Sociologist Sam Ladner was direct when she opened her session on reputation management and monitoring. “We won’t talking about reputation systems”, she said. “eBay reputations will not be part of our discussion. Read Bryce Glass’ blog if want to know more. He also has a Slideshare called Designing your Reputation System“.

In fact, Dr. Ladner led an amazing session in which she explored the meaning of reputation, attributes of reputation and the impacts and contexts of brand conversations that could be reputation impacting.

Despite the response of McNeil and the amazing brand-recovery case study that is the Tylenol cyanide poisoning incident of the 80’s, the company still took six days to respond. Can you imagine a six-day response time today? How would a six-day delay convert in a twenty-year span?

To setup the discussion, Dr. Ladner presented the three elements of the Looking Glass Self:

• We imagine how we appears to others
• We imagine how others judge that appearance
• We react to that imagined judgment

Consider that reputations cannot be managed. To help understand why, Dr. Ladner outlined the three key attributes of the Online Self:

• Hidden (online sources lack contextual cues)
• Digital (easily broken down, re-arranged, mashed-up and rearranged)
• Proliferating and Permanent-ish

What’s most interesting is where brand discussions are taking place and how that context affects the brand reputation and the opportunity of individuals and corporations to participate in that discussion (Forester Research and Statistics Canada):

• 48% of North Americans participate in social computing
• 30% of Americans have posted online ratings
• The average Canadian spent 35% more minutes talking on the phone in 2003 than in 1997

Dr. Ladner walked through a number of online tools and services that allow individuals and companies to monitor reputation.  There are several classes of tools available:

• Social Media analytics (Technorati, Google Trends, ePinions, etc…)
• Social Media observation tools (Neilsen Buzz Metrics, Cymfony)
• Social Media Response Tools (Umbria, TruCast)

Using examples that involved these tools, we learned of a number of studies in which included brand reputation of breakfast cereal based on health and nostalgic references, and brand reputation based on online attitudes on sustainability.  The examples were incredibly interesting and I would have been grateful for an extra hour to explore these examples in more depth.

Before leading a more interactive discussion, Dr. Ladner proposed some reputation monitoring best practices:

• Systematic (develop standard metrics, stick to them)
• Regular (measure at consistent intervals)
• Governed (assign accountability for metrics, create a task force)

We were all encouraged to use Google Labs to do our own research on reputation conversations, offering that we research Dell and Best Buy together over a period of time and look for when the Dell announcement on selling their computers at Best Buy.

We were able to wrap up with a more lighthearted discussion on the doppelganger effect.  I guess there are some advantages to having a one-of-a-kind name.

Note: this session will be available in slideshare.

Last night’s Third Tuesday Toronto had a more relaxed atmosphere than the events I’m familiar with in Ottawa. Perhaps that was because of the size of the crowd (standing room only in a room with a capacity of 160) or maybe because I was sitting with the wild kids at the back of the room (Jay Moonah, Eden Spodek, Doug Walker, Dave Fleet and ring leader Collin Douma).

Michael O’Connor Clarke moderated a discussion on measuring social media with panelists Katie Paine, President of KD Paine and Partners and author of Measuring Public Relationships, Marshall Sponder, Chair of the Web Analytics Association`s Community and Social Media committee, and Marcel Lebrun, President of Radian6.

Most of the discussion centered around the need for metrics and how they should be presented to clients or management as well as some of the elements that can be measured.  Ms. Paine’s direct approach which includes the ‘Suckiness Factor’ boldly displayed on charts and graphs was by far the most interesting and entertaining.  I had hoped for more discussion on methods for collecting metrics and specific case studies on the collection, presentation and influence of metrics in decision making by clients.

The panelists’ parting thoughts on social media and metrics ensured that the evening wrapped up on a high note.

• Katie Paine: ‘measure how much social media traffic you get’
• Marshall Sponder: ‘listen about you’
• Marcel Lebrun: ‘listen about the marketplace’

I’m off to the Mesh Conference.  I hope to have the opportunity to blog during the day.

Many podcasters look for ways to measure success. The most common, and possibly most validating, is monitoring statistics such as subscription downloads, web downloads and plays through an online player. For each download we can track the country of origin, date, day of week, time, and frequency of downloads. These numbers can be used to create pretty graphs — many of which are generated automatically — and software tools allow us to chart trends.

Despite all of this amazing technology, we are still unable to measure how much of each episode was listened to and whether or not the listener was actually engaged in the program when they listened. Measuring this level of engagement sounds like having the ability to generate a podcast consumption box score. Oddly, it’s the level of engagement that makes the most difference.

It’s easy to become addicted to the numbers and overlook the personality and purpose of the podcast, and the community which it serves. I know that Andrea and I have spent various periods of our “podcasting careers” refreshing the stats pages of our sites on a very short cycle. That was until recently when an upgrade of the Podpress player caused a problem that could only be overcome (bypassed, if you like) by disabling the stats collection function.

Try taking a step back from following your stats this week. Use the extra time to read a book, relax, or (if you’re like me) work on additional creative projects (podcast or other). If you make it to the end of the week, see if you can make it to the end of July and then to the end of the summer. Since stats historically decline over the summer, it’s probably best for your ego to ignore them now, anyway. If you still need your stats at the end of the summer, find a compromise between obsessing and ignoring.

It’s easy to hold on to traditional ways of doing business and measuring success. The music industry is one of the best case studies to demonstrate this. They have been completely unwilling (to be fair they may lack the agility) to identify new business models that are consistent with the digital age. Instead of investing to adapt and advance, they invest heavily into lobbying for legislative support to make a traditionally civil matter a criminal matter. All the while, they continue to hold on to dated cost structures and charting styles.

So, what are some of the ways of measuring penetration and success?

HIT COUNT

While becoming increasingly less dependable, one way of identifying penetration is by looking at how many hits a website has attracted. The home page is a terrible way to confirm that a Podcast has built an audience. By establishing a “buried” page as the measurement tool through some form of call to action within a Podcast, the Podcaster can more reliably identify how many listeners actually listened, and reacted, to the Podcast.

DOWNLOAD COUNT

Technology to measure listens versus downloads is still in its infancy and the tools that are being built and tested are proprietary. Until that is worked out, download count is just that - a count of the number of times a media file has been downloaded.

REFERRER LINKS

Social networks have established a new way of identifying how much penetration a Podcast has achieved… referral links. Referral links are links from other sites to a specific target site. There is a direct correlation between the most popular episodes of a Podcast and the number of referring links to those episodes.

FEEDBACK

Encouraging listeners to submit comments on a Podcast, or a particular subject discussed within a Podcast, makes it possible for the producer to do two things: engage the listener, and measure response. There are two popular types of feedback. Text feedback can be sent via email, typewritten in the comments section of a blog page, and contributed to active forum discussions. Audio feedback is becoming increasingly popular and can be sent as MP3 (or some other format) attachments in email, through voice messaging services that forward digitally encoded versions of the messages to the Podcaster via email, and web-based audio commenting tools.

POSTAGE STAMP TEST

Michael Geoghegan introduced me to the idea of identifying the cost of producing a Podcast and averaging it across the number of downloads, and comparing that to the cost of creating a postal mailing to reach the same number of people. What’s particularly intriguing about this idea is that if a business is going to hold on to dated ways of doing business, why stop at the beginning of the Internet?