Monthly Archives September 2014

Lessons Learned in Two Months of Social Media Engagement

Before you get the wrong impression about my social media engagement experience, I’ve been using social media for more than two months.  My first personal status update on Facebook was on 26 October 2006.  My first tweet was on 18 April 2008.  And I got an early invitation to Google+, so I was on it before the public release: 30 June 2011 (the public launch was in September).  We won’t talk about MySpace, but I was on that for a bit before Facebook came along.

In contrast, Ardea Coaching as a brand page has only two months of data on these “big three” social sites.  That gives me some insight into what new businesses confront when they’re just getting started.

Caveat: Is Social Media Engagement Right For You?

A frequent refrain when I’m training someone on social media marketing is that success depends on what you’re looking to achieve.  “What’s your goal?” I often ask.  This blog post will assume that audience interaction assists those goals.  (Maybe your business goals focus on funneling people from social media sites to visit your website for more information.  Perhaps you don’t care if they interact with you.  Or maybe you’ve found that those who interact don’t visit your site.  In that case, recognize that what I say here may not apply to you.)

From my perspective, interaction is good.  For Ardea, I look at it this way: for every interaction, even if it’s a simple +1 on Google Plus or a Like on a Facebook post, there’s a person who is indicating they’re more interested in what I do than all the (let’s face it) millions of people who didn’t take the time to click that button.  They’re that much closer to referring business to me, or signing up for a class, or writing a recommendation, or asking for a proposal.  Therefore, I find interaction one way to pre-qualify leads.

Note that although I did start my G+ and Twitter accounts earlier than FB, I gathered data starting July 7th, my first official day of relaunching Ardea.  As mentioned in a social media post recently (Google+, Facebook, Twitter), my follower growth has been fastest on G+ by a fairly wide margin.


An endorsement is the term I’m using for the lowest-commitment social media engagement on the big three sites: a Like on Facebook, a favorite (the star) on Twitter, or a +1 on Google Plus.  It’s an interaction, so it shows interest, but not much more.  I compared the total of all endorsements, as well as endorsements-per-post to see where I got the most interaction; there wasn’t too much difference by percentage.  The chart below shows percentages by total endorsements.  By per-post, Facebook gained about 8 percentage points and the other two showed roughly proportional decreases, so the chart looked similar enough I didn’t include it here.

Social Media Engagement Pie chart for Endorsements: Google Plus 66.7%, Twitter 11.8%, Facebook 21.5%


Comments & Mentions

This is the next level up, in my opinion.  If someone comments on a post, or mentions you in one of theirs (or in another comment, or whatever), we’re talking about that portion of your audience who has risen above the single, easy, one-button click to actually typing things like words or even sentences.  Sure, occasionally it’s not more than “LOL” but usually there’s more thought, time, and effort expended on the part of the audience member.   Like with engagements, I looked at totals and per-post averages.  On a percent basis the two were even closer (Facebook gained only 4% share), so again only one chart.  Note that Twitter might be so big because of a disproportionately long conversation with Danny Wong sparked by my earlier blog post riffing off one of his articles.  His comments were about a third of all the Twitter mentions.  And because only Legolas’ elf eyes would be able to discern the difference in the chart, Google Plus is at 48.3% and Twitter is second at 46.7%.

Social Media Engagement Pie chart for Comments & Mentions: Google Plus 48.3%, Twitter 46.7%, Facebook 5.0%


You should note something else here, too.  Above, I linked to my social media posts about follower growth.  The people who commented on that post in G+ were industry leaders.  I’d consider Danny Wong to be an influencer or “big name” (big enough that I’m happy to have received so many mentions by him on Twitter).  In comparison, 100% of the comments on Facebook were made by myself (replies, follow-up information, etc.) or by people that I already know in my day-to-day life.  In other words, from a quality standpoint, Twitter and Google Plus are delivering interaction and connection with new and influential people.  They are delivering high volume.  In comparison, Facebook is low volume connection with people I already know.  That’s not to say the Facebook comments aren’t from quality people, rather, that I can connect with all those people without ever using Facebook.


Here’s the gold medal level of interaction.  A share is love.  A share is required for “going viral.”  A share is, to use Seth Godin’s terms, a “sneeze” that passes on the “ideavirus.”  It’s what exposes new people to your brand, your services, your marketing messages.  It’s an advertisement, a recommendation, and an endorsement all wrapped into one.  The note here is mainly that I specifically asked people on Facebook to share, whereas I didn’t on either G+ or Twitter.  I was trying to get a minimum number of followers on Facebook so that they would give me stats about my page (a requirement the other two sites don’t have and, even if the rule applied to all three, wouldn’t have been a problem).  Again, there wasn’t much shift between totals and shares-per-post; the difference is more or less the same as I described in “Endorsements,” above.

Social Media Engagement Pie chart for Shares & Retweets: Google Plus 67%, Facebook 29%, Twitter 5%Conclusions

Based on my own pages’ data, it’s pretty clear that from both a quality and a quantity standpoint, Google+ is the big winner.  It takes the top spot in all three areas in both total numbers and per-post averages.  Twitter still provided me with better visibility by and connection to new people than Facebook, even though it came in last in two of the areas.  Facebook only really supplied me with connections to people I already know, and only after I made the ask of my contacts.

If this seems like I’m bashing Facebook, I’m not.  If I wanted to concentrate my marketing on my friends and family, it would have come out on top (remember, “what’s your goal?”).  There are, however, many people who think being on Facebook with their business is the best thing they can do online.  My intent is to show those people that they should probably explore more sites as part of their online marketing strategy.  It’s particularly true if interaction with new and influential people is important to them.  Finally, the Twitter comments example shows that sometimes the needle moves in a big way because of some chance, serendipitous conversation.

Have you seen measurable success on Facebook relative to other sites?  Or are you seeing the same trend I am?  I’d love to hear your stories in the comments below, or on one of these social media sites!

Posted by Michael J. Coffey  |  0 Comment  |  in Social Media

Embarrassing Insights Revealed by Web Data

Be Careful: This Machine Has No Brain. Use Your Own.

You do your best to look good, to be professional, and to get the job done right.  But sometimes you just mess up.  It’s human.  And sometimes, it’s a little (or a lot) embarrassing.

Luckily, a regular review of your website data can help catch some of those mistakes before too many people notice.  Here, for your viewing pleasure, are a few of the errors I’ve caught while swimming in the data stream.  The identities of those at fault, and the businesses they worked for, have been anonymized to protect their dignity.  Because really, wouldn’t we all like to be seen as a little better than “merely human”?

I’ll refer to all of the websites as “” and whoever is in charge at the company as “Mr. Owner”.

Web Data Revealed These Goofs

  • I’m Sending Away Visitors I Already Have…to Myself!  Looking at the data from this site, I noticed a strange pattern.  Lots of visitors were leaving—abandoning it to go instead to Also, the #1 source of incoming traffic to was the site  Luckily for Mr. Owner, I knew immediately what was the problem at set out to fix it.  These so-called “self referrals” are usually due to incorrectly installed tracking code.  Sure enough, after a little exploration, I found some pages that didn’t have the code on it, so if a visitor went to that page, it was like they vanished (at least from the point of view of the tracking software).  And then they came back to a page with the code on it, the tracking program thought, “Hmmm.  Where’d this new visitor come from?  Looks like it was!”  Oops.
  • It’s a Miracle!  I’m Making Sales Without Even Having A Store!  In this case, Mr. Owner had more than one site.  One had an e-commerce site as well as some other sections.  Another website was only a blog.  But the person who had installed the tracking codes had gotten things mixed up and installed the blog-only account number on the blog…but also in the store section of the other website.  From the dashboard, then, it looked like his store was making no sales (despite the fact that he was getting orders), but someone people were buying things directly from his blog.  Crazy!  I made sure all the pages on both sites had the right account number, and now everything looks like it should.
  • You Know What’s Better Than What’s In Your Cart?  Knockoff Drugs From Overseas Somewhere!  Mr. Owner was hacked!  But it was a secret hack.  All that happened was that every once in a while (not every time, mind you), a person might randomly be sent to a drug-peddling site overseas when they clicked the “checkout” button in the store.  The issue was revealed, in part, because it looked like a strangely high number of people were supposedly clicking a link that we couldn’t find on the site, and leaving for a sketchy-looking website.  We had the site’s web host do a security check and that discovered the rest.
  • Button? What Button? Oh, You Didn’t Want To Sign Up, Did You?   Yeah, speaking of clicking on a button, if you have a page that says “Just fill in this brief form and click ‘Sign Up’!” you should probably have a button that says ‘Sign Up”… or a button.  Mr. Owner felt very embarrassed about this one.  But he put it in the middle of this list in the hopes that readers don’t notice (but if you tried to sign up for the Ardea Coaching mailing list in the last week or so, there’s a button now).  The data that revealed this problem?  I’d set up an alert to specifically track button clicks as a goal conversion and I was alerted that there had been no conversions in over a week.  So hurray it wasn’t longer than that!
  • We’ll Provide Your Professional Services!  And a Russian Mail Order Bride, Too.  The last site I’ll talk about had indicators similar to the knockoff drugs issue.  There were an odd number of exits following a link.  And that link happened to go to a site that… well, it was one that would not be good for your boss to see you looking at in the office.   But in this case, the site hadn’t been hacked.  What had happened was that the person who had designed the website originally had included a link back to their site.  Over time, they must have let the domain name expire (or sold it).  Whatever happened, the domain changed hands from a website template designer to a mail order bride site.  Every page on this professional services website had a link on it to a somewhat naughty website.  That wasn’t the kind of “professional services” they were selling, either.  I quickly removed that link from the footer so nobody would question the integrity of the site owner.

What Do You Do?

While ‘saving clients from embarrassment’ isn’t what I usually say when someone asks what I do, that’s sometimes a side effect.  And a reason you should always review the data you’ve got coming in from your site because you never know what surprises you might find on your own site.

Image source: (no modifications except resizing)

Posted by Michael J. Coffey  |  0 Comment  |  in Analysis & Testing
  • Stay Connected