One of the great challenges with asking people what their preferences are on certain topics is that their answers aren't always accurate.

Sometimes they honestly don't remember what they did when asked about it after the fact. For example, families selected for Nielsen ratings research have to turn in a log each week of what they watched. If they fill it in right before it's due (rather than tracking activities during the week), they may believe they watched a program when they didn't, or that they watched all of a two-hour special when they really watched only a half hour. In the meantime, millions of dollars could be misdirected because of faulty data.

And, sometimes, people consciously choose to lie, or at least embellish. College students being asked in an interview about success with the opposite sex may decide to exaggerate the truth lest they look like dweebs to the researcher. Most of us probably would, too, especially if the researcher were attractive.

That is one of the great things about modern webcast and virtual-environment technology. Yes, you can track attendance, run surveys and tests asking opinions, or pose questions during the event in order to keep attendees engaged. All of those are important. But what if you could track their actual behavior? In so doing, you'd be able to see how your audience engages with certain content, experts, and even their peers. You can use that information to adjust the content, the conversations, and even the virtual environment itself to improve your efforts. After all, it's one thing to say you will do something; it's another to actually follow through and do it.

The most convenient and thorough way to track audience behavior is to use tools that are built into your webcasting software. With such tools you can determine what you would like to measure, then run reports that aggregate the behaviors of your attendees to present an overall picture of their actions. You can also drill down to individual users to track their actions—a very useful capability if you're trying to sell to a particular company and you see from the attendee list that employees of that company participated in the event.

Here's a good example of looking at the aggregated numbers. There are all sorts of ways of presenting information in an online event—PowerPoint slides, sharing documents on a desktop, video, audio plus photos, audio-only and so forth. By tracking audience behavior you can find which techniques create the most engagement (or at least keep everyone tuned in and focused) and which (if any) drive attendees to click out.

For example, you might assume video would be most engaging, and that HD video is preferred to standard. But if the bulk of your audience doesn't have the machines or bandwidth to handle HD video, it could cause them to disengage instead. By using the built-in reporting tools in your online event software, you might see that you lost 23% of your audience when the video came on, which would help you decide whether to include HD video in future webcasts and virtual environments.

Taking that concept a step further, you could also use reporting tools to see whether it's the HD or the use of video that is the issue: You could run standard video at the same point and check the drop-off rate. If attendees remain on the online event or in a virtual environment for standard video, you have a pretty good indication that HD is the problem. If they drop off at roughly the same rate for standard video, you know not to include any video in future webcasts or virtual environments.