Monday, July 1, 2013

Goals, Filters, & Funnels

“Every successful web metrics story starts with a very simple question: “Why do you exist? ... This question is asking why your business (or non-profit organization) exists. In a world where we are blessed with more data than God herself wanted us to have, it is often easy to skip past that question and just jump into reporting metrics. After all, metrics are so easy to find.” (Kaushik, 2010).

Kaushik, with his years of experience, has learned that for a business to be successful in its data collection and interpretation efforts it is important to “focus on a critical few metrics” in order to shape data interpretation and subsequently the conversation of the metrics with bosses and team members. Kaushik warns that “you shouldn’t have more than three of four critical metrics” (2012). If you have more, he says, then you are not doing your job well.

It is important to keep the above information in mind when beginning to look at goals towards conversion on a given site. This post will focus on creating Goals in Google Analytics. It will also look briefly at Funnels and Filters.

Goals


I explored the Goal functionality of Google Analytics this past week and discovered some of the key functionality. When setting up Goals on Google Analytics, there are 4 different of types of Goals a user can create: Destination, Duration, Pages/Screens per visit, and Events. In this post I will discuss the two I tested on my site: Destination Goals & Duration Goals. 


 1. URL Destination Goals


This function allows you to set goals around a specific page with a unique URI that you want your visitors to visit. For example, on my page, I wanted to test how many people I could direct to my latest post—a post I made about the use of “filter bubbles” by companies like Google and Facebook.  To do this, I set up a Destination type Goal and set “/2013/07/what-google-and-facebook-are-hiding.html,” as the URI of my post as the destination.  


By setting this up as a goal, I can now see how many people who visit the site go to that specific page. Here is the result after driving some traffic to the site through social media posts:


As you can see, the total number of completions for this goal was 9. That created a conversion rate of 18.75%.

Since the benefit of this for my blog is not that obvious, let us look at an example of where a Destination Goal would be considered a key metric:

Let me pretend that my site sold cat toys and related products. For the month of May, I have an overstock of 5 items that I want to increase sales of. To do this, I create a special page called “May Specials—35% off!” As a website owner, I can create a goal to drive at least 30% of the website visitors to at least visit this page. Each week, I test a different visual strategy to draw attention to the tab: In the first week, I use arrows to point to the tab. In the second week, I use highlighted colors. In the third week, I use a colorful banner. In the fourth week, I use a pop up.

By analyzing the goal completion rate, I can see which approach worked best to drive traffic to the May Specials URI. I can also look at the Traffic Flow to see the path visitors took to reach the May Specials tab in order to better understand Drop Off. By constantly comparing different approaches to the goal rates, I can refine optimization strategies and better understand visitor behavior.

2. Duration

A Duration Goal allows you to set a certain time limit that you want visitors to spend on your site or a page on your site. For example, on my site this week, I set a goal to have viewers spend 30 seconds or more on my site. One of the problems I noticed in the beginning of my installation of GA is that the engagement on my site was extremely low. Since the site is informational in nature, I wanted to understand what percentage of visitors were actually logging on and reading the content. Here are my results after some promotional efforts on social media sites:


As you can see, I had a total of 5 visitors stay on for 30 seconds of more. That results is a conversion rate of 10.42%. I can also see what specific posts were generating the goals. In this instance, the “What Google and Facebook are Hiding” post drove the highest number of goal completions.

The benefits of this metric are a little more obvious, even for an informational site like mine. In general, we want visitors to stay longer and engage with the site at a deeper level. By setting specific duration goals, we can see what percentage of visitors are sticking around for a desirable amount of time.

Funnels

A funnel represents the path you expect visitors to take on their way to converting to the goal.” (P.I. Reed School of Journalism, 2013).

Funnels help data managers understand how often visitors abandon goals and the paths they take to arrive at the goal or the path they took after they left the goal. Specifically for e-commerce websites, understanding funnels can be of great importance. For example, they inform continued optimization to check-out processes to increase final purchase and conversion rates.

Filters

“…filters can be set up to exclude visits from particular IP addresses, to report only on a subdomain or directory, or to take dynamic page URLs and convert them into readable text strings. (P.I. Reed School of Journalism, 2013).

A filter allows for inclusion or exclusion of certain sources of traffic. For example, a filter can improve the accuracy of a reporting by excluding certain IP address. To give an example, a filter can be set up to exclude information about a website tester, as he or she would be creating lots of traffic to a site that would affect conversion rates and statistics if counted as a regular visitor.


Conclusion:

 Goals, funnels, and filters help create data that can be understood on a mico-level. By isolating certain metrics that have been defined as key metrics for your website, analyzing them can inform effective and efficient optimization. In short, goals, filters, and funnels allow for targeted analysis, which can be used to understand conversion successes or failures. 


What Google and Facebook are Hiding from the World

The web is moving to one built around personalization, but is that really a good thing?

Eli Paris gave a TED Talk about the issue and sees this move in the web as a major threat to democracy and the very idea of a free and open web. He describes the problem as this:

"As web companies strive to tailor their services (including news and search results) to our personal tastes, there's a dangerous unintended consequence: We get trapped in a "filter bubble" and don't get exposed to information that could challenge or broaden our worldview. (Pariser, 2011).

Watch the talk here: 




In his talk, Pariser shares how two identical searches for Egypt brought totally different results. One searcher's results brought politically oriented news; the other's brought travel information and general information. 

Pariser's argument is that personalization brings us what we want to see, but not necessarily what we need to see. 

Do you agree or disagree? Comment below. 

Monday, June 24, 2013

Diving into Google Analytics


About one week ago, I successfully installed Google Analytics on my blog and have approached promoting it in two ways:

1. I directly asked friends to visit the site by sharing the link with them over the chatting service Google Hangouts.

2. I posted a link on my Facebook page requesting friends to visit the site.

Given my two promotional efforts, the data generated by my analytics was not surprising. I explored a number of areas, but in this post I will cover Traffic Sources, Demographics, Technology, and Visitors Flow.

Traffic Sources:

Google Analytics tracks what sources are driving traffic to a specific website. In the case of my website, I can see that Facebook was the largest source of traffic; direct traffic was the second largest; Blogger referrals were the third largest and mobile Facebook referrals were the fourth largest source. The results were not surprising as they more or less mirrored my promotional efforts.
In addition to just basic information on where the visitors came from, Google Analytics also provides information on pages viewed per visit as well as the average visit duration, percentage of new visits, and the bounce rate. The metrics show me that for the people who visited from Facebook, they almost all left immediately; this was probably because I posted, along with the link, that I only needed friends to visit the site for traffic purposes. It would have been a better tactic to arouse interest in the site on Facebook by providing information about the site and subsequently attracting readers who would be interested in the content of the blog. The direct traffic average visit time was longer, but overall, hardly anyone spent any time on the site at all.

In a professional setting, the information provided in the Traffic Sources section can be extremely valuable. The source visits help marketing managers identify what communication campaigns are working and what campaigns are less effective. The overall number of visitors compared with the average visit duration and bounce rate can confirm or raise questions on the quality of the traffic. For example, if a source referred a lot of visitors but none of them stayed for long, then that source is obviously attracting the wrong type of visitor to the site.

Demographics: Location
Another web metric that I found particularly interesting is the location data under the demographics section. Google Analytics provides location data that can be broken down into four dimensions: Country/Territory, City, Continent, and Sub-Continent Region. While I was familiar with Google’s ability to provide data on country (I was familiar with this by exploring the analytics section of YouTube on my channel page) I did not know that GA also provides information broken down by city.

The business application of knowing where website visitors are coming from can be helpful for a number of reasons. To start, it provides a big picture overview of where visitors are coming from. This data could help drive investments decisions on whether to invest in local language versions of a website for countries that are large sources of traffic. The ability to break down the data by city can measure the effectiveness of marketing efforts in certain areas. To give an example, a company could measure the results of a one-time marketing campaign at a large event in New York City.

At the top of the page, there is a tab to switch between site usage and ecommerce. By switching to ecommerce, the map will provide information on how the website’s ecommerce activity is divided up by location. Depending on a company’s objectives for being on the web (i.e. informational vs. commercial), knowing how each geographic area contributes from a financial standpoint is likely more important than just knowing the web usage statistics.

Technology
Google Analytics provides statistics on various metrics of the technology used to access a site. To start, GA gives a breakdown of the different operating systems used by visitors. The browser data can be used to decide where to allocate resources for web browser optimization. If, for example, the majority of web activity is coming from Chrome, Chrome optimization will be of high-priority in the update and testing cycle.  

In addition to browser information, GA provides statistics on the visitors’ operating system, screen resolution, screen colors, Flash version, and Java Support. These statistics are useful for similar reasons as the knowing what browser visitors are using to arrive at a site: they should be taken into considering when optimizing the site to make sure it accessible to all visitors. In particular, the Flash version would be important to know for each visitor, as efforts should be made to have Flash content be accessible to the largest group of people possible (if it is crucial to the webpage experience).  

Visitors Flow
The Visitor Flow metric was not particularly interesting for my website, as the website is rather simplistic in nature and the engagement of my visitors to date has been extremely low. However, for more complex sites with clear conversion goals, I see this metric and visual overview as being both interesting and helpful.
The Visitor Flow provides a visual overview of exactly how visitors are engaging with the site by providing marked drop points and visual paths showing where visitors are going on the site. The paths can be broken down by country, browser, traffic source, and just about any other metric.

An application of Visitor Flow is that it can allow insight in to how or why visitors are not converting in a desired manner. For example, it is possible to break down the Visitor Flow to show paid search traffic, by social media sites, by just advertising, by landing pages, and more (Kerschbaum, 2013). By examining how visitors from certain metrics flow through the site, managers have better insight on why the visitors are or are not doing what they are supposed to be doing. Additionally, Visitor Flow can reveal that different sources bring different types of traffic. For example, a visitor from a mobile device might interact with the site different than a visitor from a desktop device. Additionally, a visitor from a pay per click ad might interact with the site differently than a visitor who arrived at the site through a social media link. 

Monday, June 10, 2013

Google Adwords and Facebook Ads

Google Adwords




Advertisers have always wanted to be able to pitch a product or service with the right offer, to the right person, at the right time.

Google adwords attempts to bring this long-sought-out functionality to marketers: it provides highly-targeted messages delivered to users based on what those users are specifically looking for at that exact moment.

The platform is further compelling because of its user-directed nature, allowing buyers to set their budget, location, and keywords. Results are then displayed on a dashboard interface which can include a number of metrics, including the total amount of sales generated by a specific keyword. Campaign managers can modify campaign specifics at any time.

An example of how a marketing campaign based on adwords can be effective is seen in a case study done by the InterContinental Hotel Group that was written about a Marketing Week article:

“IHG created mobile-based AdWords campaigns for each hotel brand for localised sites in the UK, France, Germany and Italy. These campaigns contained unique messaging, which included phrases such as 'Holiday Inn Hotels', 'Mobile site - book now' and 'Call or book direct on mobile site'. De Rosa says: "It seemed clear that to maximise traffic, we needed mobile-specific ad text for the different hotels worldwide."
The adoption of unique messaging delivered great results. "By using text specifically aimed at mobile users, we saw revenue from our mobile search activity increase by 91% year on year," says De Rosa. Further analysis showed the UK produced the best results, closely followed by Germany, France and Italy. De Rosa adds: "Thanks to our current activity with Google, traffic to our mobile site is increasing by 20% month on month." (2010)

Google is the second most popular website in the world after Facebook (Alexa: top sites).  For this reason, its reach is one of the largest available to marketers. Furthermore, with search at the core of Google’s business, it will continue to refine its understanding of its users. In a Wired article, author Steven Levy wrote this of Google:

“The microeconomics of Google is more complicated. Selling ads doesn't generate only profits; it also generates torrents of data about users' tastes and habits, data that Google then sifts and processes in order to predict future consumer behavior, find ways to improve its products, and sell more ads. This is the heart and soul of Googlenomics. It's a system of constant self-analysis: a data-fueled feedback loop that defines not only Google's future but the future of anyone who does business online.” (2009)


  
Facebook


Facebook is currently the top ranked site in the world according to Alexa Rankings (Alexa: top sites). This means that for any marketer, it is an important site to understand and use effectively in order to reach your audience.

Facebook’s main advertising tool is Facebook Ads, display ads that can target users based on demographic and psychographic information.

Facebook Ads offer a number of benefits compared to Google adwords: they are visual, they can target psychographic profiles in addition to demographic profiles, and users simply spend more time on Facebook, allowing the advertised brand to get more exposure. The Facebook Ad selector tool will also inform campaign members of the number of users who will be reached through a certain campaign. To managers then have the option to either pay per click or pay for views.

To compare and contrast, Facebook and Google offer two fundamentally different approaches to reach an audience. Google focuses on specific intent of a user at a specific time. The ads are better targeted because a user is specifically looking for that information at that time. To give an example, a user might search for “Vacation rentals in Malibu” on 5 pm on a Sunday night. The ad for a service offering vacation rentals will be displayed to the person at that time. In this situation, Facebook, on the other hand, can only offer advertisers profiles of people who “like travel” in their hobbies and fit the demographic and psychographic profile of someone who might possibly be interested in traveling to Malibu and staying in a vacation rental. That being said, something like a buy one get one free coupon for a Burger King burger would likely be better on Facebook, as the target audience for that kind of promotion would not necessarily be people searching for fast food coupons. 

Content vs. Conversation

For marketers seeking success in the digital marketing era, there are any given number of strategies one could be convinced of and follow. This can be a daunting task considering once a plan is conceived, mapped out, and executed, a lot of energy has gone into one strategy and changing the approach is more-often-than-not a challenging task. Adding fuel to the fire is the fact that modern marketing strategies—i.e. content marketing—don’t always have an obvious connection to sales, so it can be hard to understand how strategies are working or not working.

Content is King
In the readings from this week, the debate between building efforts around content compared to building efforts around conversation arose. Michael Greenberg wrote, in an article titled Content is King of Social Marketing, that…

social marketing efforts need to be driven by content, not vice versa. Without content, there is not a whole lot to talk about.” (2009) He goes on to add that “More than anything, you have to think of yourself as a publisher…The reality is, if you aren’t already generating a fair amount of content, it will be difficult to keep up any sort of cadence and maintain your content promise. Plus, most e-commerce marketers’ websites are specifically geared towards conversion.”
Greenberg here makes a valid point: at the heart of a social campaign, content is king.

Conversation is King
Catherine Novak in a related article offering a different point of view argues that a campaign built around content is misguided.

Content without conversation is just broadcasting, or just advertising.  It goes to the listener/reader/viewer/visitor… and stops there.  If the sender is lucky, it may lodge as a piece of information in the receiver’s consciousness, and they may act on it someday.  If the sender is luckier, or perhaps more engaging, it may be something that the receiver wants to talk about.  And then the message gets a whole new burst of energy.  The energy behind the message is what gives it meaning, and a life of its own.  That happens because we humans like to communicate with each other.  Thus the conversation begins.” (Novak, 2010).
In her article, Novak goes on to compare content strategies without conversation as impersonal and more commercial in feel, making them resemble traditional media as opposed to being properly social.  

Who is right?
There are, of course, many truths embedded in both Novak’s and Greenberg’s articles: content can lead to great conversation, and if executed improperly--as Novak suggests--content strategies can fail to have the desired effect on the audience.

The way I see it, however, is that focusing on great content is the key to winning the social media marketing game. To be honest, we cannot expect users to sit around on social media sites chatting about products or a content-related article; that is simply not what they are there to do. We can, however, expect them to share interesting and relevant information with friends that will spark interesting conversation offline, be disseminated by the users themselves, and ultimately build brand awareness.

One example of how a well-executed content strategy can play out is Mint.com:

Mint owes much of its user adoption and brand success to its aggressively intelligent content strategy. Unlike the half-hearted, months-between-updates blogs that most businesses keep, Mint’s blog “MintLife” was a core part of the company’s operation.

Mint dedicated significant resources to its blog, including a full time editorial staff and a slew of freelance contributors. It invested time in social news sites like Reddit and Digg, and after months of seeing consistent, quality Mint content, readers in those communities began trusting Mint as high quality, reliable, and cool to share. Eventually, those users turned into Mint customers, even advocating Mint in their personal networks.
News and tips posts, link roundups, slideshows, videos, and infographics were all key components in Mint’s content strategy, and they were held to a strict editorial standard. By establishing itself as a smart resource that was easy and accessible to the financially curious, Mint was able to leverage its content credibility to convert readers into buyers of its actual product.” (Snow, 2011)

The point is that Mint, a seller of personal finance products, was able to build its credibility and brand by focusing on having high-quality content. This helped build trust in the brand, ultimately resulting in more users of its products. 

Monday, June 3, 2013

Exploring Landing Pages


Some simple questions marketing managers interested in web metrics might ask:

By looking at overall statistics for my website, how would I know if a specific campaign is successful or not? 

How can I make sure that driving traffic to my website will lead to good conversion rates for my business objectives?

One possible answer to both of these questions: create good landing pages.
Pamela Vaughn, in an article titled Why Landing Pages are an Indispensable Part of Marketing wrote that “A landing page is a web page that allows you to capture a visitor's information through a lead-capture form (AKA a conversion form)…A good landing page will target a particular audience, such as traffic from an email campaign promoting a particular ebook, or visitors who click on a pay-per-click ad promoting your webinar. You can build landing pages that allow visitors to download your content offers (ebooks, whitepapers, webinars, etc.), or redeem other marketing offers such as free trials, demos, or coupons for your product. Creating landing pages allows you to target your audience, offer them something of value, and convert a higher percentage of your visitors into leads, while also capturing information about who they are and what they've converted on.
Let’s look at how creating a landing page could be used as part of an overall SEM strategy:
To illustrate how a landing page can be used effectively in part of a digital marketing campaign, let’s take the example of a fictitious company that sells investment advice to personal investors.
One of the organizational objectives of the site is to capture leads (names and email addresses) in order to promote their various investment services to buyers through their email list. One of the company’s strategies to accomplish this is to offer free analyst reports of popular stocks.
For this case, let’s say the company has created an in-depth report looking at the high-volume stock of Apple Computers, ticker symbol AAPL.
The company decides to run PPC (Pay Per Click) campaigns on popular search site Google for keywords related to AAPL (e.g. “is AAPL a buy,” “should I sell AAPL,” “AAPL analysis,” etc.). The PPC ad copy reads “Should you buy or sell AAPL?” FREE In-Depth Analyst Report on AAPL.
When visitor click on the ad, they are taken to a landing page that has a summary of the report and a bio of the analyst. The visitor can then use a form to sign up to have the report digitally delivered to his or her email.
It is easy to see how by then using additional web metrics tools on the landing page the company can see of the effectiveness of this specific campaign. For example, the company can run various PPC keyword ads and then see which ones are the most effective and optimize around them. Additionally, once viewers are on the site, they can adjust the copy and format to test which content and format results in the highest conversions.

In short, having very specific landing pages tied to PPC campaigns and other marketing efforts can help create communications efforts that, coupled with web metrics, are easy to measure and control. This is at the heart of what digital marketing is good at: highly targeted communication efforts that can be tested, measured, and optimized. 

Web Metrics: The Basics


As this blog will focus on search engine optimization, web metrics, and related topics, it is logical to start at the beginning. In this post, a simple question will be explored:

What are web metrics and why should people attention to them?

In short, web metrics allow individuals or organizations with web properties to understand more about what is happening when users visit the sites. While in no way comprehensive, the video below produced Volusion—a company that specializes in solutions for ecommerce sites—outlines five basic web metrics that everyone should know:



Using these five metrics as a framework, we can dive deeper into each one to help form a general understanding of some of the core web metrics that make up web analytics.

1. Visits:
Number of Visits is simply that: how many visitors—new and returning—are coming to a particular site. In isolation, this metric can help site managers understand a very big picture view of a website’s popularity over time. It can also help companies identify cyclical trends and compare the traffic in a given week or month against data from previous periods in a website’s history. For example, a site might find that certain periods of the year bring in a high volume of traffic than others. Site managers might also be able to show how a website has increased—or decreased—in popularity over a certain period of time.

2. Page Views:
Page Views, as the name suggests, tracks the number of times visitors who show up to a certain page on a website. The benefit here is that site owners can understand which pages on their website are the most popular. Depending on the business objective of a website, this can be very insightful. For example, let’s say a site is focused on offering investment advice for young professionals. The Page Views metric will help managers identify content and format trends that their visitors are the most interested in.

3. Traffic Sources:
The Traffic Sources metric tracks for where visitors to a site are coming from: Did visitors come directly to the site by typing in the web address? Did they come from a Google search? Did they come from a link in an ad or Internet video? Did they come from a news article about the company?
Knowing the answers to these questions can help, of course, track the effectiveness of communication campaign efforts as well as observe sources of traffic that grew organically.

4. Referring Keywords:
Referring Keywords are the specific key words that are directing traffic to your website. Let’s again look at the example of an investment help site for young professionals. This metric might reveal that “how to invest in stocks” and “growth stocks” are two keyword combinations that are driving traffic to the site in high numbers. The site owners can then optimize for those keywords in order to result in even more traffic.

5. Bounce Rate:
The Bounce Rate shows how many visitors are coming to the site, looking at just one page, and then leaving.According to Google Analytics, “There are a number of factors that contribute to your bounce rate. For example, visitors might leave your site from the entrance page if there are site design or usability issues. Alternatively, visitors might also leave the site after viewing a single page if they've found the information they need on that one page, and had no need or interest in visiting other pages.”

Naturally, as with any thoughtful analysis, all of these metrics mean nothing without first defining a clear set of business objectives and then using web metrics as a way to answer important questions you might have about the performance of a web property. Jim Stern, a management consultant and evangelist for web analytics, stated in an insightful interview that web analytics allows us to answer key questions: “How many people show up to your website? What do they do when they get there? Do they accomplish their goals? Do they accomplish your goals?” 


The benefit to using and understanding web metrics is really quite simple to understand: web metrics can help individuals and organizations understand what is going on with their web properties. This information can then be communicated and acted upon, resulting in increased accomplishment of business objectives.