Web Metrics | Search Marketing
Site Strategy

"In a nutshell, Jason created our marketing analytics capability. He was able to figure out what data we collect, where it is, what was missing, and hook it all up so we canget meaningful, actionable data. Our marketing efforts have improved leads and conversions in some cases by an order of magnitude. He knows his stuff."
Chris Foleen, Marketing Project Coordinator, TransCore, Inc.


Tuesday, May 31, 2011

Thoughs on OMC's "Using Content Marketing To Drive Acquisition" Webinar

I just watched Online Marketing Connect's recent webinar "Using Content Marketing to Drive Acquisition".  The speaker, Chris Baggot, brought up some great points.  

He started out by making a great statement:
"So much or our social media conversation has been kinda touchy-feely...At the end of the day, we all have to grow our businesses."
This was a good strong start in my mind.  Most of what he talked about was a walk through of process. 

He also talked a lot about how content needs to flow around.  
Every distribution channel is also a content collection channel.  No content should be wasted.  Content from anywhere to anywhere.

Almost all content is blog worth and the blog should be the hub.   

Interestingly, Chris pointed out that 80% of all blog traffic is first time visitors.  This is opportunity.  Many times it is repeat visitors that are seen as more valuable because they represent a higher level of engagement.  However, Chris turns this on its head and is clearly focused on growing the audience, not just keeping the audience.

His perspective on content marketing is clear:  Content marketing is a volume game.  Have a message wherever someone might show up.  Target an ocean, then inundate it with content (2-3 blog posts a week is not enough).

In terms of practical advice:
  • Start content planning with keyword research.  This is a way to find new visitors that don't know who you are already.  Uses their language. 
  • If you want to create content that gets shared, it has to be content about people.  Reviews are over rated.  People like to talk about themselves and their experiences.   Foster this by creating "similar-situation stories".
  • Customer service emails can be extremely blog worthy.  A LOT of content is created and sent through email.
Interestingly, he kept coming back to the point that more content = more search traffic.  This was a strong way to blend the mediums.  Using social media to foster content growth that could be echo'd on the site blog to feed search engines was a nice well, rounded way of creating a content generating ecosystem.
Using the Dunbar Circles as an example, Chris noted that:
  • People actively network with about 5-6 people.
  • In general, people cap out on relationships around 150.
  • Avg. number of friends people have on facebook is also 150.

Chris's advice on getting customers/users to build your content for you:

  1. Use personalized email to communicate with users.  Track links.  Get people to tell their own stories.  Remind them to.
  2. Once the content is up, have clear calls to action.  Remember that they are mostly first time visitors.  Blogs are key entry points to websites, especially from a content mktg. POV.
  3. Make people's content easily sharable so they can broadcast your message for you simply by talking about themselves. 

He also noted the value of likes for SEO. More likes = more reach = more positive SEO = more search traffic.

Twitter got classified essentially as a real-time search engine.  The rule for ranking #1 is to be the most recent. 

This was a great idea: Answer facebook questions with customer contributed content instead of the standard company answers. 

He also advised to look for people who post a lot on your facebook page and ask them to be a "guest blogger".  These people want to be involved. 

Tracking Content Performance

  • Look for traffic from specific traffic sources + search traffic.
  • New visitors are key and represent new opportunities created by social.
  • Look at time on site and bounce rate.
  • Track conversions.

Used Western River Expeditions as the example of really good content mktg.

Overall the presentation was pretty well done.  I think he made a very strong and practical case for 
  • How to create more content
  • How to get your customers/users to create content for you
  • How to get your customers/users to market content for you
  • How the content helps push search traffic

Friday, May 13, 2011

7 Ways the New Google Analytics Gets It Wrong

Previously I wrote about the Ten Ways the New Google Analytics Gets It Right.  There are a lot of really good things that are going on in the new interface.  But there are also some ways that the new interface is not quite ready for prime time.  I marked these as "backslides" because, in  my opinion, they are areas where the new interface actually makes things harder or more clunky.  There aren't many of these that I've come across.  But some of the ones I have found are pretty critical..

Here are seven ways the new interface has problems:

Backslide #1: No Advanced Segments in Dashboards
Although the new dash boarding feature is pretty great and represents a real step forward, it is almost unusable simply because there is no Advanced Segment selector.  The benefit of multiple dashboards is almost completely lost if you can't created segment specific dashboards.  The old GA had the ability to apply segmentation to its dashboards.  This is a big oversight on the part of Google and something I hope gets remedied soon. 

Backslide #2: Profile Selection Screen UI
I personally find the profile selection screen to be more difficult to use.  For the profiles I use all the time, I've adapted to using the search box.  But for the profiles that I don't frequent on a regular basis, the navigation is harder.  Also, I haven't seen a way to "star" or flag my most common/favorite profiles.  This feature was very handing when dealing with clients that had many profiles.  Now, I just have to dig. 

Backslide #3: Comparing to Past in Spark Lines
The new way of displaying spark lines when comparing date ranges (stacking instead of overlay) is clunky, it makes the page way too long, and it is difficult to screen grab this information into a presentation. This just doesn't look finished to me. 

Backslide #4: Tiny Tabs
Some of the navigation has been moved from the left nav into tabs.  These tabs are not like the old tabs that looked like tabs.  These tabs are tiny text links.  The way these are implemented and organized takes some learning and could be more intuitive. 

Backslide #5: Table Filtering is Harder to Find and Use
One key piece of functionality that has gotten buried is the table filtering.  If you are looking for all .pdf file types in the content report, you need to filter.  If you are looking for visits from phrases that include a brand name in the keyword report, you need to filter.  Filtering used to be a little text box at the bottom of the results table.  Now it is under a button hidden to the right of  the page.  Then the "Apply" button is on the left side of the page. The buttons are neutral color so they don't stick out requiring your eyes to hunt all over the page to figure out how to use this feature.  It was so simple and effective before.  Why did they every change this?

Backslide #6: Counting Inactive Goals (bug?)
Inactive goals get counted in the total conversions counts.  This should be a bug because it makes the goal overview reports worthless.

Backslide #7: No Emailing Reports (down for now)
You can't email reports or dashboards.  I read this is going to be remedied, but why was it cut to begin with? 

Summary of the New Google Analytics
Overall, the new GA seems very "beta" to me.  It just feels like it's not finished.  Some of the issues I mentioned above--no segmentation in dashboards, inactive goals being counted in the overall conversion numbers, no emailing--make this new version impossible to rely on for my purposes.  Until these issues are fixed, the new interface just isn't ready to the primary interface.  Hopefully the old interface will be around for a while. 

Thursday, April 28, 2011

10 Ways the New Google Analytics Gets It Right

I have been very skeptical of the new Google Analytics interface.  Why?  Not sure.  Maybe I just don't like change, especially when it effects tools that I use everyday and rely on for my prosperity.  From first glance, you can tell there are some big changes in store.  This was not an update.  This was an overhaul in many ways.  And after using it now for a few weeks, I can say that I think it's a good move but seems only partially implemented.
I'd planned on doing screen shots and making a full instructive blog post, but frankly, I don't have the time.  So here is my condensed list of things I like in the new GA so far.  These are in no particular order.

Getting It Right #1: No All Visits Required
This is a huge improvement in data visualization and something I never understood.  Comparing everything to the All Visits segment crushes the scale and you lose the details in lower volume trends.  Freeing us from the shackles of scaling everything at the maximum is wonderful.

Getting it Right #2: Comparing Individual Keyword Trends
If you go to the Organic Search report, you might notice that things have changed.  The list of keywords have a check box next to them and at the bottom of the keyword table here is a button that says "Plot Rows".  Now, you can only plot 2 rows, but it's a start.  This continues with the awakening that Google Analytics seems to be having that comparing trends is central to what web analysts do.  And the more trends we can compare, the happier we are. 

Getting it Right #3: Persistent Report View when Switching Profiles
This a huge time saver. In the past, when you wanted to look at the same information in different profiles, you either had to have two windows running or do a lot of navigating.  Now, using the profile list in the upper right, you can basically swap back and forth
 between profiles in the same report.  Thank you Google, for this one.

Getting it Right #4: Understanding Importance of Segmentation
Moving the Advanced Segments from a right side little selector thing to a top of page, big obvious drop-down makes a statement.  Segmentation is important.  We've known that for a long time, and now the UI is knowing it to.

Getting it Right #5: Differentiating Reporting Styles
There are 3 styles of reports in the new GA.  Dashboards, Standard Reports, and Custom Reports.  In the old GA, Dashboard was the first report.  The standard reports were basically drill downs from the dashboard.  Custom reports were this little hidden text link below the main left nav.  Now they are all top of page, side by side, and big as day.  The UI and the functionality is completely separated for them.  Standard reports are not just the drill down from the dashboard.  And the Dashboard is no longer just a smorgasbord of snippets from the body of reports.  They are each distinctly different.  This individuation of the reporting styles is a bold move by Google.  Analysts will learn to love it.  There is a lot going on here philosophically in terms of the purpose and presentation of data. 

Getting it Right #6: No Auto Loads
Thank you. I no longer have to sit through the  loading of the Traffic Sources Overview report, when what I really want is the Organic Keyword report.

Getting it Right #7:  Event Goals
Why this was never included from the beginning is a mystery to me.  This update has a lot of impacts and really ups the ante on implementation choices between virtual page views and event tags.

Getting it Right #8:  Multiple Dashboards
I'm calling this one out separately from item #5 above because it is important and is one of the strongest indicators that Google has evolved their concept of the "dashboard". Unfortunately the implementation is incomplete, but this is certainly a strong move into new thinking about data presentation.

Getting it Right #9:  Custom Report Flat Tables
If you go into the custom reports and create a report, you will see an option that was never available before--one little drop down list with an option to create an "explorer" style report or a "flat table" report.  Thank you, Google.  Thank you for recognizing that we export a lot of the data out of your reporting interface and having flat tables makes things much easier.  I don't love the new report designer interface, but I can learn it.  Having new options for creating reports that are designed to be exported is well worth it.

Getting it Right #10:  Easy Access to Settings from Reports
That little cog button just to the right is well done.  Another time saver.  Thanks.

There may be many more things about the new GA interface that is worth pointing out and I just have not discovered yet.  But these are a start. 

How the New Interface Fails
It's not all peaches with the new Google Analytics. In the next post we'll look at all the problems in the new GA.  Some of these problems are really critical issues that, frankly, are keeping me from moving 100% over to the new interface for all my reporting needs.  More on that soon...

Wednesday, April 13, 2011

Measuring Social Media's Value -- A Simple Twitter Test

The value of social media seems to be a huge question.  Some people are evangelists.  Some people are skeptics. Some are on the fence.  Some are "gurus".  Some are full of [poo poo]. 

The reach of social media seems to be the best part of its promise.  If you tweet something and 50 people see it and some retweet it and 50 people of theirs see it, and some of them retweet it, etc. etc., you could get your message out to thousands of potential customers!  It's so efficient that its almost like a dream come true.

But the truth is that it seldom works out that way.  In fact it seems like there are reports or metrics showing up all the time that dispell the social media nirvanna so many marketers were hoping for.

I got to thinking about this, and one aspect of social that doesn't seem to come up often in articles I read or reports is the time aspect of social.  You can tweet something and all of your followers will get that message.  But they will only get it for so long.  And if they are not online when it is sent, they will not get it at all.  This really really limits the impact of social media in ways that I don't see anyone accounting for. 

A Little Test 
I am not the most prolific twiiter user by any means.  For the purposes of this test, I wanted to see how long i would be exposed to my own tweet.  This is primarily determined by the number of people I follow.  The more I follow, the more incoming tweets I'm going to recieve, the more quickly things will slide off my screen.

I follow 50 people on twitter.  This is apparently not uncommon, although I could not find any definate stats on what the average number of following was. 

Here are a couple things I found:

June 2009:   92.4% of Twitter users follow less than 100 people

March 2010:  An active or “True” Twitter user has at least 10 followers, follows at least 10 people and had tweeted at least 10 times. By that definition though, only 21% of Twitter users are active users.

Average = 398 following
slide 15

So, I am going to guess that I am around average for the number of people that I follow on Twitter.  Given that, here is what I found:

I sent out a tweet today at 3:55pm.  It took 1 hour and 3 minutes for it to reach the bottom of my screen.  Now, if I have 50 followers and they are all about average like me, then my tweet has a maximum potential of 50 hours of exposure. 

This seems pretty good.  However, this is a bit disingenuous becuase these are simultaneous hours.  At the time my tweet hit the network:
  1. Not everyone is listening.  Take some percentage right off the top.
  2. Some of these people will be following many more people than I do.  They will have less than one hour to experience my wonderful tweet.
  3. Some days and times of days are more active for twitter than others.
The concept of simultaneous hours is pretty important and really puts the kabosh on the notion of the value of a tweet. 

"Time to Conversion" -- A New Social Metric?
One interesting metric to come up with would be "time to conversion".  You could optimize for this by calculating how much time exposure you need in the social space before someone actually buys something from you. 

Using "time to conversion" you could figure out:
  • How many tweets to send
  • When to send tweets to get maximum exposure time
For example, say you did some math and figured out you need two thousand hours of twitter time to get one sale.  Then you could baseline it and test against it.  Requiring less and less "time to conversion" would represent good optimizaiton, increased social value, and real  marketing savings.

Also, if you figure in the labor costs of coming up with however many tweets are required for 2000 hours of twitter time and put that against the value of a sale, then you have the makings for a real ROI from social media.

"Time to conversion" might be a strange way to see the value of a tweet, but I bet with some further thinking it could be valuable.

Wednesday, January 19, 2011

Visio Stencil for Information Dashboard Design

I have been doing a lot of dashboard design lately.  When I do dashboard design I like to use Visio.  It is easy to use and you can output as PDF or an image file for easy sharing.  

One thing that always is a little bit of a pain is getting the right collection of chart samples together for designing a dashboard.  There are so many kinds of charts and ways to put together a display of data, that it can get tiresome to always be trying to find a sample.  So, a while back, I started collecting charts and created a Visio stencil to hold them.  This way I don't have to go hunting for chart samples anymore.

And now neither do you!

Here is the stencil that I created and have been using for dashboard design.  It comes with a selection of charts and text objects and annotations that have been useful to me.  It is not an exhaustive collection, but it has the more common elements. 

I'll be adding to it over time and will post updates as needed.  If you want a shortcut to designing dashboards or visual reports, give this a try.  This is most useful for designing Power Point style visual reports, not interactive reporting tools.