolutely need to do that so that I don't get distracted by a long (too long) list of possible stuff to do. Any system must allow this.]

[IainB]

@tomos:

...could you go into more detail about how this worked with Lotus Notes ?
...
How did you get on with IQ ? - I could imagine it being very adaptable to the ABC approach.

-tomos (June 13, 2010, 08:47 AM)

It was Lotus Agenda that I used, not Lotus Notes. The two things were completely unrelated in the Lotus stable of products.

It would take too long to describe how I had ABC working with Lotus Agenda - even if if I could remember it all after so many years. Suffice it to say that Agenda was a very powerful text database tool, complex to use and necessitating that you were able to apply some good logic as a user. I became a power user and eventually had used most of the documented features of Agenda and some undocumented features that I discovered also.

Sorry, but I'm not much of a good user of IQ: My efforts with IQ tailed off (I rather lost interest) when I discovered that it was a good general purpose d/base PIM tool, but that it lacked some of the Lotus Agenda-like features that I was looking for. In theory, since it uses SQL, then, as a user, you might be able to frig the IQ system and emulate those features, but I did not have the time to invest in trying that approach out.

That was what surprised me about the Firefox add-on "GTD for Gmail" - it had some features that made me think maybe the developers had understood quite a lot about ABC. For example, they had some horizontal buttons you could set up for "Status", and which were dynamically mutually exclusive. Thus, when you clicked (say) "ToDo" for a task/item, and later clicked WIP (meaning you had started working on the thing), then "ToDo" was automatically "unpressed". However, for reasons that escape me, the developers seem to have just ruined the possibility for doing that with their latest changes because these buttons are now not mutually exclusive and have been buried in drop-down menus to boot. Very tiresome.

One of the most useful things that Agenda could do was to dynamically auto-set a logical attribute called "category" for a task/item, depending on a rule - for example, whether a certain character string was present in the item data. In the Gmail context, this would be like Gmail dynamically setting a label for an email discussion, if (say) the word "frog" was found in that email discussion. I think this sort of capability might be built into mouser's Clipboard Help & Spell - which employs virtual folders and SQL filters - but I haven't had time to play with that to find out for sure.

Hopes this makes sense. I tend to make mistakes when tired, and I am tired now.

[tomos]

Thanks for the reply Iain,
apologies about getting the 'Agenda' name wrong  
And I see now that focusing on the software was probably veering off-topic - I may bring it up in another thread. I found your proposal btw IDEA: Create a modern version of Lotus Agenda


Armando - nice to get an idea of your system, thanks. Some good ground rules there too... will have to have another look at it tomorrow (tired here too - Germany just after winning first World Cup game 4:0 - Yeaaay . . . now I'm definitely off-topic :p )

[Paul Keith]

In any case, the thing to remember is that any of these methods, if not incorporated into a very concrete and integrated system, won't work. Whatever makes sense to someone must be integrated into concrete steps to follow (in lists, computer software...) each day/week.

-Armando (June 13, 2010, 12:47 PM)

Could you provide a visual or specific task vs. task example?

The way you combined all systems you encountered got me curious as to how long it takes for you to put it all together as well as how much your prioritization improved before and after the whole process took place.

I know you mentioned the systems you used and I apologize if I'm being vague. It's not that I didn't understand your post as much as I didn't catch the part where the concreteness and integration locked in.

One of the most useful things that Agenda could do was to dynamically auto-set a logical attribute called "category" for a task/item, depending on a rule - for example, whether a certain character string was present in the item data. In the Gmail context, this would be like Gmail dynamically setting a label for an email discussion, if (say) the word "frog" was found in that email discussion. I think this sort of capability might be built into mouser's Clipboard Help & Spell - which employs virtual folders and SQL filters - but I haven't had time to play with that to find out for sure.

Hopes this makes sense. I tend to make mistakes when tired, and I am tired now.

-IainB (June 13, 2010, 02:33 PM)

Could you provide some examples as to what specific labels and character string you had that gave the feature a much needed place in your priority system?

Descriptions like these often make sense until I try imagining what specific string I would use.

[Paul Keith]

Content, Relationship, Information, Process

Source: http://en.wikipedia....iki/CRIP_Methodology

[Armando]

Could you provide a visual or specific task vs. task example?

The way you combined all systems you encountered got me curious as to how long it takes for you to put it all together as well as how much your prioritization improved before and after the whole process took place.

I know you mentioned the systems you used and I apologize if I'm being vague. It's not that I didn't understand your post as much as I didn't catch the part where the concreteness and integration locked in.

-Paul Keith (June 13, 2010, 06:57 PM)

Hi Keith,

Like I said, I was mainly influenced by Covey (7 habits) and Allen (GTD), but also Effexis software and a lot of other stuff that I just integrate in my system as reminders, processes in task list, and other various lists (Weekly Review, etc.).

In IQ I was able to set my priority system exactly how I wanted it. It's pretty straight forward, but with a few twists and complexities most people wouldn't care about. Note that I was able to set a similar system in Outlook, minus the urgency calculations and tasks hierarchies, and a few other twists.

It took me a while to set it up exactly how I wanted it, and It's hard for me to say how much my life has improved since then... I personally feel more "in control" (in a good way), and I always know what I need to do.


===========


All the "prioritizing" parameters  I exposed previously appear as fields/columns in various grids (tasks grid, project, etc.) and so I just need to use specific filters to see the items I need to see depending on my needs.

Tasks :

Currently, when I enter a new task, I just go from column to column in the grid and ask myself a few questions for each column/parameter I encounter. For ABCD priorities, 1-9 numeric priorities : questions are of course linked to my values, projects and goals (i.e. : I'll favor the tasks linked to my most important values, projects and goals : these are important, they are As if they're "time bound" or Bs if they are important, but not so "time bound"). I'll also ask myself the questions I outlined in my last post -- but it depends on the task, and it's mostly for projects.

The due date (+ start and end date if necessary), time necessary to complete the task etc., don't need any special existential questioning, they're usually straight forward. They do affect the urgency though. A task taking 10 h to complete, due in 2 days, but that I haven't started, and which priority is an A 9, is automatically ranked very high on my priority list...
 


Projects :

This follow pretty much the same process. But I do take more time to evaluate projects since they take longer to complete and mean a bigger energy investment. I ask myself a few questions and I then decide if it's an A or B, or even a C or a D, how much time I want to invest in them depending on how much time I have left, etc. and then I give it an extra push towards the top or the bottom with numeric priorities (1-9), etc..


For both tasks and projets, all the relevant data (due date, priority, etc.) I described in my previous post is used  in a special urgency formula that ties the parameters together and gives me a result (numeric) and colors items depending on the urgency. Red is what needs to be done ASAP (now)...

Various filters allow me to see what's coming if I need to.

The weekly review is important (it's a group of tasks that's part of my task list, but I actually usually do it once every 2 weeks, and it's pretty long). This is where I integrate most organizational strategies/processes I come across, and this is where I re-evaluate my tasks and projects ABCD and numeric priorities. I'll also defer some to a later date (without actually touching the original due date...), and delete or file away stuff that I decided not to do, etc. So yes, I mainly use GTD system and philosophy (I read the 2 main books) to a- Record /collect stuff(inbox), b- triage it (next action, projects, defer, delegate, archive, calendar, trash...), and c- Weekly Review.



I think this is a pretty good summary of how I prioritize stuff. I've skipped some details, of  course.

(I intend to now integrate the military strategy outlined in the article from you first post -- using a template of items and a column equation in IQ. It seems to be simple yet effective -- as a second strategy, of course, as this doesn't tell you anything about what you value, what's important to you, etc.)


[Edited as I forgot to answer a couple questions]

[Paul Keith]

Hi Armando,

In IQ I was able to set my priority system exactly how I wanted it. It's pretty straight forward, but with a few twists and complexities most people wouldn't care about.

I'm only one person but I can guarantee you, I do care to know about these "twists and complexities".

Could you be bothered to make a screen recorded video? Man, the more I hear about your process, the more I think it would help substantially if it can be viewed in action.

I could even try to provide you with a messy list to churn out if you're concerned about the privacy of your tasks list.

[Paul Keith]

Soft Problem/Hard Problem = Cultural Feasibility/Systematically Desirable

Victims/Beneficiaries - Who are the beneficiaries or victims of this particular system? (Who would benefit or suffer from its operations?)

Producers – Who are responsible for implementing this system? (Who would carry out the activities which make this system work?)

Transformation – What transformation does this system bring about? (What are the inputs and what transformation do they go through to become the outputs?)

Weltanschauung (or Worldview) – What particular worldview justifies the completion of this task? (What point of view makes this system meaningful?)

Destroyer - Who has the capacity to abolish this system or change its measures of performance?

Environmental constraints – Which external constraints does this system take as a given?

Modified from: http://en.wikipedia...._systems_methodology

[Paul Keith]

I don't know where I got this or how I had this written down in my old notes but these sets of categories are sort of reverse prioritization categories: instead of prioritizing which task you need to do first, this prioritizes how much you need to drop or set aside a task.

Unsatisfactory -Tolerating-Longing-Graveyard

[Paul Keith]

Source: http://www.triz-jour...1997/11/a/index.html



Compromise + Idealized Final Result (or how much does this task add to your ultimate goal in life)

[raybeere]

I understand the appeal of a lot of these methods. They appear to offer a more precise method of sorting out your priorities. But the drawback to every single one of them is that an individual - or worse, a committee - must assign values to each of these supposedly objective criteria. And doing so is always subjective.

Yes, sometimes it is easy. But, to use an example, the military method of selecting targets which takes into account how quickly a damaged target could be repaired seems objective - but if you look at it closely, it only works well in those cases which are pretty obvious. I can think of potential military targets which could be repaired much more easily - or less easily - than others. But those are the easy choices. When it comes down to the hard choices, the assessment of relative ease of repair relies on estimates made by one or more officers. That isn't to say they will never get it right. Of course, experienced officers would often arrive at a fairly accurate estimate.

The point is that, in that or any other situation, there are so many things which can go wrong during the process. Being unaware of a single piece of information can affect the accuracy of such estimates. Discounting a single point can have the same effect. And so on.

My system, although it relies more heavily on gut instinct, also forces simpler choices and averages out the result. Those same experienced officers would no doubt have pretty good gut instinct. And, if faced with only two targets to choose from instead of a whole list, they'd usually make the right choice. So, if you force them to make choices of one target out of each possible pair, then calculate the most important targets from that set of "votes", the final result is pretty reliable. Yes, the method still has weaknesses - in any method, if those following the process are obsessed with a single issue, that obsession will dominate the results. But, by reducing things to a set of - mostly - easy choices, then calculating relative merits based on those choices, it does smooth out a lot of errors introduced by more complex systems.

Personally, I find the other systems most useful in the way that they provoke me to think in different ways about my choices. Over time, incorporating new ways of looking at issues into my routine tends to be helpful. But if I try to follow the actual process, it trips me up. Every time.

Of course, every individual is different. If you have a method which works for you, there's no reason to abandon it because someone else can't make it work. But just because that method works great for you also doesn't mean that it will prove useful to anyone else. I am one of those people who sees so much complexity - in anything - that if I try to analyse specific points, I'll get bogged down. The single question, if I were selecting military targets (to stick to my example above), of how easy that target would be to repair once damaged has many facets. How complex is it? How many spare parts does the enemy have? How many people trained in making such repairs? What materials are needed, and how short are they on those? You get the idea... So forcing a (hopefully informed) "gut" decision between pairs, then using the power of averaging to smooth out my mis-calls works for me.

[Paul Keith]

While your post is pertaining more to how to narrow down wicked problems, raybeere, I think you captured many of the essence of why this needs to be a survey.

I apologize if I didn't go to this detail in the TS. It's just that I felt it would hold back and keep people from sharing their methods by being tl;dr.

The bottomline is this:

Personally, I find the other systems most useful in the way that they provoke me to think in different ways about my choices. Over time, incorporating new ways of looking at issues into my routine tends to be helpful. But if I try to follow the actual process, it trips me up. Every time.

...yet by this same token, you can't ignore the fact that most productivity users want to take and get something to work rather than the opposite and share so we could all compare and discuss and actually have a bigger more concrete pool to provoke and improve our prioritization methods.

I don't mean to single you out but since you provided the meat, I hope you don't interpret this as something meant to insult you:

Even after you wrote out that well thought out post, did you list your prioritization methods?

This isn't so much to point out how you didn't contribute as much as to highlight how even those who would realize the quality of subjectivity in prioritization would still be apt to keep their own methods to themselves.

Thus for those who feel the flaw of this thread, I hope you think that through.

For just like the military needs to ask some more questions...nay even create an argumentation template map, hopefully those who might be reluctant to share might also wonder:

"How else can people know how complicated each person's prioritization methods are if I and many others keep mine hidden away?"

"How many spare categories and additional process does all my Co-DC member have compared to mine?"

"How many people are actually blessed with methods that would improve my own?"

"What categories or techniques do they use and how complex or simple are they compared to my own?"

That said, I am being a hypocrite for keeping the specific categories I use but I fear such an insertion would miss the soul and necessity of this thread. Call me a cynic, but right now, I'd rather have the bolts and pieces on the table because we all threw our knowledge and research on the table.

Later on, all this talk about subjectivity and gut instinct, if enough people really want to gather around and attempt to merge them to save those unproductive people who can't form their methods some hope. Ok, let's do that.

But right now, as much as I would like to apologize for seemingly trying to emotionally blackmail everyone to empty what's in their pockets, desperation trumps incomplete criticism. The more things are on this topic, the more chances an unproductive person who stumbles upon this topic may upgrade their gut and maybe even switch it on to the "more useful objective criteria that needs to be given subjective values" setting from the "less" or "zero" or "supremely flawed criteria" settings.

[Armando]

Even after you wrote out that well thought out post, did you list your prioritization methods?

-Paul Keith (June 17, 2010, 04:05 PM)

Well, I think he did :

My system, although it relies more heavily on gut instinct, also forces simpler choices and averages out the result. Those same experienced officers would no doubt have pretty good gut instinct. And, if faced with only two targets to choose from instead of a whole list, they'd usually make the right choice. So, if you force them to make choices of one target out of each possible pair, then calculate the most important targets from that set of "votes", the final result is pretty reliable. Yes, the method still has weaknesses - in any method, if those following the process are obsessed with a single issue, that obsession will dominate the results. But, by reducing things to a set of - mostly - easy choices, then calculating relative merits based on those choices, it does smooth out a lot of errors introduced by more complex systems.

-raybeere (June 17, 2010, 01:29 PM)

I use this a lot too. It's similar to what's called "Paired Comparison Analysis".

[raybeere]

I'm a fan of robust discussion myself. So I'm not easily insulted.

But I am afraid Armando is right. I did explain my method. It is just a deceptively simple one.

List all the items to be prioritised. Break that list up into all possible pairs, and for each pair, choose the most important item. Tally up the votes, and calculate the relative importance of each item.

Since this is very clunky to do by hand, I did write a - clunky - WordPerfect macro some years ago (1998?) to let me input a list of tasks, then display the paired choices and do the calculations for me. It is limited to a list of certain length, since the clunky routines I used to keep track of things had to be copied and expanded, and I got sick of doing that. I decided I could always do that later, if I needed to. So far, I haven't needed to.

I'm not sure I'm ready to share that, as I'm afraid the amusement it would provide the coders among you might prove so overwhelming it would turn fatal. (In other words, you'd all laugh yourselves to death.) I suspect something similar could be done in AHK, but there are also undoubtedly much more elegant solutions than the one I used.

[Paul Keith]

Oh, sorry for the misunderstanding.

I guess in this case the question would be, what general terms/questions you use to separate the pair and decide on. (unless you meant you just stick to the criteria of which is most important)

I'm actually a huge fan of paired comparison analysis but since I don't know how to make a macro,  I'm often stuck with this

Still my issue with it is that, in the end it's still just a pair and you're still just replacing numbers with pairs. I still dream of a full blown questionaire ala something like this

For example, even the Covey Matrix can branch off into 4 areas all revolving around the concept of importance.

If it just sticks to that question though, again, I apologize for misunderstanding.

[raybeere]

As I've used it to date, it just sticks to the basic question: "which is most important / has the highest priority", although that decision may be informed by particular considerations.

This discussion has me thinking, though. Ideally, I'd like to set up a system which - for each pairing - allowed me to select one among various different detailed ways of determining the answer to that question. I'd even like to have it so the user could add custom criteria sets. That's what I'd like to do, now that you've got me thinking, but I fear my skills are not up to that challenge.

At least the version of WP I used for my macro would not allow me to use any type of array (I forget whether arrays were possible but too simple for my purposes or just not possible), which is one reason it was so clunky. In place of arrays, I had to use specific routines. Trying to alter that system so that every time I am presented with a pair to choose from I am also able to choose from a more detailed method of making that choice is just not feasible.

I don't know if it would be easier to set up a routine like that, with the extra choices, in AutoHotKey. I'm still learning how to use AHK. So I have no idea if I'll update my system or not. If I do, I'll share it.

In the meantime, a little more explanation on my own needs for any such system. I often find myself needing to set priorities among a list of things which include a certain number of similar tasks - as well as some very different ones. So I find myself comparing oranges to oranges, apples to apples, and apples to oranges. That's one reason I needed such a simple, flexible system. The methods which work in one area won't work well in another. So, ideally, I'd use the basic framework of Paired Comparison Analysis (I'm pretty sure that was what sparked my own macro, although I can't recall precisely so many years later), but apply a specific set of questions to certain pairings. Say, an "orange evaluation score" for comparing oranges, an "apple evaluation score" for comparing apples, and then the more basic "gut check" when dealing with comparisons between apples and oranges.

But such a routine would only really be useful if I set it up to accept custom checklists / questionnaires, whatever you wish to call them, with custom scoring methods for each one. Each item would be assigned a category, then items with matching categories would get the appropriate questionnaire when their pairing came up, while mismatched items would fall back to the basic "which one trumps the other" question. It would be a great system, but the code to do it would be hideously complex, at least at my skill level.

So, no, I haven't been using a more complex system, but you've convinced me it would be a great idea. I don't know if it will happen, though, because doing that all manually would consume more time than it would be worth, at least for me. And unless I get a lot better with AHK than I am now, I have no chance of setting up something that complex.

In general, I have a lot of great ideas for apps to do this and that. My problem is, I can see how they'd work in fair detail - but getting them to work is a struggle. I've actually read that software development is sometimes 'split' between one person who maps out how the app is supposed to work, while another person implements that map. I find the second part much more difficult than the first, which can be frustrating. But if I do ever manage to implement the system you have me thinking about, then I'll share it.

[Paul Keith]

Thanks raybeere. Looking forward to when that day comes.

Have you had a look at either the Form letter Machine or TFdocs?

I still don't get TFDocs and I still used TFLM for a sort of casual user level way of combining questionnaires but part of what made me want to learn TFDocs then was because of the potential of custom checklists/meet questionairres.

That said, I still don't get it and haven't learned it but outside of the scoring system, some of the demos made it look like it was possible to achieve your idea. I wonder if you'll get an easier time of grasping that service even though it's different from a local application.