Keeping users coming back for fresh content requires a delicate balance on broadcast web sites. Comments have a big effect on traffic. If a story gets negative comments, should you leave it up and let the people bitch, or rotate it out for fresher content? New research coming out of HP’s labs reveals some unexpected trends. The surprise is that on-line discussion produces markedly different results than traditional face-to-face discussion.
HP tracked all the commenting activity on some of the most popular web sites in the world – Amazon, Digg, YouTube, Netflix. These are businesses where comments are the very foundation of the site. They help the audience decide how to spend their most valuable resource – their time. Digg’s comments and voting system help readers painlessly sift through the minutia of the web and get right to the good stuff.
HP’s research focuses on how people pay attention to things on the web. What do they notice, then invest their time in? This is called social attention, and it is the science of buzz in the on-line world. So on a typical broadcast web site, should you put the most novel things at the top or the most popular? HP studied the interplay between popularity and novelty.
On Digg, the average life of a story is 69 minutes. That is how long it took the average story to fall out of the most popular rotation. These stories live and die based on the viewer commenting. So once you notice these trends, how can you enhance the experience to make story buzz live longer? What kind of comments spell doom for a story and which help light the fires of buzz?
There is a very big difference between opinion on the web and off-line opinion. In the real world, there is a phenomenon called group polarization. Research shows that a group of people discussing a point face to face tend to migrate to the more extreme opinion or the more risky course of action. For example, they asked mock jurors to estimate the amount of punitive damages before deliberation, then compared those estimates with the group decision. Face-to-face group discussion tended to move an opinion towards the extremes.
What was interesting is that opinion on the web tends to do the exact opposite. When you track how on-line opinions form about books, movies and videos, you see no evidence of group polarization. Reviews and comments tend to be more moderate. And the more opinions you have, the more moderate they become.
If you sit down and really think about it, this phenomenon makes sense. If a person takes the time to write a review on YouTube, she is committing to use her valuable time to voice an opinion. It will require some real effort on her part to put sentences together and create a relevant and interesting comment.
So let’s say 100 people all watch a particular YouTube video and comment that it is fantastic. If you’re the 101st person to say the video is great, it just doesn’t seem that important to pile on your similar opinion. But if you hated that video that everyone else likes, then commenting seems more relevant. People with a contrary opinion are more likely to invest their time in writing a review. So people who disagree with the prevailing opinion are more likely to show up and voice their opinion on-line.
So a video clip starts out with top ratings and top reviews, but as time goes on, the reviews will get slightly worse as contrarians leave their mark. So whatever starts popular will gradually be brought down. Whatever starts down will gradually be brought up. Clips that are considered terrible at the very beginning will gradually gain in popularity as more people join the conversation and proffer a differing opinion. This has an overall moderating effect on everything on line. Nothing is too wonderful – but nothing is too terrible.
So what this means is you can help a video clip’s review score if you know how to time its introduction. Let’s say you have a fantastic clip that you think will be a Digg superstar. If you want a very high positive viewing score, then posting the clip during high traffic times makes sense. Get in there and get those positive reviews before the contrarians can bring your rating score down.
But let’s say you have written a book that is on Amazon and it wasn’t your best effort. Your book is likely to get slammed right out of the gate. In a situation like this, a long lead time makes more sense. The initial negative comments will eventually be mollified by contrarians who will come to your rescue with positive feedback.
The same is true with the movie industry. If you have a bad movie, then you might want to do a limited release just to get all the bad feedback out of the way before the masses check out the on-line reviews. By the time the movie is released to the general public, the group will probably bring you up to a moderate rating.
So how can you apply the knowledge to decisions on content rotation on your own site? Well, if you have a great clip that is sure to be a hit, feature it prominently and encourage commenting in the early stages. But once you’ve gotten that initial wave of positive comments, then you might want to stop encouraging readers to chime in with their opinions. This will keep the initial euphoria from the moderating effect of the contrarians.
Conversely, don’t be too hasty to pull down a great clip that gets a few bad reviews in the initial stages. If you are confident the clip is good, then stick with it. The longer you leave it on the site, the more those initial bad reviews will be moderated by the second and third waves of commenters.
–Graeme Newell, President of 602 Communications, is a broadcast and web marketing specialist. He guarantees that his teasing seminar will immediately increase your news ratings or his workshop is free. He can be reached at [email protected].
