If The Wizard Is A Wizard Who Will Serve: Using Web Metrics To Accomplish Goals

For my final blog post, I am tasked with evaluating an online retailer’s web analysis tools and techniques. Since I do not have any contacts with online retailers, I located a case study highlighting how Discount Tire used Google Analytics to increase website performance and increase online sales. In my very first post, I said:

Many companies don't know where to start when it comes to web metrics; they don't even know what questions to ask. Others know what to ask, but they are afraid because they fear the answers might be beyond their reach. So, they "pay no attention to the man behind the curtain" so-to-speak and maintain the status quo, never really moving forward or achieving their business goals (at least, not that they can track anyway). It's true that there are a lot of metrics and even more ways to apply them, but a basic understanding of key metrics can help a company evaluate their online presence and compare it to business goals and objectives (Nauman, 2012).

Before launching into the case study, I thought it was important to first determine the goals that Discount Tire was trying to accomplish when gathering and analyzing their web data since goals and objectives drive KPI’s and determine what data to collect. If Dorothy did not have a goal of going home, her visit to Oz would have been much different. Because she had a specific goal in mind, she was able to take the necessary steps to accomplish them, i.e. killing the Wicked Witch (obstacle) who was preventing her from reaching her goal. 

Case Study Overview

According to the case study (Google, n.d.), Discount Tire had clear goals of driving traffic to their website for both direct sales and for browsing before making an in-store purchase.

Online shoppers can try wheels virtually, seeing how wheels will look on specific vehicles with an Interactive Wheel System. They can also save time and money by ordering tires or wheels without paying for shipping, or by purchasing tires and making appointments for installation at their local store-all online (Google, n.d.). 

It was important to track more than just visits; they needed to determine how customers were arriving there, and what they were doing once they got there. By collecting this data, Discount Tire could create a positive web experience that would increase sales both online and in-store. 

Online advertising included a variety of search-based campaigns, including AdWords. These types of campaigns are critical for Discount Tire as it allows them to target specific customers with local, regional and seasonal messaging. They had tried a variety of web analytics tools to track and measure the effectiveness of Internet marketing efforts, with a full-time web metrics analyst and online marketing agency. Travis Unwin, Managing Director of Discount Tire's online marketing agency said “We've used everything from log file analysis to cookie tracking and pixel tagging to measure the effectiveness of our website and of Internet advertising. Unfortunately, they all had major shortcomings" (Google, n.d.). 

Working with the online marketing agency, Discount Tire implemented Google Analytics. The seamless integration with AdWords campaigns offered immediate, accurate results measurement. They were able to tag URLs easily, and set up conversion funnels and goals without having to add additional tags, which helps page load times. Unwin was able to track the progress of marketing campaigns, run keyword reports and demonstrate the value of online search advertising to Discount Tire’s bottom line.

After implementing GA, Discount Tire could see ROI on specific products, gauge customer purchasing patterns and view traffic patterns on the website so they were able to streamline customer interaction from landing page to purchase page. They also used GA when conducting A/B testing on site language to drive sales. For example, they discovered the most successful language for the checkout button. 'Purchase and make a reservation' resulted in a huge drop in sales, but changing the text to ‘check out and make a reservation’ produced a 14 percent increase in online sales in a single week (Google, n.d.).

Viewing metrics related to cart abandonment, they determined that cart abandonment rates went up when items were not in stock at a local store. By adding language to the page informing customers that out of stock items would be located and stocked at their local store, cart abandonment rates dropped and sales increased 36 percent (Google, n.d.).

My Recommendations

While the case study is not dated, language within it suggests it was written in 2005. A lot has changed since then and there are new tools that could provide additional value. Discount Tire is known for “fair dealing, reasonable prices, top-notch customer service and guaranteed satisfaction” (Jarman, 2011). With the employment of Social Media, including Facebook and Twitter handles, implementing a tool to provide a dashboard for the marketing team to quickly see the impact tweets and posts have on website traffic as well as monitor what customers are saying and be able to respond quickly would provide additional value. 

Eye tracking tools could offer insights into customer interaction with the website. While funnels provide an overview of pages users visit in sequence, eye and click tracking would offer a more detailed view of actual user experience. Data collected by this type of tool would pinpoint areas on the website that are potential stumbling blocks to actual sales, and could assist with page layout to ensure key information is accessible and visible to users.

Data related to visitor locations can help when determining new locations to open. If a lot of visitors from St. Louis, MO are searching the site only to discover there are no store locations nearby, a feasibility study might need to be completed to determine if it would be a good location for growth. Similarly, large percentages of users accessing the site via mobile device could indicate a need for a mobile site.

Finally, discounttire.com does not have a search feature. Adding a search feature and then tracking commonly searched keywords could offer invaluable insights into what customers are searching for. This data would indicate pages or information that may be best featured on the home page or perhaps even used in marketing. By comparing on-site search data with referring traffic, it could provide interesting data on how the homepage should be laid out for the optimal user experience.

Conclusion

Why was Dorothy going to see the Wizard of Oz in the first place? She wanted to get home. The Scarecrow wanted a brain, the Tin man a heart, and the Lion some courage. After allowing the Lion to accompany them on their journey, the group sings the line “If the Wizard is a Wizard who will serve, then I'm sure to get a brain, a heart, a home, the nerve!” (LeRoy & Fleming, 1939). In the case of Discount Tire, once they were able to get web metrics to work for them, they were able to accomplish, and I might say surpass, their goals.

References

Google. (n.d.). Discount Tire increased online sales by 14% in the first week using Google Analytics. Retrieved from http://www.google.com/analytics/customers/case_study_discount_tire.html

Jarman, M. (2011, January 2). After 50 years, service still drives Discount Tires. AZcentral.com. Retrieved from http://www.azcentral.com/business/articles/20110102good-service-still-drives-Discount-Tires-0102.html

Nauman, M. (2012, November 1). Start at the beginning [Blog post]. Behind the Emerald Curtain. Retrieved from http://emeraldcurtain.blogspot.com/2012/11/start-at-beginning.html

LeRoy, M. (Producer) & Fleming, V. (Director). (1939, August 25). The Wizard of Oz [Motion Picture]. United Sates of America:  Metro-Goldwyn-Mayer (MGM)

Funnels, Filters and Goals (Oh My!): Finding the Courage to Take Action

Funnels, filters and goals, oh my! For the inexperienced web analyst, these terms sound scarier than flying monkeys, which are pretty scary! In reality, these terms are fairly easy to understand and extremely important to fully understanding web and business goals and successes. To start, I will briefly explain these concepts as they apply to Google Analytics (GA). 

Funnels 

A funnel is the set of steps or page views a visitor takes to achieve a goal. By tracking a funnel, you can see drop off points and determine what needs to be fixed. For example, on an e-commerce site, a funnel might be the shopping cart, payment page, and confirmation page. If a lot of traffic drops off on the payment page, there could be an issue with inputting payment date, or it could mean that visitors are leaving to go in search of a coupon. 

Filters

Filters will remove data that is not pertinent based on the parameters setup by the administrator. It is important to note that once filters are setup on a profile, the filtered out data cannot be retrieved and re-filtered. Because of this, it is recommended that a profile be maintained that includes all data. GA allows each domain to have up to 50 profiles. Many companies will use a filter to exclude their IP address, that is, their employees. If GA tracks all IP addresses, company employees who regularly hit multiple pages every day can skew reporting. 

Goals 

As I have mentioned previously, metrics are just numbers without human analysis. After determining which metrics are important to your business, you can set up goals within GA to track those specific metrics. There are four ways to measure goals (Lofgren, n.d.):

1. URLs – when a user lands on a specified URL, like a thank you page, PDF or confirmation page.
2. Time – When a visitor spends more or less than a specified amount of time on the site. 
3. Pages/visit – Tracks when a visitor has visited more than, less than or equal to a specific number of pages during a visit.
4. Events – When a visitor completes an event. The event is set up and then tracked once its been triggered. An event could be a download, social shares, time spent watching videos, etc. 

Once goals are setup in Google Analytics, they can easily be tracked and viewed in GA reporting.

So, how do these apply to a real company with real website goals? Following is an example of how these tools can be used at my place of employment, Greenville College.

Earlier this year, Greenville College sent a postcard to current teachers in their target market directing them to a landing page for information about degree programs that could help them advance their career. The preferred funnel was 1) the landing page; 2) one of four program pages; 3) request information page; 4) confirmation page. We also included a contact form on the landing page, but we expected recipients to click-through to more detailed information before submitting contact information. In order to track the effectiveness of this campaign, I could use funnels, filters and goals. 

With funnels, I want to see how many visitors followed the funnel to completion. If the report shows that only a few visitors followed that exact funnel, I can take a step back and see which users went from the landing page to the program page, and then follow where they go from there. In this case, most users clicked through to the program page then dropped off. There are a multitude of possibilities as to why they drop off at this point. It could be that the program was not interesting to them, or maybe the request info was not clear on the page (which I believe it is not). I can only hope that these users will come back again later, from either bookmarking the page, hitting the landing page again, or a simple Google search. In the future, I could use funnels to track A/B tests and determine which landing pages resulted in the better funnel flow (Alexander, 2012). 

Filters can be utilized to see where we had the most effective reach. 

1. Filter by state. To see interest in Illinois, I could filter out all states except Illinois. Or, I could reverse that to see if our message is effective outside Illinois. With state laws and certification requirements varying from state to state, this could be highly useful information.
2. Filter by city. Like point one above, seeing city-level data could help us determine where our message was most highly received. If the cities within a 30-mile radius are the most effective, we may want to focus our marketing efforts in that radius. Or we may find cities where we have done little marketing previously, that had a high response rate. 
3. Filter by technology. An interesting filter to utilize would be the mobile filter. This filter could help me determine how many users are accessing the landing page via mobile phone, and if it affects their traffic patterns.

There are a lot of possible goals to use in this scenario. 

1. I want to collect contact information from both the landing page and the request info page. I can set the confirmation page used on these pages as a goal and track.
2. If I want to know when visitors are downloading course descriptions for the programs, I can set this simple goal to trigger when visitors click on the link, which automatically downloads a PDF.
3. I also want to know any time a visitor directly emails the admissions representative from the website. In order to view this information, I would first need to set that action as an event, then set the event as a goal. 

Google offers a library of information about each of these topics, with detailed explanations on how to set them up on your website. For further reading, click here.

So, how can you use filters, funnels and goals (oh my!) in your web reporting? Can you muster up the courage to give it a try? If so, let me know how it goes in the comments below.

References:

Alexander, D. (2012, September 10). Google Analytics goal flow: More funnels with less work [Blog post]. Luna Metrics. Retrieved from http://www.lunametrics.com/blog/2012/09/10/goal-flow-more-funnels/

Lofgren, L. (n.d.). 4 Google Analytics goal types that are critical to your business [Blog post]. Kissmetrics. Retrieved from http://blog.kissmetrics.com/critical-goal-types/ 

Blogger Metrics and the Tin Man

Now that my blog is up and running and I have convinced my friends, family and classmates to read the content I have posted, I am able to review the metrics for my site. Remember, metrics are just numbers; it takes a human element to make the metrics hold value. Otherwise, metrics are like the tin man, just standing around, not fulfilling their purpose. Blog objectives will ultimately determine key metrics or KPI’s. In my first post I discussed the importance of bounce rate. Unfortunately, bounce rate is not a good metric for bloggers since most readers will hit a page, read the content then leave, registering as a bounce. As a new blogger trying to grow my readership, there are five metrics that are of importance to me:

1. Total visits
2. New visits
3. Repeat visits
4. Referring pages
5. Comments

Total visits

This metric lets me know if people are finding, and reading, my blog. The more visitors I receive, the better. Total visits will measure new and returning visitors, which are also important on an individual level, as discussed below. The total visits metric alone does not tell me if the visitors are staying to read the content or not, but if the metric goes up, it indicates some level of success as time goes on. Also, by looking at total visitors per page (or blog post), I can get a better understanding of what content my audience is looking for (Georgieva, 2011).

New visits

New visits let me know the number of first-time visitors I have had to my blog. It indicates that new visitors are finding me from friends, search engines, or other ways. Like total visits, this metric does not indicate the quality of the visitor, but since I am wanting to grow my readership, the higher this number is, the better it is for me.

Repeat visits

Obviously, this metric represents the counterpart to the new visits metric. This number lets me know about my readership. If the repeat visits metric is high, it means I have a good number of readers who found value during a previous visit and took the time to return to read more. 

Referring pages

This metric lets me see how people are finding my blog. Are they following a link on another site? Are they coming over from my self-promotions on Facebook? Are they using keyword searches? Determining how my readers found me helps me know which promotions are working, and/or what relationships and referrers I need to focus on. If another blogger is sending traffic to my site, I would thank them and try to start a mutually beneficial relationship with them, or with other related sites.

Comments

Comments are a great way to measure engagement. It is very daunting to start writing and then wonder if anyone is actually reading. While all of the above metrics could indicate that someone (hopefully several someone’s) is reading, comments are the proof. A comment is an opportunity to not only receive feedback, but also dialogue with a reader (Kingston, 2011). Most blog readers will not leave a comment, so when someone does, it is imperative that you respond quickly or you risk alienating the reader. 

With all of this in mind, let us take a look at my blog’s metrics.

In Google Analytics (GA), I can get a good overview of the first three metrics in the audience overview section, which looks like this:

With this overview, I can see that 22 individuals visited my site over 41 visits. Of the 41 visits, 53.66% were first time (new) visitors, while 19 were returning visitors. Since the timespan dates back to the origination date of my blog, the unique visitors metric matches the new visitor metric. This will not always be the case. These numbers by themselves do not mean much, but it is providing a baseline for future analysis. 

The fourth metric, Referring Pages, can be located in GA under Traffic Sources and looks like this:

This reports shows that the most referrals came from blogger.com, with Facebook (regular and mobile) close behind. With this information, I know that people are coming to my blog because I have posted something new and told them about it. If I scan to the right, I can see the percentage of new visits for each referring site. Interestingly, of the 12 visits referred by blogger.com, only 16.67% are new visits, which means that most (83.33%) are returning visitors. Thank you, blogger.com for the continued traffic! The direct traffic is most likely from my classmates since 100% of that traffic is new visitors and the only place I posted a direct link was on blackboard. Anyone who bookmarked my site would fall into this category, but the % new visitors would decrease over time.

I can get a good overview of the first four of my key metrics in the New vs. Returning visitors report, which looks like this:

This report lets me see detailed information about visitors, visits and referrers in one place, rather than having to go back and forth between several. The downside of this report is that it does not show me a clear breakdown of new vs. returning visitors on a percentage basis.

The last metric, comments, is only available within the Blogger metrics tool. I hope to be able to set this tracking up in GA in the future, but this capability does not appear to be available at this time. Blogger provides this overview:

At one glance, I can see the total number of comments. Since my blog is new, this is a manageable way to view comments. However, since I am unable to drill down into comments for additional metrics, it will lose usefulness as the blog grows.

Additional metrics I will want to track in the future include: social media sharing, downloads or views, and subscribers. Each of these metrics provides additional insight into customer engagement and what readers are finding valuable. 

Are there additional metrics you feel would be valuable for a blogger?

References:

Georgieva, M. (2011, December 1). 5 critical metrics to measure business blog performance [Blog post]. HubSpot. Retrieved from http://blog.hubspot.com/blog/tabid/6307/bid/29315/5-Critical-Metrics-to-Measure-Business-Blog-Performance.aspx

Kingston, C. (2011, September 8). 7 tips to increase your blog comments [Blog post]. Social Media Examiner. Retrieved from http://www.socialmediaexaminer.com/7-tips-to-increase-your-blog-comments/

Some people do go both ways: Facebook ads and Google AdWords

When it comes to digital advertising, two popular avenues are Facebook ads and Google AdWords. Some people say Facebook is the way to go, others point to low click-through rates and insist Google AdWords offers better ROI. 

Pardon me; that way is a very nice way [points left]... It's pleasant down that way, too [points right]… Of course, some people do go both ways [points both ways].

Facebook Ads 

Facebook has an overwhelming amount of information about their users and allows advertisers to drill down and effectively target their audience with a very tailored message. Beyond basic demographic information, advertisers can select specific employers, industries, pages the audience likes, hobbies and interests, and much more. The advertiser selects the headline, content and photos associated with their ad, as well as the timeline and budget. When selecting a budget, advertisers can choose to pay on a cost-per-click (CPC) model or a cost-per-impression (CPM) model. Marketing goals and objectives determine which model to use. For instance, general awareness might be best purchased on a CPM basis. On the other hand, to generate Facebook likes or other conversion-related goals, a CPC model might be best. 

Since advertisers can set their ad spend, specific messages and audiences, they can easily perform A/B tests to determine which are the most effective. Different headlines, photos or ad copy can be used to target the same audience. Or, they could choose to run the same ad in several locations and see which locations performed better. Each ad offers several reports to track click-through rates, social reach, impressions, reach, frequency, etc. 

Recognizing that the most effective form of advertising is word of mouth, Facebook also offers an option of running sponsored stories. These ads are shown to current fans of a company’s Facebook page. The ad displays with “Friend x likes brand” above the ad content within the user’s news feed. Data suggests that these ads are much more effective than traditional display ads due to the “recommendation” element of the ad (Edwards, 2012).

On the downside, Facebook continues to make changes to their advertising policies, making it increasing frustrating for the advertiser. However, advertisers continue to utilize Facebook advertising because of the 536 million Facebook logins each day (Le Viet, 2012), their targeting ability and the relative low cost of advertising.

Read how Tough Mudder has effectively used Facebook advertising here. 

Learn more about advertising on Facebook at their website.

Google AdWords 

While Facebook pushes tailored messages to a specific audience based on demographics and behaviors, Google AdWords works by providing relevant search results to users in the form of a text ad. A company with a Google AdWords campaign sets keywords that are relevant to their product or services and associates each with an ad and a CPC or CPM bid. For example, a shoe store might set keywords like “children’s shoes”, “red pumps”, “wedding shoes”, etc. When a user performs a Google search for red pumps, an ad will appear directing users to the company’s website. For example, “Shoe store X has red pumps in a variety of styles”. 

The greatest obstacle with AdWords is optimizing keywords so users who click on the ad are potential customers. If a shoe store that specialized in children’s shoes had an ad running with the keyword “shoes”, they would show up in searches of people who are looking for athletic shoes, wedding shoes, winter boots, etc. Each click by these users would cost the company money without increasing their sales or increasing brand awareness with potential customers. Without careful planning and continuous optimization, a company could end up spending thousands of dollars with little return.

Like Facebook, Google AdWords has some targeting capabilities, though it is much more limited. After creating the ad, selecting keywords and establishing a budget, you are offered options for geographic reach and language (Dahl, 2009). A small business located in Michigan would not want their ad to appear in California so they would select a radius around their store location and languages that pertain to their target audience. 

Google also allows companies to advertise on their AdSense network. Anyone who spends time on the Internet has seen text ads on their webpage. Many of these utilize Google’s AdSense network. When advertising on this network, companies can specify which sites they would like their ad to appear. For example, a religious institution would probably not want to appear on sites affiliated with alcohol or gambling, so they would opt-out of these types of content.

A bonus of using Google AdWords is the built-in measurement tools of Google Analytics, which, in my opinion, far outperforms Facebook’s reporting. 

For more information about Google AdWords, check out Google's help section.

Which is better?

There is no better or worse in this scenario. Business objectives will play a key role in determining which advertising avenues to use, just as they do when using traditional media. “Given the challenge of having a product that hasn’t existed before, AdWords may not be the best tool for generating interest and sales” (Dahl, 2009). However, if your customers are not on Facebook, then that might not be the best solution either. Google targets people as they are searching for a product while Facebook targets a specific group of users. An integrated marketing communications campaign might find success using both.

Have you found better results with AdWords or Facebook ads? Or have you found that a combination of both offers the best ROI?

References:

Dahl, D. (2009, October 14). Real-life lessons in using Google Adwords. NY Times. Retrieved from http://www.nytimes.com/2009/10/15/business/smallbusiness/15adwords.html?pagewanted=all

Edwards, J. (2012, September 18). CHART: Facebook’s sponsored stories are way more effective than display ads. Business Insider. Retrieved from http://www.businessinsider.com/facebooks-sponsored-stories-vs-display-2012-9

Le Viet, S. (2012, July 21). 5 best reasons to recruit with Facebook Ads. Mashable. Retrieved from http://mashable.com/2012/07/21/recruit-facebook-ads/

Some people without brains do an awful lot of talking: The content vs. conversation debate.

A hot topic of debate in the marketing industry these days is content vs. conversation.  Both sides claim their view is the most important. I will briefly discuss each viewpoint before offering my opinion.

Content is king

The content camp believes that generating regular, consistent content is the most important function of a social media marketing strategy. If you do not have content, you do not have anything to talk about. If you do not have regular content that readers can depend on, you will alienate them. 

In 1996, Bill Gates predicted “Content is where I expect much of the real money will be made on the Internet… Those who succeed will propel the Internet forward as a marketplace of ideas, experiences, and products—a marketplace of content” (Ladd, 2012). In 2012, content is easily generated by anyone with access to the Internet. Blogs have exploded along with social media – the uploading and sharing of ideas is normal and expected. In June 2012, Twitter reported they generate 400 million tweets per day, and that number is growing (Farber, 2012). Consumers expect brands to compete for their attention by generating relevant, engaging content on a regular basis.  Penn (2012) states, “Spreading unique, helpful bits within the industry or with consumers creates brand awareness, new customers and client loyalty.” Her infographic provides some statistics about content marketing that are relevant to this conversation. 

Conversation is king

On the other side, those who believe that conversation is king believe that in the social media space, content is just something to talk about, while the conversation that arises is what is important. 

When Bill Gates made his prediction in 1996, great content was scarce. Today, with the social media explosion, content is everywhere. In March 2012, we were closing in on 200 million blogs worldwide (Nielsonwire, 2012). The problem is that this content is not always relevant, engaging, or even dependable. As the Scarecrow said in the Wizard of Oz “Some people without brains do an awful lot of talking” (LeRoy & Fleming, 1939). The conversation camp believes that content needs to be engaging to be of value, otherwise it is just advertising (Novak, 2012). By generating content that spurs conversation, a brand can truly begin dialoguing with their customers and take advantage of the most effective type of marketing – word of mouth (McCarthy, 2011). 

Context is queen

Just to throw another wrench into the puzzle – a new camp is arising that says that context is queen to the content king. Meaning, companies need to understand where the customer is coming from, how they are getting there, and what they are looking for. Once they have this data, they can provide the customer with the information they are seeking. For example, if a customer walks into a physical shoe store and wants to buy red pumps, she would have little tolerance for a sales person who brought back athletic shoes. The same is true for the web. If the customer Google’s “red pumps” and is provided with generic links that take her to a shoe store landing page filled with kids shoes, athletics shoes, etc. she will be dissatisfied. Understanding consumer preferences, location, behavior and social networks is key to understanding context (Savitz, 2011).

Personally, I feel that content is king, context is queen and conversation is the soul. Content without context or conversation is a lonely king with no kingdom. Conversation is just people talking. You need to provide content to guide those conversations into something meaningful that aligns with business goals. Having all three working together will produce a thriving kingdom.

So, how do you rate content, conversation and context? 

References:

Farber, D. (2012, June 6). Twitter hits 400 million tweets per day, mostly mobile. Cnet. Retrieved from http://news.cnet.com/8301-1023_3-57448388-93/twitter-hits-400-million-tweets-per-day-mostly-mobile/

Ladd, C. (2012, September 5). Is content king for business marketers? Forbes. Retrieved from http://www.forbes.com/sites/gyro/2012/09/05/is-content-king-for-business-marketers/

LeRoy, M. (Producer) & Fleming, V. (Director). (1939, August 25). The Wizard of Oz [Motion Picture]. United Sates of America:  Metro-Goldwyn-Mayer (MGM)

McCarthy, P. (2011, October 16). Infographic – The word and the world of customers. The WOMMA World [Blog]. Retrieved from http://womma.org/word/2011/10/16/infographic-the-word-and-the-world-of-customers/

Nielsenwire. (2012, March 8). Buzz in the blogosphere: Millions more bloggers and blog readers [Blog post]. Retrieved from http://blog.nielsen.com/nielsenwire/online_mobile/buzz-in-the-blogosphere-millions-more-bloggers-and-blog-readers/

Novak, C. (2010, July 27). Why conversation, not content, is king. Social Media Today. Retrieved from http://socialmediatoday.com/wordspring/152636/why-conversation-not-content-king

Pann. J. (2012, February 8). Marketers who share content drive traffic, gain customers [INFOGRAPHIC]. Mashable. Retrieved from http://mashable.com/2012/02/08/new-content-marketing-tactics/

Savitz, E. (2011, August 30). Context will drive the future of web content management. Forbes. Retrieved from http://www.forbes.com/sites/ciocentral/2011/08/30/context-will-drive-the-future-of-web-content-management/

Are you a good witch... or a bad witch?

It’s only natural to want to know the people who are visiting your website (or blog, landing page, etc.). While social sites like Facebook can provide detailed demographic information about fans and followers, visitors to websites can’t be pinpointed so easily. However, learning how they landed on your website can be measured using traffic source metrics. In Google Analytics, it looks like this:

Search Traffic

This segment of visitors arrived at the website via organic (non-paid) web search. Drilling down into this section will show the top keywords used in the search. Note: if you have paid search marketing running, this will show up under a different section, but the information will be similar. Google Analytics shows keywords like this:

In this particular case, the top performing keywords are not provided (this happens when a visitor is signed into a Google account). The second best performing keywords are “Greenville College” which account for 28.51% of the search traffic. Each keyword can be drilled down further to see bounce rate, time on page, etc. to determine the overall performance. 

Here, I drill down the #3 ranking keywords "Greenville College Athletics":

In the case of Greenville College (my employer), the "Greenville College Athletics" keywords accounted for 1.59% of visits, and had a bounce rate of 16.25%. I can deduce that visitors searching for Greenville College Athletics wanted an update on scores for a game or to read a story about a recent win or athlete. These visitors will spend several clicks locating the information they are looking for, and several minutes reading the information. If the bounce rate goes up over time, I would visit the page to determine the cause of the bounce. 

Referral Traffic

Like before, drilling down into this section will show the top 10 referring pages. Many times the top referring pages and sites will not surprise you, but on occasion there will be an unexpected listing.

The screenshot above reveals that Facebook is the top referring page for Greenville College. This is great news for Facebook and indicates that it is an area where we need to continue to spend time and money. The other referring sites are predominantly organizations where Greenville College has been cross-promoted via conferences and partnerships. Google.com comes in at #4. This could be an indicator of good, or poor, SEO performance (Kaushik, 2010, P. 86).

There may be times when a source is listed that you are unfamiliar with. It is worth the time to investigate. Depending on your business and the nature of your website, this list can provide information you can use to leverage key relationships. If a blog is generating a lot of traffic, perhaps there are ways you can continue the relationship with the blogger to create a brand advocate.  A brand advocate is a user or customer who tells their friends about your brand – for free. Cultivating these relationships is priceless. For more information about brand advocates, check out this article from Phil Mershon (2011).

Like in the search traffic section, each source can be drilled down to see bounce rate, time on site, pages per visit, etc. Some sources will have a high bounce rate, others will be low. The ones with lower bounce rates are the more effective sources. Visitors from these sources are engaging longer with your brand. Try to determine why they are so effective and if there are ways to increase traffic from these sources. It will be worth the effort.

Direct Traffic

This metric shows the percentage of visitors who arrived at the site directly – by typing in the URL, using a bookmark, or clicking on a link in an email, text or chat message. These visitors are already familiar with you, are usually returning visitors, and are most often your current customers (Kaushik, 2010, P. 86). This can be a good indicator of the strength of your brand or effectiveness of an email campaign (Roggio, 2011). Drilling down into the direct traffic will show the top landing pages on your site. Often times the homepage will be at the top, but not always. If an email campaign is running that includes a link to a specific page, that page could be a top performer. Or, if a virtual link is set up for advertising purposes, that page could be in the top three. If any of the top 10 pages are a surprise, it is worth some investigation to determine why users are landing on that page, and if it is generating a high bounce rate, etc.

References:

Kaushik, A. (2010). Web analytics 2.0: The art of online accountability & science of customer centricity. Indianapolis, IN: Wiley Publishing, Inc.

Mershon, P. (2011, June 13). 9 Reasons your company should use brand advocates: New research. Social Media Examiner. Retrieved from http://www.socialmediaexaminer.com/9-reasons-your-company-should-use-brand-advocates-new-research/

Roggio, A. (2011, July 18). Understanding ‘Traffic Sources’ in Google Analytics. Practical ecommerce. Retrieved from http://www.practicalecommerce.com/articles/2916-Understanding-Traffic-Sources-in-Google-Analytics

Start at the beginning

Web Analytics seems like a big, scary, mysterious thing - like the great and powerful Oz. Many companies don't know where to start when it comes to web metrics; they don't even know what questions to ask. Others know what to ask, but they are afraid because they fear the answers might be beyond their reach. So, they "pay no attention to the man behind the curtain" so-to-speak and maintain the status quo, never really moving forward or achieving their business goals (at least, not that they can track anyway). It's true that there are a lot of metrics and even more ways to apply them, but a basic understanding of key metrics can help a company evaluate their online presence and compare it to business goals and objectives.

So, where do you start? Like Glinda tells Dorothy before her journey down the yellow brick road, "It's always good to start at the beginning” (LeRoy & Fleming, 1939).

The yellow brick road starts with objectives. Without objectives, all the metrics and analysis in the world will mean nothing. For help in setting business objectives as they relate to web metrics and analytics, visit Avinash Kaushik’s blog. He offers great insights and step-by-step instructions for setting measurable goals and it's well worth the read.

Once objectives are established, it becomes easier to determine what metrics need to be measured; unique visitors, visitor loyalty, conversion rate, etc. But, before going in too deep, it’s important to understand the basic metrics measured by most analytics programs. Without an understanding of introductory metrics, it will be difficult to understand the more complex metrics and how to measure key performance indicators (KPI).

Google Analytics, a popular free analytics tool, displays the following metrics on the homepage:

Basic definitions:

Visits

The total number of visits, or sessions, started on the website.

Unique Visitors

The number of individual people who visited the website. If Mary visits the page five times and Tom visits the page once, there are six total visits and two unique visitors.

Page Views

The number of pages within the website that were viewed during the visits in the specified time period.

Pages/Visit

The average number of pages viewed during a visit. This is calculated by the number of page views and divided by the number of visits.

Avg. Visit Duration

The average amount of time spent on the website during each visit. The longer the better – it means that people are spending a lot of time consuming the content on your site.

Bounce Rate

The percentage of visitors who came to your site and left immediately. As Kaushik (2010) puts it, “They came. They puked. They left.”

% New Visits

The percentage of visits from visitors who have never been to your website before.

% Returning Visitor vs % New Visits

Shows the percentage of new visitors who have never been to the website before compared to the percent of returning visitors.

Of all of these basic metrics, a key one to review is bounce rate. The bounce rate helps determine where to focus your attention. In his article, Quinn (2011) says, “As a rule of thumb, a 50 percent bounce rate is average. If you surpass 60 percent, you should be concerned. If you're in excess of 80 percent, you've got a major problem.” Review the bounce rate of your website, landing pages, keywords, etc; it will reveal the areas that are not meeting expectations. It could be outdated content, technology errors or information that does not meet the needs of the visitor. It is imperative that the reason for the bounce is analyzed and corrected. It could take several tries to optimize the page - do not be afraid to experiment. If visitors aren’t sticking around, you won’t meet your objectives.

The path to the great and powerful Oz is a little dizzying, at least at first. But, taking small steps forward will eventually get you where you need to go. 

References:

Kaushik, A. (2010). Web analytics 2.0: The art of online accountability & science of customer centricity. Indianapolis, IN: Wiley Publishing, Inc.

LeRoy, M. (Producer) & Fleming, V. (Director). (1939, August 25). The Wizard of Oz [Motion Picture]. United Sates of America:  Metro-Goldwyn-Mayer (MGM)

Quinn, M. (2011, January 31). How to reduce your website’s bounce rate. Inc. Retrieved from http://www.inc.com/guides/2011/01/how-to-reduce-your-website-bounce-rate.html