More talk about social media ROI and measurement? You know it.

This time Mitch Joel of “Six Pixels of Separation” wrote a post about measurements and metrics. As much as I like Mitch (his articles are great and I highly recommend his blog), I really disagreed with his take on social media ROI.

credit

He states that…

“Thinking about business objectives instead of ROI makes the whole strategy that much workable and doable. In the end, instead of looking for the ROI in Social Media, maybe the smarter thing to do is to focus on what the business objectives are, and then figure out if Social Media is an effective means (from both a strategic and budgetary point of view) to help you and your business accomplish your goals.” [Source]

He uses two phrases that stand out: “business objectives” and “effective means”.

Here’s where I disagree with Mitch. I don’t agree that business should be run by “effective means”. It should be run based on decisions that lead to the most “efficient means”.

What’s the difference? Well, imagine my business objective is to build awareness about my product. I could do a number of things.

Picking up the phone and going through the phonebook would be extremely effective. Nothing beats a personal connection. But is it the most effective? Most likely, no.

Credit

Without being able to assign an ROI to marketing efforts, managers can fall into the trap of allocating resources to measures that do not efficiently achieve business objectives.

The Problem:

If we agree that determining the ROI of social media is a necessary evil to making intelligent marketing decisions, then how do we measure it? Therein lies the problem.

Currently people have measure social media loosely, if at all. We’ve been using metrics as indicators of succes.

I hate to use the Dell example because I think they’re doing a great job with social media, but the response they gave about ROI (“Binhammer responded that ROI was nothing more than an accounting term and probably has little to no place when it comes to measuring the success of any Social Media marketing initiative”) seems unsustainable.

Dell is also well known for there announcement, stating that they “surpassed $3 million in sales via links from one of its Twitter accounts”. It’s important to note that this figure is slightly disingenuous. You don’t know how much they spent on social media. How much is spent to have a team dedicated to social media? What is the profit?

A second form of anecdotal “ROI” that doesn’t actually tell the whole story is when managers say “Company x is making more money and they’re using social media, therefore social media must be making money”.

It seems logical at first, but when you delve into the statement you realize the hole. How are you to assume that it wasn’t consumer interest, a by-product of more efficient marketing, or just being lucky? Without examples or data, you can’t prove or disprove that social media helped the business be more profitable.

Credit

Without data and without being able to clearly track sales, social media can’t be proven to work. It requires a leap of faith, which businesses should not be forced to make. Efficiency is dictated by testing, by data, and by understanding.

The Marketing Mix Model:

One of the most important tools a marketing manager will have at his disposal is a marketing mix model. The model, an equation, helps marketers “optimize promotional tactics with respect to sales revenue or profit.”

Credit

Social media marketing doesn’t fall within the model. Because of its dynamic and fluid nature, social media becomes an outlier. But, without being able to measure social media’s effectiveness, we won’t be able to assign it a value in the marketing mix model.

Herein lies our goal. We must identify a way to measure and assign social media efforts a value in terms of sales revenue or profit.

Finding a Solution:

There’s no point in complaining about social media measurements without offering some kind of solution. The first step is outlining exactly what we need.

To accurately measure the effectiveness of social media, we must be able to:

• Identify users that have engaged through social media
• Assign the appropriate value to social media interaction in the click-stream
• Determine if the user converted to a sale

In my mind the best way to achieve these measurements is through a cookie.

A cookie (also tracking cookie, browser cookie, and HTTP cookie) is a small piece of text stored on a user’s computer by a web browser. A cookie consists of one or more name-value pairs containing bits of information such as user preferences, shopping cart contents, the identifier for a server-based session, or other data used by websites. Source

Why use a cookie to measure social media? With a cookie we’re able to determine which social media profiles the user interacted with and when they finally converted.

credit

With this kind of data, marketers would better be able to understand and optimize all online marketing campaigns.

There exists two major problems with this potential solution: privacy and the social media sites themselves.

Addressing Privacy Concerns:

Most people want some kind of privacy on the web. I agree, I tend to clear cookies when I close my browser. I’m the person that Comscore and Quantcast hate.

credit

Nonetheless, being tracked online has become second nature. For example, social sites like Facebook and Twitter already cookie users to track engagement and other site metrics for themselves. Most ad networks use some kind of cookie to retarget users and show ads across the network.

Allowing individual brands to create and deploy cookies that almost “piggy-back” on the cookies already being used shouldn’t lead to a huge outcry from privacy supporters. The information is already being gathered, the question is how can marketers get it, and how can it spread across multiple platforms?

Concerns of Social Media Sites:
Another obstacle to overcome is the mere fact that most social media sites would never want to freely share this information. Data is their money and collecting it is extremely important.

It’s interesting that social media sites, the herald of change on the web, have maintained an old model for monetizing their sites.

Instead of using the data to target advertising, why not allow marketers to optimize their social media campaigns?

I know a number of brands that would pay for a “brand Facebook page” that allows for useful metrics. Throw in the ability to cookie users and an integration with Google analytics, and I wouldn’t even have to think about the decision to pay for a fan page.

Twitter has a unique opportunity too. Cookie users as they click links and give me click through rates. These kind of metrics and measurements help social media marketers allocate a value to social sites.

By not working together you hold back social media from growing and becoming a data driven marketing path. Which leads us to the next section.

Isolated data isn’t enough, it must be comprehensive. If we’re to create cookies that work across platforms, we need some kind of standard.

Proposing a New Standard:

So why do we need a standard? Well it’s simple really.

No successful social media campaign covers just a single social media site. Social media marketing thrives when multiple platforms are used in harmony.

It would do us little good if we could only track conversions through Facebook, but not Twitter. Also, it’s no help if we’re tracking Facebook and Twitter independently.

Marketers should be able to allocate each site’s involvement in the sales process. Imagine that a user finds a Facebook page, then follows the brand’s Twitter account. Next thing you know they buy a product that is Tweeted out.

Should only Twitter get credit for the sale? It was Facebook that originally introduced that user to the brand.

With a cookie that is standard across platforms, similar to OpenID, marketers would be able to properly identify the value of each individual platform.

A Short Term Fix:

The standard cross-platform cookie might be months, if not years, away. So what do we do in the mean time? How do you measure social media to the best of our ability?

When it comes to social media measuring right now, it breaks down into three sections.

• The first is measuring outreach effects.
• The second is measuring profiles reach.
• The third, and I think most difficult, is measuring the effect of status updates.

Measuring Outreach Effects and Profile Reach:
Before we can explore how to best measure outreach and social media profiles, there are a few things you need to have in place ahead of time.

The first is to be able to create virtual pages or 301 redirects. If you’re on WordPress, you can use the Redirection plugin. The idea is to create a “vanity” URL that redirects to a “tracking” URL.

Secondly, you need to have Google Analytics installed on your site for this technique.

So how does this technique work? Well we can’t actually track the user through the entire social media funnel, instead we can track them as they jump from one profile to our site.

This gives us an idea of the final step of the process and if people are converting from the social media site.

The best way to explain the practices is to use an example.

Imagine I have a Facebook page. On the page I have a landing page that highlights my latest product, a new Dell Monitor (The Dell Ultrasharp 3008WFP to be exact).

Instead of posting on Facebook the URL [http://accessories.us.dell.com/sna/products/Monitors/productdetail.aspx?c=us&l=en&s=bsd&cs=04&sku=223-4890] that is the exact product page, instead we create a vanity URL. The new URL we use can be http://promotions.us.dell.com/fb/dell-3008wfp. This vanity URL will redirect to the product page URL, but with a few additions.

The additions come from the URL builder from Google. Paid Search Marketers have been using this tool forever, to tag and identify campaigns. Social media marketers can do the same.

In this instance, because we’re posting the link on Facebook we name the Campaign Source – Facebook, the medium is our Landing Page and the Campaign name is Social Media.

So let’s walk through the user workflow. The user comes to Dell’s Facebook page. Sees the landing page and clicks on the link, http://promote.us.dell.com/fb/dell-3008wfp.

This link redirects to:

http://accessories.us.dell.com/sna/products/Monitors/productdetail.aspx?c=us&l=en&s=bsd&cs=04&sku=223-4890&utm_source=facebook&utm_medium=landing-page&utm_campaign=social-media.

Google analytics tracks this as a Facebook sources referral with a campaign assigned as “social-media”.

It becomes powerful because as people share the vanity URL, we’re still able to attribute any conversions to the original promotion and effort.

The same can be done for blogger outreach. When recommending a page bloggers link to, the link could be a vanity URL which redirects to a UTM encoded page. This allows you to track the overall effectiveness of your blogger outreach campaign, as well as conversions from these links.

Although not perfect it gives us a basic idea of outreach and profile reach. Mixed with advance use of goal funnels and event tracking in Google analytics, you can get a better understanding of the ROI of the social media campaign.

Measuring Effects of Status Updates:
I stated before that tracking the effectiveness of status updates is more difficult. The reason for that is because it’s so open.

A link shared on Twitter can easily spread to Tumblr. Also, the life of each individual link is so short that it’s difficult to gather a comprehensive amount of data.

We can’t perfectly measure the effect of status updates, but with a few tricks we can attempt to gauge a campaign’s success.

The previously mentioned technique of “vanity” URLs could work for status updates, just as it does for profiles and outreach campaign. Awe.sm is an application which lets you use your own domain to create shortened URLs and automatically add the UTM code.

Another trick is to use coupons to gauge conversions. Sharing Twitter or Facebook specific coupons via status updates allow you to track the number of coupons used and measure the status update’s reach. The coupon needs to be published multiple times to take into account variables such as time and overall Twitter noise.

Finally, another quick way to gauge overall effectiveness of status updates is to create a specific landing page for the status update. Let’s use the Dell monitor again as an example.

The goal is to test the effectiveness of our reputation and clout. The first step is to create a landing page separate from the current product page. Next, we tweet out a link to the page with a message like “Just sat next to my new dell monitor, it’s nice and bright! [link]”. Now we track click-throughs to the page, but also over-all traffic to the page.

credit

More than likely, the number of click-throughs will be less than the the traffic the page gets. The reason is because people re-shorten links, or use different tools, like ow.ly to retweet the link. By creating a separate landing page for this test we’re able to collect the total data.

Also, we test our “clout” by not pushing the user to actually buy the product in the tweet. The only thing influencing the consumer to buy the monitor is that the Dell account stated it was a bright monitor.

We’re looking to track impulse buys that are based solely on the fact that the product was tweeted by the account. The number of conversions will be extremely low. This shouldn’t be disheartening since the goal of status updates isn’t to create large conversions, but to slowly convince a consumer to buy product over time.

As you can see tracking status updates can be a difficult task. Instead of attempting to continually measure it, we’re forced to spot-check the success of campaigns to gauge effectiveness.

Conclusion

It’s clear that social media measurements need to mature. The problems are numerous and obstacles great. The question is, will we pass this phase where marketers are saying “Social media works, trust me…” and enter a paradigm where social media is measured and optimized thoroughly.

credit

Until social media sites work together to offer full tracking and data, we’re forced to using incomplete metrics as indicators of success. We’re using snapshots to track entire campaigns.

Social media metrics must mature to a point where smart social media marketers can clearly track ROI so they can make informed data driven decisions. Most large brands are risk-averse, and without clear data, social media becomes a huge business risk.

For social media to continue moving forward its effectiveness to build a business profit needs to be proven. Otherwise, why would I invest in social media when I know a TV ad will net me a profit?