I’ve thought for some time that what we’re learning from the use of tactical web video in the 2008 presidential campaign will inform and inspire marketing and communications well beyond politics in 2008 and beyond. And as creators and producers of Mitt TV for the Romney for President campaign for the last 14 months, we’ve been privileged to have our hands in the incubator, so to speak.

So with the Romney campaign over and the experience fresh in my mind, I thought it would be useful to start writing down the key lessons we’ve learned from building the most sophisticated internet TV channel (and supporting Tactical Video Unit) for this campaign cycle.

Here, first of all, are my ten most important observations:

1. YouTube is a two-edged sword
2. A YouTube channel is necessary but not sufficient
3. A content-managed video platform is vital to success
4. When choosing a platform, back-end workflow is a key decision criterion.
5. Prepare for success: use a CDN (content delivery network)
6. Maintain one asset collection, many faces
7. Build a fast & flexible production process
8. Seeds and feeds build viewership
9. Don’t believe everything you read about clip length
10. Listen to the data

Each of these will merit its own entire post in the days and weeks to come, but let me take a moment to write an additional sentence or two about each; the elaboration on each topic will come later in this blog:

1. YouTube is a two-edged sword. A lot of campaigns are relying largely on YouTube channels generously “given” to the campaigns by the video site. And there’s no question that a lot of independent traffic saw our clips based on YouTube searches. But if part of the idea is not just to inform but also to inspire people to act (give money, sign up, give us their email, etc.) then YouTube was weak at the “call to action” part. And it’s just as likely that the NEXT video presented might be a competitor’s “oppo” clip.

2. A YouTube channel is necessary but not sufficient. Because of #1, we decided to invest in our own very rich channel that eventually had more than 400 video clips on it, which we control and tailor much more precisely and which were closely associated with our “calls to action” in the campaign. And the most remarkable statistic of all is that more people watched the Romney campaign’s clips on Mitt TV than on our YouTube channel.

3. A content-managed video platform is vital to success. Just as content management systems are key to maintaining lively, dynamic text websites, a content-managed web video publishing system is vital to building something as sophisticated as Mitt TV. We used PermissionTV as our platform with great success.

4. When choosing a platform, back-end workflow is a key decision criterion. One of my costliest mistakes in choosing a publishing platform was allowing the decision to be swayed more by how the channel looks than how it works on the back end in day-to-day production. The PermissionTV channel was beautiful and created a wonderful user experience, but its back-end was hellishly challenging in the early days (they’ve made good strides forward, based on our feedback).

5. Prepare for success: use a CDN. One of the most common mistakes I see webmasters and IT guys make is the decision to host and serve video from their own servers. And then they wonder why the lights dim and nobody can access the internet when a video clip gets popular. With many Mitt TV clips viewed simultaneously by hundreds of viewers (do the math: 100 simultaneous viewers = 50 megabits per second), a CDN is not optional – it’s essential.

6. Maintain one asset collection, many faces. With a large collection of video assets and a content management system, we were able to quickly create multiple sub-collections of Mitt TV, tailored to particular issues (see “The Briefing Room”), states, and even fans of Ann Romney via Ann TV.

7. Build a fast and flexible production process. As a documentary filmmaker, my mindset is often locked onto projects that take anywhere from six weeks to six months. But back in the seventies, I was also a daily news reporter…and it is that mentality of quick turnaround that I’ve had to bring to our tactical video unit. New tools help this: capturing to disk rather than tape, portable editing systems, on-the-fly video compression. These tools enabled us to measure camera-to-net times in minutes, not hours or days.

8. Seeds and feeds build viewership. While we had our share of “I want my Mitt TV” people tuning in every day to see what’s new, the key to building Mitt TV’s audience (which got as high as 70,000+ viewings a day in the late stages of the campaign) was outreach – to bloggers, through press releases, through RSS feeds. The result is that there are more than 23,000 references to Mitt TV on Google and more than 2,800 sites linking to Mitt TV.

9. Don’t believe everything you read about clip length. The conventional wisdom is that video clips need to be under 2 minutes to have a prayer of getting watched. But looking over the viewing statistics, we see that many of the most popular clips were complete speeches or events that were as long as 20 minutes or even more. For example, the 21-minute “Faith in America” speech was the third-most-highly-viewed clip on Mitt TV in December and January, and nearly half of the viewers watched every minute of it.

10. Listen to the data. One of the advantages of having your own internet TV channel is that there is a tremendous amount of data available about what works and what doesn’t. You can watch patterns of viewership and correlate it to different outreach efforts. You can see which clips are being viewed and for how long. And you can see where traffic is coming from. All this is useful in making a more effective channel.

The lessons of Mitt TV are valuable ones, and it’s too bad we won’t be able to apply the valuable data and learning to complete the campaign. Such is the nature of the winnowing-out process.

But the lessons are equally valuable to a wide range of our clients, across commercial, education, and not-for-profit domains. We don’t have to wait till the next campaign cycle to apply what we’ve learned; we’re already reviewing every one of our video production and web video clients to see how we can take the lessons of the campaign out to their marketing and communications efforts.

The incubation period is over.  Let the real fun begin.