Articles in the Analytics 101 – The Basics Category

April 20 2011

Using Website Analytics and Research to assist in Marketing Decisions

by

Last week, I had the complete and total pleasure of speaking at the Florida Association of Convention and Visitor Bureaus “2011 Destination Marketing Summit“. You can visit their web site to learn more about who they are and what they do. The conference was held at the Plantation Golf Resort & Spa, roughly an hour north of Tampa, FL.

The title of this blog post was the title of my presentation to this great organization of marketers and IT folks. Clocking in at just under one hour, my presentation was divided into four parts:

A. Preliminary Steps to Take
B. Things to Avoid at All Costs
C. Researching
D. Making Decisions

The first two parts have very little to do with using web analytics at all. Before a marketer even starts doing anything with web analytics, it’s critical to have a proper foundation set. Otherwise, marketers will quickly find themselves in a tailspin of useless reports, meaningless statistics, and possibly, unable to perform necessary technical tasks (like, implementing marketing tags on an Email marketing effort).

The last two parts have everything to do with web analytics and the measurement industry. While we are a Google Analytics Certified Partner and love our Google Analytics platform, I made it clear during my presentation that the platform itself is of little consequence. Users of Adobe’s Omniture SiteCatalyst, WebTrends, Yahoo! Web Analytics, or even those who use server logs and nothing but Excel spreadsheets will all benefit from insights on how to do research and how to make informed decisions based on your research.

Below are the individual slides:

A. Preliminary Steps to Take

1. Find the Right Person – Every company, every organization has that “right person” to champion the web analytics crusade. In all likelihood, that person is you, reading this blog post.

2. Collaborate – Collaboration is huge, because if you are that “right person” who will take over the reins of your company’s web analytics efforts, you will find that you can’t do it alone. Make friends and get buy-in from others in your organization.

3. Tech Check – Are all of your web site pages tagged for your web analytics program? All pay-per-click landing pages tagged, too? What about your marketing URLs – are those carrying referral data? Are your PDF files tagged for inclusion in your tracking tool? Performing routine sanity checks on all things “tech” forces you to collaborate with your friendly, neighborhood IT administrator, and helps avoid post-marketing launch head-scratching.

B. Things to Avoid at All Costs

1. Meaningless Reports – Reports that do not provide context, insights, or do not solve a problem are most likely meaningless reports. Take a long, hard, critical look at the reports you’re generating, and ask yourself if they are doing anything at all for you. If they’re not, stop running them.

2. Lacking Insight – Insight is awesome because it adds such a nice flavor to any statistic or any report you generate. Google Analytics lets you insert insight using Annotations. Omniture SiteCatalyst and WebTrends let you do it via inserting notes into a report. Find a way to incorporate your own analysis and insights.

3. Your Conversion Rate – Please, do not throw anything at me! Your conversion rate is a paradox – a very evil one at that. It’s the metric that we all chase and strive for, yet it is the most harmful and unfair metric of all-time. Got a 2% conversion rate? Good for you – what about the other 98% of your visitors who you’ve neglected, who have most likely performed other important actions? Conversion rate is conversions divided by the number of visits (visitors), and it, by default, pushes aside the overwhelming majority of your online audience.

C. Researching

1. In Your Tool – Great research doesn’t mean great expense. In Google Analytics, you can create custom dashboards and perform a seemingly unlimited number of operations with your website data. You can literally invent your own statistics and computations with Omniture SiteCatalyst. You don’t have to spend any more than you already have spent by researching within your own platform.

2. Not in Your Tool – One of the best, free tools out there right now is called Google Insights for Search. With a few clicks of your mouse and a few keystrokes, you can get trending data, regional interest data, and also a bit of forecasting analysis on what Google thinks will happen, volume and interest wise, on the search terms you insert into the tool. It’s fascinating – and extremely helpful.

3. Design Your Own Tool – Eventually, your web analytics efforts will mature to the point that you’ll need to start creating Custom Reports. Every major web analytics platform allows you to customize your reporting needs based upon what you see fit. Don’t settle for what the web analytics vendors show you by default – crack open their custom reporting solution and pave your own road to wisdom and intelligence.

D. Making Decisions

1. What to Change – Example: Suppose you moved your home page’s main call-to-action from the bottom of the fold to the top-right corner. How did that change impact your bottom line? Using a visual overlay tool, you can clearly see where visitors are clicking around on your web site, and where they are converting from. Google Analytics has the In-Page Analytics report, Omniture SiteCatalyst has the ClickMap report / browser plug-in, and if your tool doesn’t let you see click-stream data on top of your web site, request a free trial from CrazyEgg.

2. What to Invest In – Web analytics tools have gotten so good that they’re starting to become human. WebTrends gives you a few sentences on your dashboard of what’s important and what’s happening with your data. Google Analytics has the Intelligence section, where you can review all significant events that happened on your web site. Use these highly-specialized features to know where to place your marketing dollars (and, where to not place your marketing dollars).

3. What They’re Saying – What people are saying to you in the form of surveys and voice-of-customer tools, and what people are saying on social media is more important and more critical than ever before. Some of the ways that you can evaluate what your visitors / customers are saying is to scan your Twitter and Facebook accounts for certain keywords and hashtags. You can use free tools like Klout and Twitalyzer to evaluate how influential you are to those who may be talking about you. Heck, even your URL shortening tool like Bit.ly or Goo.gl has its own analytics for every URL you shorten, which again can put you “in the know” in terms of your social / brand influence. Who doesn’t like knowing how influential one really is, based upon what is being said?

April 7 2011

Clean up your Google Analytics data with these 5 filters

by

Attention all Google Analytics users around the world: you don’t have to be an expert in regular expressions to use filters. Why? Because this post will help you, that’s why!

No long and drawn-out lead-in to the story this time – here are 5 filters that you can create for your Google Analytics profile(s) that will tidy up your data and make you a happier analyst.

1. Excluding your own traffic from reports
Why: Chances are that your own visits to your own web site aren’t racking up that many visits and page views. Nonetheless, you can still permanently remove your own traffic statistics from appearing in your Google Analytics profile(s).
How: First, grab your IP address from whatismyip.com (or, ask an IT person). If you have administrative access to your account, click on your account’s name, then click on your web property’s name. Next, click on the filters sub-tab (within the profiles tab), click on “Add Filter“, and do the following:

Method: Create New Filter
Filter Name: Exclude my IP Address
Filter Type: Custom Filter >> Exclude
Filter Field: Visitor IP Address
Filter Pattern: ^192\.168\.25\.25$
Case Sensitive: No

Replace the IP address in the example above with your own IP address, but leave the ^, the $, and the three \ symbols (just replace the numbers). Click Save, and you’re done!

2. Lowercasing your hostnames
Why: A hostname is a domain that has sent you visitor data. In other words, a hostname is a URL where your Google Analytics tracking code is present and has at least sent you 1 visit during the selected date-range that you’re looking at. If you ever toggle your report dimension by hostname, or switch the viewing table to show hostnames, you could see mixed cases (upper and lower), which leads to many different variations of your same domain name appearing. That also means you need to work on your SEO re-directs – but that’s something for another time.
How: Go through the same steps as you did in the last filter to get to the filter creation screen. Once there, do this:

Method: Create New Filter
Filter Name: Lowercase Hostnames
Filter Type: Custom Filter >> Lowercase
Filter Field: Hostname

Click Save, and you’re done! You can also create additional lowercase filters to do the same thing to other pieces of data that may look unsightly (one of them might be the Request URI filter field, which represents everything after the .com part of your URL).

3. Search for long, bulky page name; Replace with short, clean page name.
Why: Page names can get long and bulky. There’s probably an important page in your top ten that’s just an eye-sore. How about we shorten it and clean it up some?
How: Follow these filter creation steps – but remember to change the page names to your own, as the following is just an example:

Method: Create New Filter
Filter Name: Search & Replace: Long page with “/john.php”
Filter Type: Custom Filter >> Search and Replace
Filter Field: Request URI
Search String: /your-very-long-and-bulky-page.php?id=1234567
Replace String: /john.php
Case Sensitive: No

4. Add the visitor’s browser to the visitor’s operating system
Why: Why not? Google Analytics lets you create some powerful, advanced filters that let you do something cool (and efficient) like adding the visitor’s browser to the operating system that they’re using. This way, you can see a visitor’s browser along side a visitor’s operating system, without having to apply a secondary dimension (saving your secondary dimension option for something else).
How: Here’s how you do it:

Method: Create New Filter
Filter Name: Operating System + Browser Platform
Filter Type: Custom Filter >> Advanced
Field A -> Extract A: Visitor Operating System Platform -> (.*)
Field B -> Extract B: Visitor Browser Program -> (.*)
Output To -> Constructor: Visitor Operating System Platform -> $A1 – $B1
Field A Required: Yes
Field B Required: No
Override Output Field: Yes
Case Sensitive: No

For Field A and Field B, choose the filter field as described, and then in the blank form field, type in (.*) as shown.

5. Include your domain (and, ONLY your domain!)
Why: Unfortunately, server caching and having your tracking code outright stolen and placed on someone else’s web site is something that we sometimes have to deal with. So, from time to time, you must write a filter that will prohibit the collection of data from every domain except for your own web site.
How: Create your include filter like this:

Method: Create New Filter
Filter Name: Include my domain
Filter Type: Custom Filter >> Include
Filter Field: Hostname
Filter Pattern: mywebsite\.com$
Case Sensitive: No

Click Save to stop the nefarious ones from sending you irrelevant data!

We could write about filters until the next Presidential election, because there is just so much on the topic, and, so many different things that you can do with filters. Even though you can copy the steps outlined in the above 5 filters directly, I still urge you to use caution. Filters are sensitive, temperamental, and must be precise, to say the very least. A poorly-created filter can cause permanent damage, so tread lightly.

What about you? What filters do you like to use? What problems are you experiencing? We’d love to hear your thoughts below!

March 30 2011

Annotate your way to success with your web analytics account

by

I have some good news and some bad news.

First, the bad news: Your web analytics tool can’t answer all of your questions for you.

Now, the good news: There’s something that you can do about it!

As the marketing manager, the research analyst, the IT administrator, or simply the person who has been tasked with all things “web”, you need answers, ratios, percentages, KPI’s, AOV’s, CPA’s, and many other three-letter acronyms and important-sounding phrases. And, you need them now!

So, you log-in to Google Analytics / SiteCatalyst / WebTrends to get this critical data, and the inevitable happens. Your eyes open wide, your eyebrows nearly pop off your forehead, and you say “What happened on that day?!?!?” You thought you were going to sneak in to your web analytics account and tip-toe your way out with everything you needed, only to find yourself smack in the middle of a mystery – why are you staring at an incomprehensible large spike, a puzzling large recess, or drastic change in your report’s line graphs?

No, your tool is not broken.

And no, you didn’t forget to tag your marketing URLs.

Oh, and by the way, your tool can’t help you decipher this riddle because it is not Rain Man, nor does it have the advanced heuristics of Lieutenant Commander Data.

So, where’s the answer? It’s found within the four walls of your office building. Someone decided to send out a marketing newsletter, causing a large spike in visits. The pay-per-click campaign was paused for a day, leading to a big drop in conversions. Your web site’s homepage had a facelift, so your bounce rate decreased to a new low.

Now that you know the answer, or, at least, where to find it, you can most likely deduce what I’m about to say next – integrate the knowledge that has been stored in your company’s HQ with your web analytics platform using annotations, notes, or whatever the equivalent is in your web analytics platform.

Advantages of using annotations / notes are:

1. Non web analytics (yet highly relevant) information becomes a part of your click stream data
2. Everyone in the organization who uses web analytics can stay in-the-loop on what directly or indirectly impacts the web
3. Eliminates the need for guesswork, frustration, and all those wasted hours of trying to solve a problem with no solution

Annotations or notes are always easy to do, and they are usually available for any user that has even the most basic access to your web analytics account.

The following screen shots show you what annotations look like in Google Analytics, and what notes look like in both Omniture SiteCatalyst and WebTrends. Don’t wait until the mystery presents itself – solve the mystery of your data spikes, dips, and shifts before they happen!

Annotations in Google Analytics:

annotations-01
Notes in Omniture SiteCatalyst:

annotations-02

Notes in WebTrends:

annotations-03

© 2017 MoreVisibility. All rights reserved