Build a Website Analytics Report from Scratch

What are your website analytics reports really providing you? So many people think that the aim of analytics reporting is to reflect on what happened (e.g. who came to our site and what did they want? how is our campaign performing? did we win or did we lose?). Yes, there are some historical elements, but I would also argue that the best reporting and metrics provide a window to what is next. In essence we can use the past to predict and control our future.

Last year, I prepared a presentation on how to Create a Marketing Dashboard. This information below contains pieces from that presentation, but primarily focuses on the process to build out a website analytics report for the first time. Please keep in mind that while there are reporting standards, there is no standard report. I prepare 10+ reports for clients on a monthly (and sometimes weekly) basis and none of them look alike. Your report should be as unique as the business for which you are preparing it.

To demonstrate each step more clearly, I’ll provide examples from a make-believe business. Let’s say we are preparing a report for a dentist.

A. Define business objectives
This may seen obvious, but I’ve seen plenty of metrics reports that don’t provide a clear indication of what the business is working to accomplish. Your business goals must be at core of your reporting metrics or else you are merely reporting less than meaningful data.

Sample Objectives:

  1. Increase the number of new patients by X%
  2. Retain Y% of all current or previous patients
  3. Improve per patient revenue to $Z

Keep in mind that in most cases the website alone cannot accomplish your business objectives by itself. However, monitoring visitor behavior and website results can help to inform your overall business strategy and help you stay on target.

B. Identify questions the report must answer
Try to avoid pulling random site metrics out of a hat. Think carefully about what you want your report to tell you independent of any specific statistics or analytics-oriented data types (e.g. time on site, bounce rate, total pageviews, etc.).

Sample Questions:

  1. How many new patient inquiries did we receive?
  2. How many new patients called?
  3. How many new email newsletter subscribers did we earn?
  4. What other sales oriented actions did visitors take (e.g. visits to “Contact” page, driving directions, new patient Insurance forms, monthly special page)?
  5. What is the ongoing trend for visitors?
  6. How many visitors come to the site for the first time? returning visitors?
  7. From where are visitors and new leads coming?
  8. What dental services are most popular amongst visitors?

[Read more...]

Marketers Can Learn from Scott Bakula

I was recently asked to speak to a marketing class here in Phoenix about Consumer Behavior. Admittedly a very broad topic, I talked with them about classic definitions, research methods, measurement and monitoring techniques. I also provided them with the story below to help them grasp the concept further as well as other aspects of the marketing profession.

Quantum LeapA popular television series ran from the late 80s to the early 90s called Quantum Leap. The show’s lead character was played by the immensely talented Scott Bakula, probably one of the most underrated actors of our generation. Quantum Leap was a comedy-drama-science fiction hybrid, but it was altogether amazing. The show’s premise was based on a time travel experiment gone awry in which Mr. Bakula’s character, Sam, randomly “leaps” into the body of another person in history. Throughout the series, Sam would move from person to person with each show providing a new setting and experience. While Sam would normally not take the place of any important individual in history, he would often influence them by coming into contact with the likes of Jack Kerouac, Buddy Holly and an adolescent Donald Trump.

Here are a few ways in which I think marketers can learn from Scott Bakula’s character in Quantum Leap.

Get in Character
To avoid upsetting the historical apple cart, Sam would have to take on the persona of his new host after each leap. If he leaped to a baseball player, he’d have to play baseball. If he leaped to a pilot, he’d have to fly a plane. If he leaped to a woman, he’d have to behave, dress and act like a lady. By making an honest attempt to take on the persona of each individual to which he would leap, he got to know them and their surroundings intimately.

Scott Bakula

As a marketer you must also strive to understand your audience. My skin crawls every time I hear a marketer start a sentence with, “Well when I search, I…” or “when I visit a website, I…” or “when I receive an email, I…”. Shut up. Stop thinking about how you would behave in a particular situation. It is paramount that you begin to feel, act and react like your customer. Become a secret shopper. Try your best to imagine what the customer’s feelings, motivations, potential influences and past experiences regarding your product might be. Sometimes this might require research, observation, or god forbid, actually talking to a customer.
[Read more...]

Buckets to Oblivion

How much customer segmentation is too much? Are you slicing the pie into too many pieces? Is there real meaning and significance to your buckets?

To adequately make this point, first allow me to explain marketing segmentation in manner more formal than I am usually comfortable:

Try to avoid bucketing your way to inefficiency.Customer segmentation has often been used by marketers to break a large audience into smaller subsets. The practice of creating audience segments, sometimes referred to as ‘buckets’, allows skilled marketers to deliver unique messages to a clearly defined group and monitor that group’s response to various marketing stimuli. Each group is defined by a common set of criteria that distinguishes them from other types of customers. The lines of demarcation when classifying customers are often aided by:

  • Demographics (gender, age, employment status, location)
  • Psychographics (personality, attitudes, interests, lifestyles)
  • Technographics (ownership, usage patterns and attitudes toward specific technologies)
  • Brand-specific behavioral measures (purchase history, purchase frequency, purchase amount), also known as the RFM model
  • A mix of the four above.

If executed correctly, audience segmentation allows marketers to provide more relevance to individual groups of customers which in turn spawns greater ROI for the business.

Now, here’s the more laid back approach to describing segmentation:

There was always that kid in high school who could get along with every clique. He could talk about beer and girls with the jocks, compare notes on lyrics from The Cure with the goths, play hacky sack with the stoners and share TI-82 calculator tricks with the geeks. He could even make nice with the teachers in a way that wouldn’t alienate himself. Proper use of segmentation is the ability to be that kid.

In a perfect world, we would have the time, resources and knowledge to communicate to every individual customer in a unique way.  For most businesses, that is near impossible and most of us are not ready for retina scan marketing.  However, examination of what matters most to your customers about your offering (and why) will provide some guidance to create the right buckets.  While the merits of segmentation can be described in much more finite detail by people smarter than me, it is possible to be too good at segmentation or perhaps just overzealous. Here are a few common problems and some potential solutions. [Read more...]

Foursquare Analytics Are Shallow

Foursquare ClingLast week Foursquare provided an update to its platform along with an accompanying announcement.  The location based service that allows customers to check-in to local venues and businesses, earn rewards for activity and share status with their friends noted that they had upgraded their specials interface for businesses as well as their analytics.  True to its word, Foursquare 3.0 does carry some new features that allow business owners (and the agencies providing guidance) to discern what kinds of specials they should create to encourage new customers as well as loyalty.  Meanwhile, Foursquare 3.0 also promised to deliver “an all-new analytics dashboard”.  Here’s a comparison of the old analytics dashboard versus the new.

Foursquare Analytics Dashboard Before March 11, 2011

foursquare insights

Foursquare Analytics Dashboard Now

foursquare analytics dashboard

Here’s what changed:

  • foursquare specialsInterface design and layout: easier to view with relevant color, graph and icon usage
  • Specials statistics: see total days run, views and unlock tallies
  • More information about age breakdown: pie chart splitting audience into four age segments
  • Ability to export: download daily check-in and specials stats for the venue you manage

While this update is a step up with respect to aesthetics, it still lacks some important insights that local store owners could utilize to enhance their Foursquare efforts and to educate themselves about their customer base.   Here are some thoughts on what else Foursquare analytics should deliver.

More Detail on Specials – Are they working?  Are they not?  Is my current unlock rate decent compared to other venues in the area?  Were there any significant trends over the time period in which they viewed or unlocked?  Foursquare could also offer recommendations for low view-to-unlock conversion rates.

Where Else Have Visitors Checked In? – Such data would allow store owners agencies providing support to build out personas for various customer archetypes. This would aid in customer service and efforts to increase average ticket totals. There is also an opportunity here for local business partnerships and shared campaigns among venues that have the exact same customer (i.e. “visit us both and save” specials).

Foursquare MayorReturn Frequency – What percentage of my customers are loyal?  Who are my regulars other than this mayor character? Speaking of which…

Customer Ratings Beyond Mayorship – Who has more friends? Most frequent check-in behavior over varying time periods? Most likely to share?  Customers may always be right, but they are not all created equal.

What Kind of Phone? – Foursquare promotes the different phones on which their service can operate right on their Home page. Why don’t they share this data in analytics? This could have implications for mobile advertising and application development.

Tip Sentiment Scoring – Those who check in to a venue can leave tips for others (e.g. “the sashimi rolls are to die for” or “all the nail techs seem nice but my cuticles beg to differ”). What is the relative level of positive vs. negative commentary? Recent press suggests they’ve been at least tinkering with sentiment analysis.

The 3.0 update implies that Foursquare is taking its relationship with businesses more seriously. Perhaps they should provide more meaningful metrics that would allow those same businesses to return the favor.

What other insights should Foursquare be adding to its analytics dashboard?

Finding & Creating Relevant Content

Over the past year, dozens of public relations and marketing agencies are using a new phrase to more aptly describe what they offer to clients: Content Marketing. Essentially content marketing is the art of storytelling with the consumer as the audience and the brand as the muse. Content marketing involves the development of sharable, multiformat, multimodal assets to attract new customers and retain existing ones. For some, the end product of this effort could come in the form of a video, press release, tutorial, informative list, blog post, infographic, opinion paper, original research document, microsite, etc. etc.

However, content marketing that is worthy of positive mention and imitation cannot be likened to throwing a handful of noodles on the wall in the hopes it will stick. Good content should be:

  • oriented to a specific marketing objective,
  • relevant to consumer needs and questions,
  • unique amongst content developed by competitors,
  • appropriate for the distribution channels utilized, and
  • executed in a manner that supports the brand message and builds trust.

Why is Content Marketing Important?
It plays nice with Sales. Relevance and meaning spurs on motivation.
It works well with Search. Content marketing possesses excellent attraction components and link building opportunities.
It has a reciprocated crush on Social Media. Good stories are told over and over again.
It leads to More Content. Good content breeds better content in time with proper testing and measurement.

Content marketing does not overtly sell a product or service. It demonstrates, educates and assists. Often times it ties a product/service to related themes that have meaning and impact on our everyday experiences. Here are just some examples:

Man of the House – a site created by Proctor & Gamble chock full of compelling articles for married men and fathers.

Personal Budget Planner – a free tool from Mint that helps individuals create and manage a personal budget.

Photo Tips from Kodak – a comprehensive collection of photography tips that mentions nothing of buying a camera but rather teaches people how to use their cameras better.

Where Does Great Content Come From?
How can your organization identify those unique storytelling elements that will resonate with all your consumers from the most loyal to the oft fleeting? Here are a few suggestions:

[Read more...]

8 Google Analytics Reports for Paid Search Marketing

Peanut butter and chocolate. Simon and Garfunkel. Apache Junction and a competent dentist. Paid search and analytics.

All marriages made in heaven.

One of the most underutilized resources in a PPC professional’s tool box is a great analytics platform. Here I’ll provide three basic and five slightly advanced reporting techniques that you can use within Google Analytics to effectively monitor and improve your paid search campaigns.

The Basic Analytics Reports

1. Simple Search Status
There are two simple methods to see how your paid search program is performing at a high level within analytics.

First, apply the paid search advanced segments option located at the top right of the Dashboard. Deselect All Visits and check Paid Search Traffic.

Another way to show paid search performance by engine is by accessing your search engines report. This is located within Traffic Sources / Search Engines.

Simply click the ‘paid’ option underneath the line graph to see performance for your paid search campaigns.

What This Report Tells You
This report is especially useful when comparing visits, time on site, revenue, transactions/leads and bounce rate between the major engines. Conversion rate and per visit value are also important metrics.

2. Branded vs. Non-Branded
Many paid search marketers bid for their own brand name. While many feel this may be a cannibalistic approach to driving traffic from your own brand name, it can certainly work in many cases. However, there are many good reasons to do so. I recommend testing to determine the ROI of brand keyword bidding vs. avoiding this method completely.

If you do choose to bid on your own brand or company name, here’s a simple way to track performance:

Go to Traffic Sources / Keywords
Choose to show ‘paid’ as in the example above.
Near the bottom of the keyword list within this report, you will see the following:

Filter your list by entering your brand name and hit ‘Go’. Here you will see a new report containing stats for keyword phrases that include your brand.

Now run a separate report excluding your brand name. Hit ‘Go’, and compare the data from the previous report.

What This Report Tells You
Know how much of an impact your brand name has on your paid search campaigns. You may also find unique keyword variations such as ‘your brand name product’ or ‘your brand name service’ that will help determine how popular different facets of your business are among the searching populace.

When trying to determine whether or not to bid on brand terms, again my suggestion is to test. Use flights carefully (e.g. one week on and one week off) and rely on percentages as opposed to aggregates. Run the same brand keyword filtering reports, but remember to show paid vs. non-paid performance for the appropriate date ranges.

3. Campaign Tracking
Campaign tracking is very helpful but requires consistent campaign naming convention within Google and Bing. You must also tag your URLs appropriately to include the correct tracking code.

I typically use one of two methods for appending a destination URL with the correct tracking code.

Method A: Google’s URL Builder

Method B: An Excel Formula
See columns A through G below. You can populate your sheet with as many rows of data as you like. Please note that the highlighted row should contain your own data.

In column G, paste the following formula on each row. The example below will work for Row 2 in your Excel document (hence the references to A2, B2, C2, etc), but just paste the formula in each corresponding row and you should be fine.

Hint: Use Paste Special / Paste Formula. Then replace all open spaces with a “+”.

=CONCATENATE(F2,IF(ISERROR(FIND(“?”,F2,1))=TRUE,
CONCATENATE(“?”),CONCATENATE(“&”)),
“utm_campaign=”,A2,”&utm_medium=”,B2,”
&utm_source=”,C2,IF(D2<>”-”,CONCATENATE
(“&utm_content=”,D2),),IF(E2<>”-”,
CONCATENATE(“&utm_term=”,E2),))

Once you have tracking code placed and all Campaign names are consistent across engines, you can build out campaign data in analytics. To view individual campaign performance, simply to go Traffic Sources / Campaigns. Here you’ll find an aggregated list of campaigns for a selected date range.

What This Report Tells You
Simply put, know which campaigns are performing and which need improvement. Adjust bids, daily budgets, ads and keywords to modify visit tallies. Focus on your landing page and site experience to improve conversion rate and transaction efficiency.

Advanced PPC Reporting Techniques

4. Engine Advanced Filters
The report referenced in #1 above provides specific information for each engine’s performance. However, there is much more that can be gleaned from the Google Analytics platform. To find additional information about those visitors coming from specific engines, you can create advanced segments.

In the upper right-hand corner of the interface, click on ‘Advanced Segments’ and select the link that reads, ‘Create a new advanced segment’. This will bring you to a screen where you can drag and drop certain measurement attributes. Here’s what the advanced segment for Google CPC traffic looks like. I’ve called this one ‘AdWords’.

And since Bing powers itself and Yahoo, you can create another for adCenter.

What This Report Tells You
Once each segment is created and selected individually, you can begin to explore other areas that will give you a better indication of quality of visits from AdWords and adCenter. For instance, review visitor loyalty and trending within the Visitors section. Check out most trafficked pages, navigation summaries and exit rates in Content. Finally, review Goals and Ecommerce statistics to determine just how well AdWords and adCenter are performing with respect to individual goal conversion and/or specific product sales.
[Read more...]

Analytics Presentation: Data Drives Decisions

It is the general nature of consumers, business owners and human beings to want more for less. As the economic climate has forced organizations to tighten belts and look for the best ways to optimize sales and marketing operations, fleeting answers on how get more juice for the squeeze may be staring them right in the face. Is your business seeking methods to make marketing/sales more efficient, grow revenue from the right sources and understand who your best customers are? If so, I present to you Analytics.

Most of us know analytics already, but we immediately think in bar graphs, pie charts and data tables. However, we sometimes overlook amazing opportunities hidden in the numbers to identify threatening issues, uncover opportunities for growth and substantiate the value of current programs and marketing strategies.

I kept all that in mind when I was asked to speak about analytics to the advertising and marketing staff of a local publisher, The Phoenix Business Journal.

Let’s be honest though. Sitting through a PowerPoint that discusses the nuances of marketing metrics could be seen in the same light as attending a seminar in the library all about toenails. To help me from putting the good people at the Journal to sleep, I incorporated several well-known axioms from the late John Wooden. Wise beyond his 99 years of life, Wooden was a legendary basketball coach for UCLA. He was also famous for delivering classic and logical sayings relevant to athletics, life, business, and in this case, analytics.

[Read more...]

The 3 B’s of Web Analytics

I’m an admitted digital marketing generalist. I love to write, sketch up ideas, and come up with creative strategies. But I also love me some numbers! In the past week it seems I’ve been living in web analytics tools helping current clients, training new ones, participating in The Analysis Exchange and sifting through analytics data for a Fortune 500 company I am verboten to identify.

Mzelle Biscotte / flickr CC

A common question I get when helping clients for the first time is, “so.. where do I start?” Depending upon the popularity of a site, there can be massive amounts of data to mine through. Here’s a list of three items to help you on your way – all of them just happen to start with the letter “B”.

1. Business Goals
Keep this at the forefront: your site has a purpose. What is it? Some websites are built to sell, and others are designed to collect valuable information from customers. Many websites are constructed merely to inform. Identify your business goals and attempt to allow analytics tell you whether you are on track. Are people buying, submitting or finding the information you want to share? How well do relevant, action-oriented statistics align with your objectives?

Ultimately, don’t get caught up in the minutiae. Entrance paths, bounces, referral traffic sources, etc. all have relative degrees of importance, but focus first on what moves the needle from a business standpoint.

2. Behavior
Just as your site has a specific purpose, your visitors have a reason for coming to your site. Analytics allow you to quantify behavior. The numbers always tell a story about where they come from, what they did once they arrived and where they were when they left.

Digging deeper, you can determine what kinds of content people find to be most relevant. You can uncover the methods in which they go about finding that content (navigation, content links, site search, etc.). Further, analytics will show, in many cases, whether each visitor was successful in their specific task(s).

3. Because Why?
After solidifying website objectives and examining visitor behavior, the next step is to derive meaning. “Actionable analysis” is a term that is thrown around quite often in digital marketing circles, but it is important. Too often numbers are tracked and reported as a routine. The act of recording history without interpretation and activity almost guarantees that you will repeat it.

Use analytics data to translate site performance, evaluate marketing programs and learn more about audience behaviors and interests, but don’t stop there. Make positive changes based on these findings or create tests to validate your hypotheses. Once you have done so, start the process again. Analytics is not a beginning or an end, it’s just a turn in the cycle.

Ode to Taguchi

Good marketing professionals are capable of creating compelling messages. They know where to place that message so it will reach the right audience at key moments. Decent and capable marketers know how to measure results effectively to prove success and, in some cases, failure.

Even better marketers understand that not every idea will be a winner. They know that sometimes plans don’t pan out. Really talented marketers know that they don’t have all the answers, but they do know how to find them.

“He who knows best knows how little he knows.”

- Thomas Jefferson

In fact, when speculating on what creative will resonate, what call to action will generate the most activity and what elements will encourage the customer to take the next step, the only voice that really matters is that of the customer. The best marketers create testing programs that help them to listen to the customer, who will always point them in the right direction. Simply put, the smartest marketing minds in the world test everything.

From simple to extremely complicated, there are many kinds of tests you can run. All these can be divided into two primary categories:

A/B Split Tests
An A/B Split helps you determine how one specific element will impact a customer’s decision (e.g. Subject Line A vs. Subject Line B, red button vs. blue button, % off vs. $ off).

Multivariate Tests
More involved than simple split tests, multivariate tests take into account that it’s not always just one element that makes the difference. Rather, the proper combination of elements drives the best results.

The one drawback to multivariate tests is the time required to get a statistically significant result. To use a pay per click text ad as an example, let’s say you want to test two values for three separate variables: Headline, Description and Display URL. To run a full multivariate test, you would have 8 different ad combinations.

  2 Headlines
x 2 Descriptions
x 2 Display URLs
8 Ad Combinations

Google AdWords Testing

For campaigns with limited traffic and/or budget, getting some real results from this testing effort could take weeks if not months. That’s where Genichi Taguchi comes in (not to be confused with a baseball player with the same name).

Taguchi Method

Mr. Genichi Taguchi

So Taguchi

Different Taguchi

[Read more...]

Create a Marketing Dashboard

The presentation embedded below was given at a recent agencyside workshop. agencyside trains traditional advertising and public relations firms as well as budding digital agencies on interactive marketing disciplines, agency operations and business development. Their goal is to help aspiring digital agencies become experts in the space. I’ve been helping them for a little over a month by producing and editing content for their site, webinars and live events.

This particular presentation focuses on marketing dashboards. Agencies and internal marketing teams alike typically create these kinds of reports to provide insight to the performance of marketing programs and campaigns. In this presentation, I’ve identify three common characteristics of successful dashboards, common problems and suggestions for specific data displays. My goal here is not necessarily to show you how to build a dashboard that would reflect something I would do personally, but rather to arm you with some basic tenets to create your own metrics reports that convey meaning quickly and clearly.

The key takeaway here is that your goal in developing measurement reports should not be to merely create a “pretty” visualization of data. The dashboard is simply a tool necessary to the process of determining what steps should be taken next to improve and/or maintain marketing results. Let me know if you have questions…