[adwade]

I'm interested in a strange idea I've never seen implemented, but think it would be useful to a lot of admins if it was available.

Basically, I'm after two pieces of information:

1)What's the best thing about our board?

2)What's the worst thing about our board?

In the world of statistics, there's a something known as a pareto chart. It's used to rank how important (or unimportant) various problems are.
See Pareto Chart

In theory, this would tell me what's working correctly and pleasing folks, as well as indicate what's wrong and needs fixing too.

So what I'm looking for is FEEDBACK from various (random?) users about how we're doing as a website. What I envision is some form of hack that would either email, personal message, or send the user to an online form to fill out where they could answer questions #1 & #2 above.

I guess the trick would be to determine a frequency for doing so. I don't want it to become nag-ware, so maybe poll a random user after they've visited the site like 20 times. Then again, say after another 50 visits. Later, after 100 visits, etc. Might be nice to be able to control the frequency and/or the number of members I want to poll from the Admincp.

Anyway, that's my idea. I glean this from Tom Peter's IN SEARCH OF EXCELLENCE book, principles of business and am just looking for honest/forthright feedback from our user base.

If there's something out there that does this, or could easily do this, nevermind. Just point me in the right direction to it.

Thanxx, AD

[Code Monkey]

Not that I have heard of. How would you categorize individual answers? In the pareto charts you show there are specific questions asked. So maybe it's just some variant of the current poll system that you are looking for.

[adwade]

Well in real life, you typically do field interviews (or bean-counts) to determine: 1)What the catagories are, and 2)Then priortize them from worst problem to least problem. For instance, if you have field work orders that are failing to get done, why? Often, it's because of manpower, materials, lack of support from other groups, other higher-priority work took precidence, etc. Then, you just go about sorting those out to find what your worst problem is that's causing you to miss so many completions on work orders...and that's where you concentrate on a fix first.(i.e. most bang for the buck, when it comes to fixing things)

However, that's far more than I need. I don't need a pareto chart at all. For 'my' purposes, I just need the feedback so I would know what to work on.(i.e. "LINKS suck, they're too complicated, not sortable, and many are no longer even valid") The 'best' feature of the board is more of a fishing expedition to learn what you're doing right, so maybe you can apply those practices to other areas of your board.

The problem with relying on volunatary feedback is you won't get much, and what you do get may just be by the 'whiners'.(and therefore not even valid) A good cross-section of your members giving you feedback should help out, if you're interested in improving your board.

So in the end, I'm really just after one-liners to point me in the right direction to fix things that I may be overlooking, or not consider as important as my members do. The important feature of such a mod would be the randomness (member-wise) and controlling the repeating frequency, at least in my mind.

[adwade]

Pareto basically said that, 80% of problems usually stem from 20% of the causes. And In most cases, two or three categories will tower above the others when it comes to problems. taken from Pareto chart helps you focus on the vital few - those few things that cause the most problems.

So what I'm after in a MOD like this, is to simply learn what do the members perceive as 80% of the problem with a given board. Given that, I know where to concentrate my efforts and get busy fixing things.

[Code Monkey]

I don't think such a mod exists. It would have to have it's own logic set and the ability to learn to sort all the variances.

In the online/computer world, surveys can be done by non human observers.

For instance, your example of the links. In this world here we simply run a link checker application and it returns fair and unbiased results on the status of all links and a complete list of those that fail and those that function. We can even keep track of which ones have been clicked and which ones take up space.

We can check google and see how many pages we have. This can tell us if we have a relavent structure to our data and we can resond accordingly such as implementing Search Engine Optimization methods.

I think the only thing that matters to humans in the online world is content. Everything else just manuevers them to your content. Such as this site is still very young and doesn't have a lot of content. So, people aren't inclined to stay if there is nothing for them here. As it grows the content will hopefully grow as well.

So it really it boils down to a few things.

1. How to steer the internet traffic to your site. Can be automated with proven methods.
2. How to navigate that traffic through your site when it arrives. Can be automated with proven methods.
3. How to continually increase and improve your content. Needs lots of human intervention.

With number three you need to find your audience and give them what they want. They will tell you what they want by activity not polls and pareto charts. Activity is the best responce you will get to any efforts. Polls and surveys just start the chain of complaints and let the loud mouths and bullies know you are offering a soap box for their wares.

That's really all you need to focus on. Leave one of the three out and you're like the little town on route 66 that has been bypassed by the new highway.

[adwade]

Just wanted to add, I found a MOD that will pretty much do this. It won't allow for repeat inquiry's, but it will collect one-liners about how you're doing on your website -AND- also keep a running average for you!

See easyFeedback - How do your users rate your community? - Page 2 - vBulletin.org Forum

[Code Monkey]

Yeah, it may work for you. Of course, the reason it has 11 installs is that noone likes those javascript things that follow you around the screen. Plus, you will not get data from the few members that have js turned out.

Let us know how it works out for you.

That's still just finite fixed replies though.