Analytics, That Dirty Little Word

Now that I have my health back (after that bad little fish…you know who you are), I find myself very grateful for my health and enjoying every breath I take. I find it sad that sometimes our lives get so crazy with all the noise around us that we take the most important things for granted, like HEALTH, family and friends. Other than food water and shelter what do we REALLY need? ……………and how do we get these essentials for life in this day and age?

Yes, I would argue that we don’t really need to complicate our lives as much as we do. Our pursuit of ever grander reward, status symbol and leisure; places a lot of pressure on us and the system as a whole. All the noise generated from all this activity can be overwhelming at times, but it’s an unavoidable side effect of a system whose organism’s backbone is increasingly built and supported by technology and our NEED to be current  valuable cogs in that machine.

Just a little over a decade ago cell phones were the realm of the privileged and were the size of a brick and the internet was a way for universities to share and communicate research. Computers as a rule were not used in homes and look at us now! Thanks to our advances in “efficiency” we have created an interconnected WORLD, not just a few terminals in some network. If the internet and the cell phone infrastructure were to collapse, society in general would fall into chaos as all the old systems (pencil and paper, wired phones etc) have been replaced and done away with. Just witness what happens at a modern airport when a given airline’s computer system goes down….it’s not pretty.

So how to talk about analytics after an intro like that? Hardly anybody even knows of their existence but they are now essential and just 5 years from now business will not remember what it was like to be without analytics and will not function without them. Systems adapt to efficiencies and the survival flow below has evolved as well. Society can no longer live off the land; there are not enough naturally available resources to feed all of us. Without modern agriculture there would be massive famine and starvation; fish in lakes, deer in the fields and wild fruits and edibles would last for a matter of weeks if not days.

Systems that are invented to make things easier or efficient just have the effect of driving society to grow to the new ceiling and thus these systems then become indispensable. In the survival flow below the items in bold are new necessities and those that aren’t soon will be.

Current requirements for subsistence:

Food+water+shelter–>Money –>economy and business –>networked systems –>computers and Internet –> web sales and marketing –> analytics –> Social/Viral business and marketing.

Analytics are already firmly entrenched in most major corporations in some way or another and those that are not using them are increasingly at risk. Analytics make those who use them MUCH more efficient and these users will devastate a competitor that has not realized their potential. So even as far down in the eco system as analytics are they are now crucial and will soon be a requirement for business survival.

Analytics And You

Before we get into the heart of the matter, today we need to look over a few terms I will use:

Cookie: no, not the yummy kind…is an identifier generated by a website that is stored on your machine that for the sake of this discussion consists of a unique identifier (random number), time stamp and a visit counter added to it.

Session: another word for a visit to a given site. If you leave a site and return you have started a 2nd session.

Bounce: much like it sounds is a visit to your site that terminated prior to visiting any other page. A visitor “landing” on your site and instantly leaving is a bounce.

Landing: A visitor arriving at one of your site’s pages creates a landing event.

Keyword: words that you use on your site that hopefully are the sort of words people search for on search engines to find the content they are looking for.

S.E.O: Search Engine Optimization. A method for increasing the exposure of your website when people execute searches on the web.

Account: You need a location to store and analyze the data, for this example you need a Google account.

Profile: You need at least one Google Analytics (GA) profile in an account. The profile is where you actually store the data you collect. This allows a single “Administrator” to collect data on many efforts or web pages. You might have several businesses that have their own web pages and you would want a profile for each business, instead of having a ton of data for different businesses all in a big pile.

Administrator: A person responsible for the analytics account. The “admin” can create profiles, manipulate what is reported and define “Users” that have access to this data and what level of access they have (you wouldn’t want the web designers who need metrics on the effectiveness of visuals to have access to financial analytics).

User: Anybody given access to the analytics data by an administrator.

Conversion: A visitor that has gone from just being a visitor/prospective customer to actually becoming a customer of your company or content.

Though in my previous post I already gave a high level description of some of what I will touch on, I decided to go into a little more detail because of an overwhelming response on this topic. As analytics are still a subject that could take months to learn in detail, I will try to at least paint a common case of how you setup analytics and what the data you collect means to you and your business.

Analytics setup

The most important thing to learn is that analytics don’t just magically happen, you can’t just go to Google for example and get the data you want about your website. YOU have to do some lifting on your end as well. For data to be collected on your site you have to actually add some JavaScript instructions to the header of your web pages that tell your browser to execute code that sets up communication with Google Analytics. The code is generated by a utility in your Google Analytic account that makes it unique to your website. Detailed instructions on installation are available at google.com/analytics.

Data Collection

As I explained in my previous post the first step to making sense of the data you collect is to first establish a “normal” baseline. You should collect data for a predetermined amount of time in order to understand a normal pattern of current traffic and the metrics associated with your normal day to day traffic. Then you have established data that you can compare to results once you start making changes to your web site’s appearance, content and start other efforts like SEO.

Once you have a clear picture of your website’s current behavior you can start making changes and use analytics to monitor the effects of these changes whether positive or negative to your business. Once again I can’t stress enough how important it is to track what changes were put in place and when, so that you can correlate these with the data you collect. Your web initiatives need careful thought on who makes changes and how these are recorded.

For Example: If you suddenly changed your websites backgrounds to psychotic green and three days later actually made a relevant positive change to content, you want to remember that the increase in bounce rate started with the color change and not your new content.

How it Begins

When a visitor arrives at your site your GA code will generate a cookie of varying kinds. The cookies will contain a prefix for the kind of cookie that you want set (Permanent, temporary, Campaign etc). The cookie will identify the user with a random number your website will now assign to them and time stamps to indicate when they arrived, the time when they last visited, the time of the beginning of their current visit and a counter that tells how many visits they have had to your site.

GA will create a cookie if this is a first visit or will read and increment an existing cookie if it already exists. The combination of cookie management with associated pages and site events is the basic mechanism on how GA then starts collecting data that you will need to analyze to determine how effective your web presence is.

Data Analysis

I will now just cover SOME of the most useful concepts that are used most often as this is not a tutorial. As such I will constrain my descriptions to generalities and not “click this and then click that” etc.

Once in your profile in Google Analytics (GA) you will have a bunch of tabs that are organized to group data and function categories in ways that make sense. You will have areas that report how many visitors you have had in a given time period  and if they are new visitors or returning, you will have an area that reports time metrics like how long they stayed on any given page as well as their progression within your website from one page to another. This information is very effective for determining if your keywords and SEO and even traditional marketing are doing a good job at getting eyes on your web pages.

Quantity Vs. Quality

Now just because you go the visitor section and see a huge number of visitors have suddenly come to your site, the analysis of that metric is not done..……Quantity does not equal Quality. Yes, there is a metric for that as well; it’s called “Bounce Rate”. If you are getting a ton of visitors to your site but you have a high bounce rate (they instantly leave your site) you probably have at least one of several factors at work.

  • Something is forwarding visitors to your site that have nothing to do with what you sell (guys being forwarded to a site that sells women’s clothes for example just because of a keyword match)
  • Your product is not appealing
  • Your website look, feel and content are inadequate
  • Landing page is incorrect (e.g. people are landing on the “Contact Us” page but were trying to get to one where you display shoes for sale)

In the graphic above you can clearly see that the most common term that routed somebody to a site that matches people up with nonprofit jobs was “I Want To Volunteer” which generated 45,172 visits; but the bounce rate was very high at 56.68% above your mean. The Term “Non Profit Jobs” was the most effective term as it yielded a decent visit count of 5,838 visits and they remained on your site 35.94% MORE than mean. The Term “Nonprofit Careers” was the one that matched your content the best at 65.43% above mean but the quantity of hits was too low. This means that you might want to invest money in “Non Profit Jobs” as a keyword string and don’t waste your money on “I Want to Volunteer” as it will have a high CPC (Cost Per Click) because of all the hits but they will arrive and not stay.

Once you are beyond bounce rates, you start to analyze what pages people are visiting the most and if they are returning to them and how long they spend on them. The Time metrics are some of the most useful and clear content metrics that tell you that you have the products/content people find worth their time. Low time on page or worse a high bounce rate after a given page should tell you that the product or content on that page needs to be improved.

Organic Vs. Adwords

The term AdWords is a product sold by Google to help you get your website to appear at the top of search results. The metric that tracks this is CPC (Charge Per Click). I will deal with the subject of AdWords in another post later as it’s a pretty involved topic as well. Just know that if you have decided to “bid” for certain keywords and yet your Organic results are higher than CPC then chances are you are paying for the wrong AdWords. This concept is one of the main influentials that drives your S.E.O. efforts.

Data Funnels

In my humble opinion Funnels are one of the most powerful tools in GA. Funnels paint a powerful picture that shows the overall health of your website. It’s not good enough to know your total visitor count, Bounce Rate or time on page etc. These are symptoms of the health of your website but Funnels have the capacity to holistically track the progression of a “visitor” all the way to a “conversion”. It gives a graphical representation of this transition.

In the graphic above you see a funnel. It describes in detail how 164 people arrived on your site and the paths they took once there. At the opening of the funnel you see that right off, 94 or 57% of your visitors bounced and left your website on arrival. On the left you see what pages they “landed on” when they arrived at your site and on the right, pages that were visited by those that exited. The funnel proceeds downward and you next see that 70 visitors actually made it to your Registration page. At this step you lost another 40 visitors or 57% of those that arrived at your Registration step were lost, this number might tell you that your registration process is too cumbersome. At the output of the Registration funnel you see that 30 visitors out of 164 actually achieved your goal of downloading the white paper you created.

Funnels are amazing at diagnosing the overall performance of your site and should be designed carefully in order to be able to see all the steps that matter. GA has a great feature that can also be enabled called Reverse Goal Path. In essence it’s a funnel in reverse. GA will show you your conversions and track upwards through their engagement funnels to their entrance and possibly show you that you constructed your funnel in a way that neglected another path….Cool Huh??!!

Alerts and Goals

You can use GA to plan and set operational goals for all your efforts and the metrics that define them. These goals will be displayed in your GA graphics as well as be tracked realtime by GA. Because of this capability you can also associate “Alarms” that are triggered by a goal threshold having been exceeded. Alarms can be attached to positive as well as negative numbers that you are monitoring. Alarms can generate an email to you or even better send you a text message right to your cell phone.

As an example you could have a negative alert such as daily visitor count falling a certain % on a given day. This alert would make you aware that something possibly may have happened to you brand that day, like an influential person said something negative about your product. Knowing this the day it happens allows you to react before real damage is done.

A positive alert could be something like you set an alert event to trigger when your “Sale Complete” page count for a certain item exceeds a certain level. This is a very good alarm as it means you have exceeded a certain amount of sales but you need the alert because you might have to instantly place an order for more inventory to meet future demand in a timely manner.

Filters

GA allows you to filter data that you collect. There are many reasons to want to filter data. Data can be filtered to route certain data to one profile (maybe E-Commerce financials) and another profile (visitor counts, time on page etc). The financials profile would be setup to display data for those in the sales channel and not web designers. The visitor counts profile would be accessible to all users outlined previously. Its important to understand that once the raw data is filtered there is no going back to it. Because of this its advisable to always have an extra profile (maybe called RAW) that would collect all the raw data with no filters. So in this case we would have 3 different profiles for one product; A RAW profile accessible only by the account administrator, A Financials Profile accessible only by those in Sales, Marketing and Finance and last a General Visits Metrics profile that would be accessible by all those previously listed as well as other users given this level of permission. Obviously the Account Admin can access any profile at any time.

Filters support Regular Expression or RegEx. You could also set a filter that would only collect data for index pages 1,2 and 3 on my website if I had 10 for example. The filter would be set to URL/index[1-3]. The data on dwayneholst.com/index4 for example would not be collected.

There are many ways to use filtering and the limits are your imagination. You can use them to reduce the amount of data in a given metric or even to do something like a Search and Replace function. A good example of that would be renaming referring web pages. This way when you generate a report you would have “News and Observer” as your data label instead of http://www.newsobserver.com/2012/07/23/2217110/smithfields-new-human-resources.html  for any referrals that came from The News and Observer.

Summary

I hope this post at least whets your appetite and gets you to start taking a look at analytics and what they can do for you. I personally am always fascinated at the insight I can glean from these reports and the picture they help me paint when I present my findings. As always you can comment on my blog or continue to do as you normally do by dropping me a note at dwayne.holst@yahoo.com. I will be taking a 2 week break from my blog….this last post made my little brain ache!…SHEEESH!! I will talk to all of you soon. Go GET’EM!

 

Advertisements

2 thoughts on “Analytics, That Dirty Little Word

  1. Do you mind iff I quote a coouple of your articles as
    long as I povide credit and sources back to your website?
    My website is in the very same area of interest as yours and my users would really benefit from a lot off the information you provide here.
    Please let me know if this okay with you. Thawnk you!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s