Matt delivered as promised. And, as I wrote to him, I think I asked the exact right guy to answer this question. Of course, with a handle like Techno//Marketer, what else should I have expected?

This post concludes the first QAD series, thanks to all who participated and contributed. And a special thanks to Philippe for starting it all, to Kris to challenging me to answer and Matt for going above and beyond with this guest post. Now, on to the final question:

What does interaction rate tell me about the impact of my rich media campaign?

Answered by Matt Dickman

In this Web 2.0 world, smooth interfaces and transparent technology have introduced a new set of measurement challenges for marketers. If you think back to the early days of marketing on the net, the metric you heard about (and scarily some misinformed people still say this) was hits. People would ask â??How many hits do you get?â?�. The problem with hits was that you could inflate the number just by adding more elements to your web pages. Page views swiftly moved in to take the place of hits and all was right with the world. The same measurement challenge is happening in online advertising.

Flash and AJAX technologies have gone a long way to creating more engaging ads. The challenge with the new technology is that traditional metrics arenâ??t making sense anymore. Flash technology is allowing much more content and interactivity to be placed in ads. So, how do you measure engagement beyond the click through?

Interaction rate (IR) has stepped forward as the de-facto measurement for rich media ads. Letâ??s look at what it means. First, itâ??s interesting to note that IR is not a standard and different companies and websites measure it in their own home-brewed way. I think PointRollâ??s definition is the best to date. They determine IR as â??the total number of unique interactions per impression divided by the total ad impressionsâ?�.

So letâ??s say we have an ad that gets served to 100 unique users. Depending on your message there are a couple of ways this could play out:

1. Display ad asking for a click-trough. Letâ??s say we get 10 clicks. So 5 interactions divided by 100 impressions gives us an IR of 10%.

2. An ad that asks users to expand the ad and play a video. For this one letâ??s say 10 people expand the ad and 5 of those play it and 2 click through to your site. Thatâ??s 17 interactions total. So 17 interactions divided by 100 impressions gives us an IR of 17%.

The second ad is much more engaging to the user and theyâ??ll be more likely to remember the ad in the future. This is the key to any ad campaign and online advertising allows you to create more interactions with your message in a targeted, consumer friendly manner.

So what is an interaction? Well it depends on your campaign goal, the ad you create and the type of options the user has. Straightforward display ads serve one purpose, to draw your attention and get you to click. Getting the user to (possibly) see the impression is all youâ??re looking for here. These ads can be built in Flash or with static graphics. Flash, however, allows for deeper engagement with the users. Embedding video, presenting content on multiple tabs and embedding forms are standard faire and they create the ability to have a micro-site inside a standard ad size. See how this changes the game? Itâ??s possible for you to present all of the information in the ad that you would on a landing page. Seeing how the users interact without a click-through is crucial.

Not all interactions are created equal. Turning on and off the audio (you donâ??t have audio on by default do you? I hope not), pausing video or closing the ad are not true interactions. Also, there are actions that automatically happen inside ads that are not true interactions like video playing on load or ads which expand automatically. Be clear when creating agreements with companies who are basing your campaign on IR that these false positives and automatic loads are not counted. You are looking for real numbers.

What makes a good interaction? We need to find out a) when a user sees your ad and b) if they are driven to engage with you. Here are some examples of good interaction in ads:

• Teasing users to click a new tab with more information
• Adding video to the tab which needs the user to click play
• Allowing a user to click through from the video to get more information
• Adding simple games to the ad which the user can play without clicking through to your site
• Embedding mapping and/or directions to the ad itself where the user enters their information
• Teasing users to expand an ad to get some of this content
• Including a form to receive more information or sign up for a newsletter

So what does this rate really tell you? IR is one of those metrics where data can confirm marketerâ??s subjective notions. There will still be users who block ads, users who see your ad and donâ??t interact or click, or users who tune out ads completely. There is nothing you can really do about those groups except fine-tune messaging. Also keep in mind that the interaction rate depends on the ad being in front of the right people. If the media buy is bad, the IR will suffer. But for those people who do see your message and do find it interesting and do click around in your rich media ad, itâ??s a powerful way to see how engaged they are with you.

All Questions

• How far should I go in the dialogue with the users? Can I accept controversy on my website?What moderation level is acceptable?
• Does online advertising make sense without a decent website?
• Are there examples of 2.0 initiatives made by traditional brands that went totally out of hand?
• How can impressions be compared to television GRPs?
• How intrusive should I be? (expandable formats, videos with sound on by default)
• What does interaction rate (only available for rich media formats) tell me about the impact of my campaign?
• Does the long tail change anything to the way I should communicate with my target group?
• Why on earth do people use sites like Second Life?

This entry was posted on Thursday, May 3rd, 2007 at 4:49 pm and is filed under Question A Day. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.