Moderator: Marshall Sponder, Senior Web Analyst, Monster.com
Use comscore Conversational media category
- dont forget that the numbers are panel based
Let google do it for us
- Trends down to the site are now available
CMO’s will end up being WEB analyts or havint them nearby, high on the organizational food chain.
Bake in analytics to all marketing campaigns by auditing them, changing the campaigns before anything else is done. Right now, Social media has no
se place in most organizations
- whenever you sit down in a campaign, sit down and think about if this is something you can measure.
Web 2.0 is about empowering users. Its not the technology, its about letting people have a voice and communicate with each other.
Measure conversations - it can be done and this panel shows some examples of where we succeed.
Tools like Radian6 and others can measure certain things.
WAA is working on stadards for social media and the beginning drafts will be available this fall.
Rob Key, CEO, Converseon
Social Media Measurement
*warns that he is going to go quickly*
Expanding social media universe
- 45% of adult interenet users have created content online
- 1.2 million blog posts per day
Designing Social Media
- Phase 1 - listen
- Phase 2 - engage
- Phase 3 - measure/optimize
First know what you want to know
- how are people feeling about our brand
- who is influential
Reports: Product attribute tag clouds
Limitations of Automation
- a pure machine based solution cant pick up sarcasm
- technology is a long way away, you will need human intervention
How to use it?
- Extension of Customer Service
- some companies are mining data on a daily basis. They are looking for venues that they can firefight and avoid dell hell type issues
- its a mix of PR, CS, and marketing all at once
- Search results
- *shows chart that has social venues ranking well for big keywords*
- built blog strategy based on 6 months of listening
- positive sentiment increased 15%
Where does this all go?
- disparity of capabilities in social media monitoring and analysis will fall
- are all metrics really part of the same elephant
Start to look at how these things work together: sales, traffic, brand trackiing, conversation mining, conversion tracking
Breanna Wigle, CRM Manager, Military Advantage & Todd Parsons, Co-founder & CPO, BuzzLogic.com
- profiles and communities
- discussion boards
Have found that social media traffic converts 6% than non social.
Strategy: isolate the influencers and reach passionate readers of military defense news and information
Campaign goal: increase product awareness to (new) influencers and their (audiences) convert visitors into rss and newsletter subcribers
Challenge - finding influencers and advertising to their audiences manually is daunting, given the fragemented nature - there were 1000s of sites
and all in the long tail. They wanted to know the top areas to spend their time.
Step 1. Uncover conversations - started looking at same keywords they used on the site. They then went out and found sites that were about those
keywords. Other site targeting: same terms, different outcome. Networks in google adsense were forums, photosharing sites and reference information.
In the data, they found active conversations that linked out to other locations.
Step 2. Rank the influencers. How often do they blog on a topic? what is their reach? They targeted the influencers.
- creative was compelling, informative and had a clear call to action
They ran on a cross section of 250 blogs which included influencers and sites linking into the conversation
Results: 86% higher CTR campain compared to historical average for targeted banner campaigns. Direct conversions: 5.32% lift from pre to during. The
goals included newsletter and rss subscriptions. 90% of visits were new to the site. Users from the campaign stayed on the site 6.25% longer.
Key observations and learnings:
- active conversations about specific topics attract passionate audiences. Highly targeted display ads can perform in this environment.
- influencers and their network relationships
- this can get you closer to a search like intent
- the nature of conversation can impact ad performance
They say there is a ton of inventory. The problem is that there arent a ton of quality ad networks that work with blogs.
Q. price wise - how does it compare to yahoo/google?
A. blog content is less expensive, but what they do is go after the influencers
Edmund Wong, Vice President, Strategy, iCrossing
Case study - tech forum engagement
Problem: unhappy customers were online talking about how unhappy they were
solution: worked with client to identify, monitor, and develop an engagement strategy. The goal was to be helpful and not market.
outcome: negative sentiment is decreasing. Alot of traffic was going to the postings where they were reaching out. Program learnings and
recommendations are being shared with the entire organization. Management, Development, and their knowledge base are all being updated.
Measuring and reporting
- Monitoring metrics
- tonality of user postings
- site traffic for the forum sites
- Engagement metrics
- direct metrics: number of company postings, number of converstaions engaged
- indirect: page views of postings, number of links posted to clients website, amount of traffic from the links
track links posted and clicks received by site - they are using Omniture data and combining (by hand) a count of which links are creating
Categorize and analyze discussion topics
- non technical (they dont engage) or technical topics(they do engage)
- decide when and where not to engage
Example: forum users were complaining about confusion about being over charged for customer support
- one path took you to a free self help, another took you to a fee based.
- Solution: they created a landing page that explained the two
- there is no one killer metric
- track anything possible
- its not just about the numbers
- its all relative (focus on benchmarking)
- view monitoring social media as a social intelligence program involving the worlds largest focus group
Live blogged by Dave Rohrer