Community Management, Social Media Measurement, Campaign Testing and Optimization, Conferences and Events

The 4 Marketing Metrics Mandates

There are four mandates that companies are adopting when they are examining and analyzing the success of their online marketing and social media programs. I recently gave a presentation to the Measured Marketing Roundtable at Techpoint, Indiana’s technology and economic development association, that outlines these new metric must-dos.

It’s Not a Lie if Everyone Agrees to It

For a very long time, measuring marketing was a needle in a haystack approach, whereby we all agreed to lie to each other about data and extrapolations of it. In Charlotte, North Carolina there are something like 1.7 million households with a television. Yet, the behavior of just 641 households are used to calculate the Nielsen ratings that determine the prices for advertising market-wide. This is clearly less than ideal mathematics, but it’s all we had at our disposal, historically.

Then, digital marketing came along. Among its many benefits was the ability to measure the behavior of every impression, visit, click and so forth. No extrapolation necessary.

However, as is so often the case with online marketing, we overdid it. Because so much data was at our fingertips, we tried to measure everything – because we could in fact do so. This resulted in a 10-year period (that we are just now emerging from) where the glut of data made it paradoxically more difficult (not less) to know what the hell was going on with our online marketing results.

Today, marketers are smarter and more discerning about their data needs and desires. Now, there are 4 Metrics Mandates:

1. Behavior-Based instead of Aggregated Data

Katie Paine coined the bon mot “HITS means How Idiots Track Success” in 1997, and unfortunately it’s still relevant today. Even though most website owners aren’t using “hits” as a success metric any longer, we’re still falling into the trap of counting aggregations.

Twitter followers and Facebook fans and the most egregious example. The number of Twitter followers you have means practically nothing. What you can do with them matters a lot.

Thus, the best metrics are those that measure behavior. Website leads from people that clicked on your tweet. Sales from Facebook fans. 5-star ratings on Yelp.com from blog readers. That’s what we need to measure.

2. Integrated Metrics instead of Siloed Numbers

We have been able to track so many things in digital marketing that we created a series of data lakes for ourselves. SEO stats. PPC stats. Website stats. Banner ad stats. Email marketing stats. But which are the most important indices?

Today, the mandate is for integrated metrics that combine data from multiple sources to provide a seamless, instructive dashboard. Companies like PostRank (with its analytics product), and Klout (with its online influence measure) and leading this integrated metrics parade.

3. Optimizable Metrics instead of Happenstance

4 marketing metrics mandates The 4 Marketing Metrics MandatesPerhaps the most powerful aspect of online marketing has been the ability to precisely test and optimize results. Given enough time, I can determine for you the best possible day, time, subject line, from line, layout, and link placement for your email newsletter. I can do the exact same thing for your paid search campaign, banner ad campaign, and your website home page. Online marketing and scientific method go together like Big Bird and Snuffleupagus.

But in the social media realm, we haven’t quite gotten there yet. I can tell you at what times of day more people click links on Twitter, and I can tell you how many people commented on a Facebook status update. But, I can’t test  (at least not easily) hypotheses about headlines, link placement, time of day, etc. Social media results are still too much about trial and error, not optimization. Let’s hope that changes – and fast. Maybe Twitter’s new analytics feature will help?

4. Benchmarkable Metrics instead of No Frame of Reference

We have an insatiable desire for context. A search of Google for “good email open rate” fins 74,600 matching web pages! Why? Because most companies know their open rate, but they do not know whether it’s any good in comparison to other companies.

That’s one of the reasons Twitter followers and Facebook fans are such popular metrics; because they allow for on-the-spot manhood measuring between your company and your competitors. Even though Daily Story Feedback, Active Fans and several other metrics are better measures of Facebook success, those stats get very little mention because you have to log-in to see them, so companies can’t one-click measuring stick.

Postrank is now allowing you to compare your blog’s performance to that of up to four other sites. And “Share of Voice” is becoming a much more popular calculation (measuring the neutral + positive social mentions of your brand vs. competitors). Let’s hope more social metrics are developed or published that give us the frame of reference we crave.

(Note: to see an example Share of Voice report, check out the example graphic we commissioned for The NOW Revolution from Chris Sietsema at Teach to Fish Digital)

The 4 Characteristics of Marketing Metrics: Behavior-Based, Integrated, Optimizable, Comparable. Do you agree?

(image from Shutterstock, a Convince & Convert sponsor)

  • http://www.maples.com Bryan Howland

    I couldn’t agree more with “the number of Twitter followers you have means practically nothing.” While in a perfect world you’d have a ton of followers who are all quality followers, a majority of followers tend to be worthless. I do think measuring Facebook fans is worthwhile just as a high level way to track the effectiveness of your efforts – and more importantly to be able to present an easy to understand stat to the C-level. Trying to explain Active Fans to people who have no concept of Facebook (yes they are still out there) can take up more time than needed. Also, If I know I’m adding an average of 150 new fans a day through ads, posting content, etc. then I have a baseline to measure if things are starting to fall off or pick up.

  • Anonymous

    Jay, I love the dive into measurement here. I would add a couple cautions, traps that are easy to fall into when following these metric mandates.

    First, understand the role of your various activities. If you are measuring late-stage behavior (you mention leads, sales and reviews), multiple touchpoints are important. Your early stage activity won’t drive a lot of late stage behavior, but it is critical to overall success.

    Second, always step back and look at the big picture. A laser focus on any metric or handful of metrics, regardless of relevance, creates tunnel vision. Is the way you are managing towards ‘Active Fans’ moving your business forward, or have you muddied the measurement, using controversy or incentives to drive activity? Is remarketing training your audience to abandon shopping carts for the discounts that follow?

    – Eric
    @wittlake

    • http://www.convinceandconvert.com jaybaer

      Excellent suggestions Eric. Indeed, I can’t encapsulate all of my metrics recommendations in a single post. Hell, I wrote 50 pages on it in The Now Revolution. The big picture is critical. You can’t be a slave to your metrics.

      • Anonymous

        Thanks Jay. I just may have to get the book!

  • http://bit.ly/HowToRun_TWITTER_CONTEST Don Power

    Here’s an interesting anecdote to support “the number of Twitter followers you have means practically nothing.” comment from this post.

    I just noticed that a local computer tech company unfollowed me. This is a tech support, hardware support kind of business that recently started selling “Twitter followers” for their clients interested in social media.

    That company has 11,000 followers in a town with 80,000 people. I have 1900 followers.

    The tech company “selling social media packages” has a Klout score of 18. I have a Klout score of 59. In this case, I think this measurement is very instructive and telling for any prospective client who is interested in getting help moving into into the social media game. In other words, you can talk and big game and look like you have a big game but you don’t have any game at all.

    ‘Be careful what you measure ’cause you just might get it’…

    Oh, for more conversation on a similar topic, I invite your readers to read “Why Social Media ROI Doesn’t Matter…Yet” from Sprout Social’s INSIGHTS blog.

    Thanks!

    Don Power
    Sprout Social Insights

    • http://www.convinceandconvert.com jaybaer

      Great stuff Don. Thanks!

  • http://www.facebook.com/nicolesorochan Nicole Sorochan

    Hi Jay. I love your segmentation of how to measure social media success and the importance of interaction over fans/followers. A few frustrations I have is working with a smaller unknown brand from ground zero – which means that they have to work extra hard to attract their audience and then work to engage them. (Obviously the company wants to see a decent audience to prove the campaign is worth the cost. 100 engaged fans does not make a successful campaign). Engagement measurements do not show much variance on a week over week basis. I would love for you to write an article based on a case study of a smaller brand and their “t life cycle” of a new social media campaign. How long should it take (and not the outlier viral campaign that takes off instantly). What types of metrics should we start the life cycle with? How should the metrics mature as the campaign grows overtime? What are some tell tale signs that the strategy is not working and should pivot?

    And a bit unrelated- but when using tools like socialmention.com for “share of voice” how do you get an accurate reading if a company’s brand name is an actual word and not a unique name/spelling?

    • http://www.convinceandconvert.com jaybaer

      Thanks for the excellent comment Nichole. Great idea for a post. I’ll see what I can do. As to the question about social mentions and common words, it’s a very tough row to hoe. Probably you’ll get a ton of false positives, and will need to do manual sorting of results to focus in on the ones that are actually about you. Paid listening software like Radian6, Alterian, etc. can do some of that sifting automatically for you.

  • http://myburleson.com Burlesonbrad

    Just outta curiosity, how’s the wibiya bar working for you?

  • http://myburleson.com Burlesonbrad

    Just outta curiosity, how’s the wibiya bar working for you?

    • http://www.convinceandconvert.com jaybaer

      I’m liking it so far. Easy to use.

  • http://dannybrown.me Danny Brown

    Interesting post, sir, and great information to take away and chew on.

    Just curious – how well do you feel automated tools can truly tell sentiment and emotion, compared to actually asking someone how they feel? Where does automation filter human emotion and all its nuances?

    • http://www.convinceandconvert.com jaybaer

      Automated sentiment will always be difficult. It has a hard time with sarcasm, for example. Even if you have automated sentiment capabilities through your listening software, it’s imperative that you manually scan and correct where necessary.

  • http://www.mitash.com/blog/ Raj – SEO Australia

    All Genuine Metrics. You are feeling sorry for those who measures their metrics from HITS but what about companies that offers some prizes/rewards when they reach significant amount of ‘Likes’? Where do you think these companies are actually leading to?

  • http://twitter.com/garious1 Garious

    I think I have to agree with you that numbers don’t matter unless you can do something with it. What’s the use of having thousands of followers who are just there for the popularity contest? I’d rather have a hundred followers who are engaged with my brand. It’s really hard to tell who are really loyal to your brand when it comes to social media marketing. I will then have to agree with you on all characteristics you’ve explained here.

  • http://twitter.com/Collectual Collective Intellect

    Jay,

    Thanks again for sharing your insights. I totally agree with the points you’ve made, especially around “numbers don’t matter unless you can do something with it.” I’m wondering if we may need to include an engagement metric – the volume/impact when a client reaches out to a customer. What’s the end result of a customer engagement, including where you interacted, what was said, how often, at one point in the lifecycle is the customer? I’m just kind of throwing out ideas here but it seems that to have a more complete picture social media metrics of we also need to include the data points of these types of engagements.

    Again, thanks for sharing!

    • http://www.convinceandconvert.com jaybaer

      The holy grail is being able to ID the social interaction and tie it back to revenue action. The challenge is that we don’t yet have a unified view of the customer/prospect. Rapleaf and Flowtown were on the right track until Facebook shut it down. The challenge is, if you interact with a customer via Twitter, you don’t really have a reliable way of tracking an eventual purchase back to that person, especially if they use a different email address. Some day. Some day.

  • letstalkandchat

    I just found a great company that builds websites for info products. To keep your costs low, they’ll mentor you on how to create your site, design a marketing funnel (one of the guys works in Hollywood and makes really slick videos), and the other guy programmed Myspace. If you’re looking to have professional web design for your small business and not waste any time or money then check their site out. Check them out: http://www.mikelmurphy.com/easy-info-product-site-system/