by Irfan Kamal
Category: Digital Influence, Measurement, Research & Insights
Earlier today John Bell and I formally introduced the Conversation Impact(TM) measurement model at the Advertising Research Foundation’s Audience Measurement 4.0. Here’s a brief overview of the model, its goals and planned evolution.
The model was developed by our team to provide brands with a comprehensive, recognizable framework for tracking social media campaigns. We relied heavily on our experience with a range of social media campaigns for both B2B and B2C clients, and considered the types of questions and reporting requests we receive with every new project or request for information.
We focus on simplicity and comparability across media - the latter, to help guide marketers with media allocation. We categorize our metrics into 3 areas, corresponding roughly to objectives and “marketing funnel” stages; each is shown below, with representative metrics (the metrics are selected based on unique client needs). Included are both familiar and new metrics.

Image courtesy of Crimson Hexagon
Reach and Positioning
Preference
Action
Some of these metrics we measure using software such as Crimson Hexagon, Radian6 or TruCast; other metrics we measure with more traditional panel or intercept survey-based instruments.
We invite discussion - this is the first phase of what will be an iterative process to arrive at a simple, effective way to help marketing and communications executives make better decisions on social media spend allocation and scaling.
Compassion in Hong Kong
June 23rd, 2009 at 6:22 pm
Irfan - Glad to see our some of our data and BI rolling up into your Conversation Impact model. Looking forward to continuing partnership.
@bcahill
@visible_tech
June 23rd, 2009 at 10:08 pm
Same here, Irfan. Glad to be on board, and happy to contribute our view toward a comprehensive solution.
June 24th, 2009 at 8:12 am
all of this is fine, but aren’t you missing a big chunk of the blosophere, and esp Twitter? How are you getting this data from sites that are either too small to show up on Compete or Quantcast and/or don’t publish their statistics?
June 24th, 2009 at 8:21 am
@KDPaine:
We include several sources of social media mentions in our calculations, including blogs, message boards/forums, and Twitter.
The issue of how to handle smaller sites is an important one. The unique monthly visit (UMV) measurement is only one part of our influence assessment. We consider several other characteristics of sites that affect influence, including content, relevance, engagement and recency of content. Based on these criteria, we often find influence even absent UMV data.
June 24th, 2009 at 2:23 pm
Hi Irfan,
Congrats to you/John et al. I’d be intrigued to know more about how you evaluate “share of positive voice in social media, within category.” Are there links to that information anywhere?
Thanks.
JCH
June 25th, 2009 at 12:40 am
Very interesting, will be following this development for sure.
July 3rd, 2009 at 1:09 pm
[...] Introducing Conversation Impact - Social Media Measurement for Marketers - Ogilvy [...]
July 7th, 2009 at 11:46 am
This is a great article for measuring social media success, however I’m not highly educated on the correct software to use. I would love to hear more about Crimson Hexagon, Radion6 and any others that offer such critical information for measuring success. Where can I find a comparable on the software?
July 9th, 2009 at 4:53 pm
@JCH:
Thanks very much. We’ll be posting more information shortly, including details on the model.
@Navah:
Since these software products evolve pretty rapidly, it’s hard to find reviews that are completely up to date. I’d recommend you check out the software company web sites (they are linked in my post) to get the latest info. You may also find it helpful look at John Bell’s blog for a recent post on how to choose social media software.
July 14th, 2009 at 4:08 pm
[...] positive word of mouth and even recommendation for a product will drive product preference (see our Conversation Impact measurement model), using social to drive sales will become a new focus for many of [...]
September 13th, 2009 at 7:15 am
[...] must have a system for measuring the effectiveness of any campaign - and followers or friends or hits isn’t it - and the ability to change strategy if things [...]
October 2nd, 2009 at 10:35 am
[...] read Agency Approaches to Measuring Social Media by Nathan Gilliatt recently and his post comparing Conversation Impact , Fluent 2009 and Digital Footprint Index, lead me to think deeply about all three approaches and [...]
October 5th, 2009 at 2:51 pm
[...] the long term value of social media was integral to creating sustained WOM programs. I shared Ogilvy’s Conversation Impact measurement model, which maps to the marketing funnel and focuses on metrics like reach (that tie out to existing [...]
October 13th, 2009 at 3:24 am
[...] reales, que luego se orientarán a generación de ingresos. Y existen herramientas (una de ellas Conversation Impact ) que va mucho más lejos que las métricas suaves para demostrar el valor real de los social media [...]
January 6th, 2010 at 4:26 am
I picked up on this article from a link in the latest post from “The Digital Influence Mapping Project” (6th Jan 2010). The point made by KDPaine above and the response from Irfan Kamal (24th June ‘09) seems to strike right at the heart of Social Media’s ability to make a real impact on the financial bottom line. My team and I agree with KDPaine that monitoring user generated content that ‘needs’ to be covered, rather than only what ‘can’ be covered is the elephant at the social media party. That’s why we’ve developed a web-scale ‘Influence Search Engine’ that resolves this problem.