A Webinar that I sat in on last week points to some answers.  Is Your Social Media Program A Boom or A Bust? How to Use Analytics to Find Out What’s Working and What’s Not, by Meteor Solutions, put on in cooperation with Word of Mouth Marketing Association (womma.org), starts by stating that in the recent past online marketing was all about buying ads, SEO and paid search, i.e. push marketing.  What’s changed is that we’ve become more sophisticated by creating killer content, i.e. pull marketing.

In order to analyze the effects of online marketing, we typically use tracking and optimization software.  Taking a look at a few of these products,  Google Analytics provides traffic source, but it also leaves out some things that I would like to see, such as what links people are clicking on*.  For my blog I use WordPress Stats, which does show that key piece of information.  There is also StatCounter.  All are free.

*Update – there is a way to see outbound clicks afterall. You have to tag the link with a bit of JavaScript:

For example, to log every click on a particular link to www.example.com as a pageview for “/outgoing/example_com” you would add the following attribute to the link’s tag:

<a href=”http://www.example.com” onClick=”javascript: pageTracker._trackPageview(‘/outgoing/example.com’);”>

Omniture offers two pay-for products, SiteCatalyst and Discover.  I am not deeply familiar with either product, but in my digging about SiteCatalyst I came across a site, Actionable-Analytics.com, which has an article that takes a look at the difference between Google Analytics and SiteCatalyst.  Taken from this article, the primary benefits of SiteCatalyst are:

• Real-time data (others are very near real-time)
• Importing external data, such as postal codes
• More custom variables
• Creating custom paths and funnels

Omniture’s Discover seems to offer some expanded features on SiteCatalyst, allowing for drilling-down to individual visitor level, comparing visitor behavior side-by-side, increased segmentation and site navigation views.

I’d also like to mention Yahoo! Web Analytics, which is free only for Yahoo! advertisers.  I don’t use this, but perhaps someone who does can share what benefits it offers that Google Analytics does not.

I came across another product, NetInsight, which is interesting. Check out this article:  12 Cool Things you can do with Unica’s Affinium NetInsight.  I could not find out how much their license fee is, but I did find this comparison of the top 10 web analytics tools if you’d like to delve into pay-for analytics packages further.

Getting back to the point of this post, Meteor Solutions goes on to say that measuring earned referrals is important, that is, when someone shares something with their friends about your company.  The top 5 sources of earned referrals are: email/IM, blogs, video, forums/message boards and aggregators/bookmarking.  Social networks comes in 6th place (Facebook, Twitter, etc.).  15-20% of unique visitors to a website are coming from earned referrals, and offer an average of 1.5-4x conversion lift.

Some tools that can be used to measure earned social media referrals:

Facebook, Twitter, Blogs, eNewsletters, et al

• Meteor Solutions – Their program uses tracking scripts to track the “pass-along” of content from one person to another via links on web pages, bookmarks, email messages, instant messenger and mobile devices.   They then provide a report on earned referrals, which can then be converted into dollars (eCPA aka effective cost per action).
• Bitly Click Trends – tells you how many clicks you have received on your bitly URL shortner, plus how many clicks for all bitly links to the long URL. For example, my blog post How to Measure Soft ROI, shared on Twitter, received 10 clicks from my bitly link bit.ly/4AsFZL, but that was then shared with others who created their own links and it was clicked 17 more times (it says, “10 out of 27″ – you have to do the math).
• As mentioned in my previous blog post, Monitoring Chatter on the Web, you can use a product like SM2 to measure whether you are receiving more mentions of brands/products/services.  They have a limited, free monitoring service and their pay-for service allows you to evaluate positive/negative sentiment of those mentions.
• *Update, from Techcrunch - Viralheat, a pay-for social product monitors YouTube, Hulu, Vimeo and other video sites, blogs, websites and Twitter by keyword.  You can filter content by location, importing analytics to Excel, and receive alerts if the “profile” (an individual’s or company’s name) you are tracking is seeing a increase in mentions or a spike in traffic.
• *Update, from Techcrunch - Startup KnowEm is helping companies manage their presence on social media.  On KnowEm’s site, brand owners can instantly check the availability of their branded usernames and keywords on more than 330 social media networks (for free). For a flat fee of $349, KnowEm will create and insert relevant information into profiles on the top 150 social media websites. For $99, KnowEm will create the profiles for the brands but won’t fill out the profiles. Brands can also pay $49 per month to protect a brand or username on emerging social media sites. So if any social media site launches, KnowEm will stake out the brand’s territory automatically.
• Wassup for WordPress – track your visitors in real time and understand what your visitors do when they surf your blog.  It has a “current visitors online” view and a “visitors details” view.
• From Bob Cortez, Tracking Online Referrals, If you have your own domain you can set up a default email address so that “anything”@yourdomain.com gets routed to you instead of bounced.  You can use that to your advantage…to track referrals for subscriptions. He says, “When I recommend that someone subscribes to [my friend John's Markethink Digest]  I tell them to mailto: bob.cortez@dakine.demon.nl?subject=Markethink.  Ask your referral partners to do the same thing.”
• From Avinesh Kaushik, Occam’s Razor:
  • Use redirects (vanity url’s) in advertising.  For example, a magazine ad with the call to action “Visit www.usequickbooks.com” redirects to www.quickbooks.intuit.com/tracking_code=newsweek_dec_2009.
  • Use unique redeemable coupons/offer codes.  For example,  ads for 1800flowers.com in a taxi cab offered the coupon code “taxi” to get $5 off a $50 order.  Another great strategy is to use the same coupon code [across] channels…so you are providing people a choice in terms of channel preference (e.g. phone vs. website) but since the code is the same you can track it.
  • Use online surveys / market research (I couldn’t agree more as the company I work with, C.A. Walker, is a full-service research firm!).  He says that even more important than website analytics, is source tracking. He goes on to say that the most important questions you can ask a customer are (I have revised these just a bit to be offline as well as online):
    1. What is the purpose of your visit today?
    2. Were you able to complete your task today? (or, Did you find what you were looking for?)
    3. If no, why not?
    4. What is the likelihood that as a result of a visit that you’ll make a purchase?
  • Correlate traffic patterns with offline ad times / patterns.  For example, what offline media stimulus – such as an infomercial – causes people to run to the Internet to search for your site?  Once you have the magazine / tv / catalog / postal mailings / radio / billboard plan for your company then do the correlations with your website traffic and see what the impact is.  I’d like to also add my own comment here – What other searches are they running beside your brand?  I know when I see something interesting on TV I always search for “product” reviews.  9 times out of 10 the results stop my interest.  Is that happening to you?

Twitter Only

• Twitturly – tracks URLs in the Twitterverse and provides a real-time view of what people are talking about on Twitter. Each time someone tweets a URL to their followers on Twitter, Twitturly takes note of it and applies it as a vote for that URL. The more votes a URL has in the last 24 hours, the higher it ranks on Twitturly’s Top100.  My note – a good use for Twitturly is to search on a keyword to see the top shared links on that topic and connect with Twitterers in your business space.
• Klout – performs a search of Twitter to calculate your reach and to analyze who specifically you influence and who influences you.
• TweetEffect – tells you which of your Twitter updates made people follow or leave you.
• CoTweet – allows up to six Twitter accounts to respond to customer tweets through a single login.

Digg Only

• Digg Spy – lets you “spy” on what articles Digg users are submitting and voting up.  More on Digg Spy here: Digg Spy 2.0 Released.  While not  helpful in tracking earned referrals, I include it because it’s interesting to click on the different views at top (e.g. Swarm, Arc) to see what people are talking about online.  It may give you some insight in what you could write about that might catch people’s attention.

And finally, some ways to increase earned social media referrals from actual case studies (my boss is going to love this!) of the best practices of the world’s top brands, such as Starbuck’s, Dell and Toyota.  From ENGAGEMENTdb:

• Launch a blog, forum or Wiki where customers can submit, comment on and vote for their favorite ideas to help your brand, with someone from every department at the company acting as a liaison.  Encourage customer engagement by implementing a reward point system for different activities, such as maintaining a blog, responding to forum questions, or adding to a Wiki page.
• Create an official Facebook brand page, even taking over ownership of the largest customer-created communities.
• Use Twitter as an “in the moment” channel to deliver timely customer support and spread word about the latest breaking news and contests.
• Be cautious about allowing employees to engage with customers via social channels, to make sure there is consistency in approach, and be ready to shut down anything not sanctioned.
• If you are going to engage customers online, you have to have a plan, make sure that resources are available, and be in it for the long haul.
• Start a YouTube channel to showcase video content.  Pull video content from around the organization, such as training videos.
• Create an “activity stream” of company-generated content such as videos, community and industry news.  That information can be pushed into blogs, tweets, video, photos, etc.
• Work closely with outside blogs not affiliated with your brand, by providing access, information and support.  This can be especially effective if you have no plans to launch a blog of your own.
• Monitor Tweets that mention your brand, to respond from respective areas of expertise within your company using CoTweet to manage a “brandlistens” account (see above), or you may choose to have people respond from individual profiles.  Same goes for Facebook, by emailing around comments that need response.
• Treat social media engagement as part of the job, just like phone and email.
• Lastly, when choosing to engage in a channel, be fully engaged.  If you have resource constraints, it is better to be consistent and participate in fewer channels than to spread yourself too thin.

I hope you’ve enjoyed this post and learned a few new things that will improve your social media ROI.

One good way of arriving at social media metrics is postulated by Heidi Sullivan of Cision Blog: “There are two basic types of ways to measure your impact on the Web: passive and active.” Furthermore, she attempts to break it down into the five senses: sight (where are you seen), sound (who’s talking about you), scent (do they stick around to “smell what you’re cooking”), taste (can they taste what you’re offering through search engines) and touch (are they taking the next step to touch others with your information).

Going back to the passive vs. active concept, here are metrics she suggests as well as from other sources (numbers refer to sources at bottom):

Passive

• Number of unique visitors to your blog.1
• Monthly site views aka page views.1 – Definition: Request to load a single HTML page.  Page views are only important to the degree they play a part in a site’s revenue model. If a site earns much of its revenue from advertising, then page views are important because of their contribution to ad inventory. If a site only earns revenue on sales, then page views are not a key statistic.
• Number of page views per visit.6 – Try landing people on an engaging page that funnels them through a messaging sequence that builds your brand and escalates sales. Measure the results and optimize the path that gets people to the end of the sequence. Also, measure the lift in total page views of your site’s high-value product pages.
• Monthly unique sessions.1 – Definition: In tabulating statistics for Web site usage, a user session (sometime referred to as a visit) is the presence of a user with a specific IP  address who has not visited the site recently (typically, anytime within the past 30 minutes). The number of user sessions per day is one measure of how much traffic a Web site has. A user who visits a site at noon and then again at 3:30 pm would count as two user visits.
• Number of return visits.6
• Search trends.7

Active

• Number of influencers (track who they are for targeting) that share your content through inbound links, citations, tweets, viral videos or other forwarding of information that result in transactions.1,4,6 – put a call to action in your campaign that will lift this metric.
• Number of company/brand mentions in blogs, on Facebook, MySpace or Twitter, as gauged by a social media monitoring solution, especially when supplemented with sophisticated sentiment analysis that discerns positive brand mentions from the negative. 1,7
• Number of people clicking on your images, watching your videos, using your widgets, and spending significant amounts of time on your site.1 – by tagging videos and applications, companies can track each occurrence of sharing and estimates frequency of exposure to particular media.6
• Number of and which keywords/other sites referring visitors to your content.1
• How much money social media has saved or created for your brand: How many issues that you solved, questions that you answered, leads that you created, products that you sold, call volume that you decreased, etc., through social media engagement versus traditional resources.2
• Time spent across an entire campaign.3
• Click rate [on an ad].3 – not the best metric to use.  People can accidentally click and then drop off, or they can click and see it’s not for them and drop off.
• Interaction rates [with an ad].3 – may be influenced by the creative and not a true representation of their interest.
• Conversion rates – doesn’t take into account return conversions or other offline results, e.g. visit to a store for further research.
• Time spent throughout cross channel engagement with the brand – forecast the number of touches a customer can have with a brand and the resultant amount of time spent with the brand, e.g. total number of exposures, visits you’ll drive to the site, increase in total mentions that can be measured with social media monitoring tools.3
  • Once you understand the total touches, you can either measure directly or assume through general patterns what the time spent is across each of these vehicles, and then create a cumulative time spent for the campaign. When you break down the individual media vehicles, time spent is actually quite easy to report on.3
• Lifetime value of customers based on their purchasing activity, which could be higher depending on their level of social media influence, e.g. if a customer is worth $9 dollars alone, but that person has 500 Facebook friends, and is able to drive even 1% of them (5) to make a purchase, that individual’s value could be as high as $54 dollars.4
• Number of people who you capture data from – Eyeblaster has found that the ability to capture data within a banner is nearly eight times more effective at addressing conversions than click-through rates (CTR), revealing that consumers are more likely to fill in a banner than click on it.5
• Number of people who opt into your e-newsletter/email updates (e-newsletter metrics to be discussed under separate, future posting).6
  
• Number of people who download materials from your site.6
• Number of phone calls you receive and source.6
• Number of people that share comments and opinions.6
• The impact individuals exert on wider audiences – a combination of data obtained from buzz-tracking with data from surveying (see below), to determine the probability of viewing buzz by taking into account the volume of relevant posts and amount of time spent by panelists on a website.7

And there are also metrics that can only be obtained via surveying, which is what the company work with, C.A. Walker, does:

• Awareness.
  
• Interest.
• Desire.
• Lifts in brand awareness and attitudes attributable to a social media campaign.
• Various actions taken throughout engagement with the brand.
• Usage – may be of a particular website or of a product.
  
• Consideration – may also be inferred from measuring the things that drive people into your (or your client’s) stores.  Actions like store locator queries and rep searches should always be measured and are clear indicators that a future purchase may be imminent.6
• Intent to purchase – may also be inferred from Time spent throughout engagement with the brand. If consumers spend more time than the average, they must be interested in your product or service — and if they are interested, that is a measure of intent.3

If I’ve missed any applicable metrics, please let me know so I can grow this list.

Sources:

1Cision Blog: Five Senses of Online Measurement

2Mashable: Twitter Brand Best Practices

3MediaPost: Time Spent Is The Right Metric To Measure Engagement

4MediaPost: Marketers Search For Social Media Metric

5Eyeblaster: Conversions Remain Important Metric

6ClickZ: 11 Things to Measure Besides Clicks and Conversions

7MediaPost: Media Metrics: Let’s Get This Party Started

I came across a company today, Techrigy, with a product called TechrigySM2 that offers this service.  They offer free to monitor five search terms and up to 1,000 results…not per term.  If you want more than that, you’ll pay $600+/month.  I’m assuming they use a sliding scale depending on the size of the company that wants it.

They monitor the following:

• Blogs
• Wikis (Such as Wikipedia)
• Message boards/Forums
• Video/photo sharing website (such as Flickr and YouTube)
• Mainstream media blogs (such as the New York Times blogs)
• Microblogs (such as Twitter, Pownce, Friendfeed, and Plurk)
• Social networks (such as Facebook and MySpace)

And, of course, they offer a whole bunch of additional options such as “Sentiment Analysis,” “Discussion Clustering,” and it all ties into their new parent company’s platform that includes Campaign Management, Web Content Management, Email Marketing and Marketing Analytics tools.

I’m checking it out myself to see what I can get out of it…you should too. One thing that I’ve found it’s great for is to find other bloggers and Twitters who are discussing marketing ROI, so I can follow them.  It’s just too bad they don’t offer a more limited package for small businesses at around $50/month.