Ongoing debates surrounding the topic of “free news versus digital media pay walls” present an interesting opportunity for web analytics professionals.  If there’s one thing I learned during my stint in new media, it’s that news organizations have historically had a poor grasp on establishing sustainable revenue models to support substantial distribution costs and sizable staff counts.  I say “historically” because savvy media companies are becoming increasingly capable of correlating content consumption to revenue streams and optimizing their advertising models for greater ROI.  The latter have been forced to cut back on content acquisition or workforce, and even now consider the use of pay walls.  But that’s not necessarily a bad thing! For analysts, pay walls introduce a great opportunity to test the efficacy of both models.

Model 1: Ad-supported content

Needless to say, there are significant challenges facing media sites that operate on ad-supported content.  Recent months have given rise to the worst decline in media jobs the industry has seen in it’s history.  News rooms are cut to the bone, local affiliate stations are closing their doors, the public in general is changing the way it consumes the news — but I digress.  The main problem ad-supported sites face is correlating visitor engagement to true ROI:

• Integration of ad-serving platforms with web analytics systems may not be straightforward
• Costs associated with ad serving platforms are minimal and relatively fixed/known, but custom advertising sponsorship deals have many moving parts
• Publisher and advertiser goals may be in conflict, i.e. visitor engagement with a media site sells ads but visitor engagement with an advertiser reduces engagement
• Calculating true ROI involves knowing the operating costs of the content provided, including but not limited to manpower for article creation, multimedia distribution rights, bandwidth, etc.

Clearly, savvy media sites that are able to manage the complexities of calculating and optimizing user experiences and maintain positive net ROI can make this model work, however it takes a brilliant analytics ninja to implement the myriad of data required to make it truly sustainable.  Avinash’s 90/10 rule might have to be magnified a bit… maybe 98/2 would work?

Model 2: Pay walls

The reason pay walls are such an appealing model for large media sites is that they can reintroduce a subscription model akin to selling newspapers, magazines, and specialty channels on TV.  Logically, using a pay wall follows the age-old business model that built large media empires.  In addition, pay walls provide a number of measurement opportunities that seldom exist on media sites:

• Actual direct revenue numbers from paying customers, and optimizing landing pages and user experience based on dollars and cents. I.e. rather than loosely correlating advertising dollars to visitors, getting real-deal dollar numbers on specific users.
• Measuring distinct user behavior and profiling and recommending packages or custom subscriptions based on user preference or actual consumption. I.e. the iTunes approach; selling content per song, per episode, or per movie.
• Engagement essentially becomes less ambiguous, users are authenticated, they pay for access to certain assets, and a relationship is established.  Data comes to the rescue and can assist in decision making to establish and nurture that engagement.  I.e. visitors now have skin in the game and a voice.

Regardless of the approach, it appears as though several key media destinations will carry-out testing some kind of pay wall model this year.  It’s uncertain what they will be measuring and how sophisticated the approach to maximizing ROI will be, but given this approach, the barriers to testing should be significantly lower, if and only if these same companies abandon ulterior motives and truly adopt data-driven decision-making.

The key is that there is no key.  There is no right way, there is no magical bullet.  Each site’s audience will be different, each traffic segment will be different.

DY62PTM2278U

An unfortunate recent statistic unearthed from the Factual Blog has shown that less than 30% of businesses worldwide are using Google Analytics on their websites out of a 4 million site sample set (via Manoj Jasra at Web Analytics World Blog).  Unfortunate only because there seems to be a booming interest in search engine marketing, namely search engine optimization.  By no means does one need Google Analytics to measure indicators of success in SEO, but it sure helps.  Any web analytics tool should top your list of priorities when embarking in online marketing, but there are lots of other resources freely available to webmasters.

Web Analytics: Google Analytics or Yahoo Web Analytics

As mentioned already, the focus should be to start collecting web analytics from day one in order to benchmark any optimization effort.  In fact, collecting historical data a few weeks or a few months prior to optimization efforts can highlight key seasonality trends that you may mistake for successes or failures in website changes.  For example, did changes to your site enable it to start ranking for a keyword, or is it because you always ranked for the keyword?  You may not necessarily know for sure without historical data, but this shouldn’t be a stalling point.  A good guy (Avinash Kaushik) wrote a great article on search analytics, which is a super resource to answer common questions about measuring SEO efforts, I recommend giving it a read as well.

Metrics to watch:

• Visits, time spent on page, and bounce rate by top entry page, segmented by Natural/Organic Search)
• Reverse goal funnel, segmented by Natural/Organic Search
• And to find a hint at visitor intent, top internal search queries segmented by Natural/Organic search
• Visits by non-paid keywords, but filtered for branding, trademarked, or copyrighted terms.  It’s relatively easy to rank for keywords in your company’s name, domain name, or product names, but it’s often those long-tail generic terms you are optimizing.  Remove anything else that will contribute to noise.
• Google Analytics “Intelligence” is a new report that does significance testing for you.  Use it to quickly identify and build segments related to significant changes in visitor behavior.

Google Webmaster Tools

Let’s face the facts, according to recent studies from various sources, the majority of search traffic still originates from Google, on the order of 60-70%.  Google Webmaster Tools gives a ton of insight into which keywords your website ranks in results for, as well as which keywords are clicked on.  Keep in mind, Google tests rankings on search engine results pages (SERPs) and weighs long-term ranking through several factors, one of which being click-through rate and bounce rate.  So by keeping an eye on the keywords your website is ranking for, but not generating click-throughs for, might let you shift optimization efforts accordingly.  Again, Google Webmaster Tools is totally free to use, and easy to setup.

Metrics to watch:

• Number of keywords triggering results for your site
• Position of your site in the search results for specific keywords
• Number of clicks to your site from keywords
• Position of your site in the search results for specific keywords clicked
• Diagnostic crawl stats can also provide insight into increasing, decreasing, or unchanged crawler activity to your site, which should be taken with a grain of salt

Competitive Intelligence

There are a multitude of competitive intelligence tools, free and paid, available online.  Due to sample collection and sample size, each tool has it’s pros and cons in terms of accuracy and precision, but they’re helpful nonetheless.  Free competitive intelligence tools include Alexa, Compete.com (free reports), and Google Trends. Paid tools include Compete.com, Hitwise, ComScore, Nielsen, and a host of other premium services.

Metrics to watch:

• When comparing two sites side-by-side in Google Trends, for example, it’s important to avoid getting wrapped up in the actual values attributed to metrics, but focus on the trending of page views, unique visitors, etc.
• Compare visitors referred from external sites to two or more competitors, but filter out large referrers such as search engines (premium tools such as Hitwise and Compete.com offer this functionality) to increase the resolution of those niche sites your links aren’t reaching.

Link Intelligence

Aside from competitive intelligence, which provides insight into which sites are sending traffic to you and competitors, consider link intelligence as well.  Search engines still value links on other websites as a ranking factor in their algorithm, so even if a link does not provide any traffic, it may still provide a boost in page rank (or SEO juice if you subscribe to “page rank” being dead).  Obviously there is no shortage of sites that track links between sites, however one great resource that is currently high on my list is Majestic SEO.

Metrics to watch:

• Back-link discovery trending graphs can provide insight into how your link-building efforts are assisting page rank.
• External backlinks and referring domains are also useful, but if links to a website are on each and every page in an external site’s template, trending these numbers could be meaningless.  Instead, consider focusing on external back-links PER referring domain.  The goal here is to ensure the ratio doesn’t become inflated with links to your homepage, but links embedded in actual content that’s relevant to deep pages on your site.

We truly live in an age of data overload. By no means is the above list exhaustive, as these tools and services are only a minuscule subset of resources available to marketers and web analysts.  Feel free to share your favorites below!

Divorcing yourself from opinion, or qualitative analysis, or wishy-washy suggestions can be a tough task.  Some stakeholders are statistically savvy and can look at numbers objectively, opting to do their own analysis – perhaps by using separate business intelligence. However, most people want to read interesting insights, even if it’s just to look smart at the water cooler.  For everyone to get on board, the trick is to stop saying, “I think” and start saying “the data shows.”

Finding valuable insight in segments of your traffic is a great first step.  Focus on one product, one traffic segment, and one conversion pathway through a site.  Slice and dice the traffic different ways until you notice a trend, or even something interesting.  You might find that first time visitors coming through organic search terms convert five times better than paid search visitors.  A perfect starting point for deeper dives across alternate product lines.

Oh oh, what if your data is sketchy?

What if your segments look a lot alike, and you think that 5% difference could mean something but aren’t sure?  That’s where significance testing and maybe even multivariate testing can help:

• If you DON’T have access to web development teams, test theories using a statistical significance test, such as PRC’s statistical significance calculator.
• If you DO have access to a friendly web dev, consider working together to run an A/B or multivariate test using Google Website Optimizer.  Using it is free, significance testing built-in.

All of a sudden, the data starts showing room for improvement.  Creativity shifts to testing to find a winning combination, instead of intuition- or experience-based decision-making.  Perhaps for the first time ever, you can start attributing measurable difference in revenue and start prioritizing projects by potential shifts in “ROI” (using the term liberally here).

In addition, the data analyst can save measurable results to their war chest.  A bit different than intuition or experience, the war chest becomes a repository of educated guesses.  As you proceed to make improvements in ROI, you can start to reliably predict lost revenue for projects that are put on the back-burner.  Data (not the Star Trek android) becomes your friend; an ally in the fight (ok, tone that down to “challenge”) in development resources.

A new ally in data, new friends in web development, new supporters across business units, new weapons in your war chest.  Becoming a data diplomat is win-win-win.

Return on investment is a sexy KPI right now; everything and everyone claims to provide easy calculations of ROI, value added services that show ROI, and other far-fetched promises.  Probably one of the hardest things a web analyst can do is measure ROI — true ROI.  We’ve become accustomed to using the term “return on investment” to include subjective value propositions; soft and cuddly marketing statements.  True ROI is just not available through analytics tools and web analysts do not have the luxury of full financial disclosure.

Unfortunately, as data-driven decision-making has become more popular, so too has the search for more meaningful business-oriented metrics.  You know, stuff that actually means something to say, your CFO — money.  Somewhere along the way, we lost sight of the true definition of ROI; people have been bastardizing the term ever since.  Let’s look to Wikipedia for a refresher:

Rate of return, or return on investment is the ratio of money gained or lost (whether realized or unrealized) on an investment relative to the amount of money invested.

What’s missing in this definition?  Oh right, all those wonderful value-add statements marketers (myself included) use to inspire confidence in products and services.  Statements describing potential positive ROI should never include subjective benefits, because that’s not ROI.  With respect to web analytics, the same is true.

Regardless of how much data you collect using your web analytics weapon of choice, you’re not seeing the whole financial picture.  If I had to wager my salary towards the truth of that statement, I would probably win 99% of the time.  Web analysts simply aren’t plugged into each and every profit and loss source within an organization, and frankly we shouldn’t be.

The typical web analyst is a highly detail-oriented individual, that includes figures in context of what is being observed online.  So we have to ditch the use of ROI as a KPI and look to more specific calculated metrics that make sense, such as return on ad spend (ROAS), average revenue per order/user (ARPO/ARPU), and maybe just aggregate revenue figures.

Does that mean we should live a sheltered life?  Hell no.

There are wonderful departments known as finance, with highly skilled numbers people just like us that have access to a heck of a lot more business information than we do.  I encourage you to reach out to these financial gurus and ask a ton of questions.  Don’t expect each and every profit and loss center in the company, but if you’re lucky, they might just give you an aggregate ratio to aim for to determine a winning versus losing product, campaign, or change initiative.

Happy New Year and welcome back to what will be a make or break quarter for many of us. It’s time to hunker down and turn what was a pretty terrible economic run last year, into the building blocks of recovery.  Many of us will have to focus our efforts on hitting hard in early 2010, gaining an important early foothold amongst strong competition.  That makes measuring continued success so much more difficult, especially if you haven’t yet established clear KPI’s on which efforts will be judged.  So with all the vague “just measure ROI” discussion aside, I’m predicting the following to be the most important key performance indicators for early 2010.

Why the hate geared towards “ROI” as a metric?  Simple, it’s pretty obvious that everyone brandishing the term about does not use it appropriately.  When most people say “ROI” they don’t mean to take the conversation up to the next level, and don’t have research to back it up.  Measuring true ROI can be pretty difficult, especially without tag-teaming efforts between a web analyst and financial controller/comptroller.  So let’s all put the term “ROI” aside for now, and refocus on a term someone smarter than me used once upon a time, “business impact”.

Response or Conversion Rate

The perennial favorite!

We’ve all been measuring conversion rate for years, and that’s not likely to change anytime soon.  However, what will change is the attribution any one department or campaign or strategy is awarded when a response or conversion is achieved.  Departments do not live in a bubble, be prepared to share and share alike as divisional lines begin to blur.  Marketing folks will have to talk to sales folks, support folks will have to talk to IT folks, and everyone will want a piece of the organization’s success metrics, regardless of who reports on those KPI’s.

Social Influence

You can’t ignore social media anymore.  To web analysts, social media is just another platform that can be measured independently using external sources, or analyzed with software by segmentation.  Avinash Kaushik goes into greater detail on this subject, and while I agree his equation of influence on Twitter is useful, “influence = RT’s per 1000 followers”, a greater holistic metric is needed that can be applied to all other social media platforms.

Why not blend external and internal metrics?  Do they blend?  They sure do!

Consider segmenting social media sources as a group, and apply ball-park cost figures to the time you invest on social platforms versus lead or order volume to your site.  In this manner, you can measure social media cost per acquisition (CPA) the same way you would banner ads, paid/organic search traffic, and email campaign success.

Voice of Consumer

As we delve deeper into advanced competitive intelligence reports from sources such as Compete.com, Hitwise, and even Google Trends, we can start measuring the impact of external traffic patterns, get additional context to put other metrics in perspective, and establish checks and balances for branding efforts.  While this is great for marketing departments, it doesn’t do much to help other folks in the company.  Enter your new favorite tool for 2010: the survey.

You may be aware of vendors that offer survey solutions for abandoned shopping carts, or as a last-ditch effort to people leaving a site, but what about using surveys elsewhere.  Partnering with research houses can give you the best measure of voice of consumer because they go beyond the scope of your site, include all your competitors, and normalize the data.

Not the surveying type?  Depending on how sophisticated your analytics solution is, you may be able to relate segments of authenticated users to your CRM or billing software.  By applying external competitive intelligence metrics to internal web analytics, and then web analytics to real customer information, you can provide ample insight to customer services, sales, and support teams.