Originally Posted by
kipper
I might reconsider where I stand on this if there was a clear plan for how to define success and failure, what happens if it's still split 50-50 at the end of the trial, and what happens if this is tried out sitewide and found to be a disaster very quickly. I see the complaints about the Premium Deals forum, and I'm not convinced that if this is found to be a disaster, TB would reconsider quickly.
Thanks for the reassurance that we're not playing Whack-a-Mole here.
We didn't have a detailed road map of the sort you want when we tested allowing images in posts. The fact that the horrible situations we feared never happened was sufficient. Even though only a small minority of members used the new feature, the images they posted had indisputable value. Once it was clear that disaster would not occur, a wider rollout occurred, if I recall correctly, by the Community Director without TalkBoard involvement.
Reader feedback is a bit different than images, in that its value is not as obvious. So we have something to demonstrate beyond just avoiding disaster.
In my mind, if a feature pleases 10% of our members and doesn't bother the other 90%, it adds value to FlyerTalk. If a feature improves post quality without bothering members, it adds value. If a feature improves post quantity without degrading post quality, it adds value. These are the three ways in which I expect reader feedback to improve FlyerTalk. This is why I want to test it.
A member survey could address the question of whether members are aware of the new feature, and if so whether it pleases them, bothers them, or neither. That's an easy one.
Improvements in post quality will almost certainly be subtle and essentially invisible. I don't see any way to quantify that, but I'd like to hear any ideas.
Increases in post count will also be small, since reader feedback is such a minor change. We could measure post count, but I doubt the numbers would be meaningful. There is also the
Hawthorne Effect to consider.
I want to make this proposal the best it can be, but there will be limits to our knowledge both before and after any trial. At that point we need to apply our own judgment. That's why TalkBoard exists.
A time-limited trial would add information. The only reason not to add information is if one's opinion is immovable no matter what the new information is.
I expect that TalkBoard will be able to see how IB's existing reader feedback capability looks soon. That may alleviate some concerns about a trial.