A little over a year ago, I wrote a blog post about why you shouldn’t freak out about people opt-ing out of being tracked in Google Analytics.  Yesterday, Google announced that they are using SSL to encrypt search queries and responses for people that are using Google.com and are logged into their Google accounts.  The result of this change is that referrals from Google organic search in Google Analytics and other clickstream measurement tools will not be able to determine the keywords used in the search that brought a user to the studied site.  The reporting of the fact that the visit was referred by Google organic search will be maintained.

Although this has widespread implications for both Search Engine Optimization and site optimization activities, I’d encourage you to not freak out about this change for a lot of same reasons that I outlined last year.

Why I’m Not Freaking Out – And You Shouldn’t Either

I have no idea what to expect with this change – there is no way for me to predict how many of my sites’ visitors are going to be coming from logged-in Google account holders.  I do know the current impact of Google organic search to my portfolio of sites – it is the single biggest driver of organic (as opposed to paid) search traffic, providing, on average, 75% of visits. According to both Comscore and Hitwise, Google had a 66% market share of U.S. searches in September 2011.

This traffic is important to the overall goals of all of our sites.  We spend a lot of time and effort on building and maintaining our sites’ search traffic and this change has serious implications for both the quality and quantity of data that we use to drive these efforts.  Despite this, I’m still not freaking out.

Using the same thought exercise as my previous post, imagine if 50% of Google’s organic traffic had its keywords obfuscated because those visitors were logged into their Google accounts when they performed the search that ultimately brought them to your site:

It (Still) Isn’t about the Individual Visit (or in This Case, Search) – Search data, like all other clickstream data, is useful in aggregate.  By looking at the keywords searched and grouping them by theme, we can calculate the value of certain types of keywords vs. others (for example, brand terms vs. product terms).  Losing 50% of this data will affect the number of Long Tail keywords that get reported and potentially affect the reporting of narrow segments of traffic that have few reported keywords, but not radically affect our conclusions on the aggregate “search intent” of our visitors.

Aggregate Data is More about Precision than Accuracy - With this thought exercise, we are again losing a bit of Accuracy without losing Precision.  There is no reason to suspect that individuals search and post-search behavior is going to change because they happen to be logged into their Google account.  The assumption that these individuals as a group do not significantly behave differently than all Google organic visitors is one that is easily tested within Google Analytics by comparing the group with reported keywords with the one whose search terms are obfuscated.  Because of this (testable) assumption, we can draw conclusions that the “search intent” of the missing 50% in aggregate is going to be similar to the fully-reported 50%.

Perhaps most importantly, from a privacy standpoint, this is the right thing to do.  I spend a lot of time connected to Wi-Fi in other offices, coffee shops, hotels, at conferences, and other places where a nefarious system administrator could easily snoop on my search queries and other non-encrypted web usage data.  Google’s new two-factor authentication makes me secure accessing Google products (including Google Analytics!) while connected to potentially sketchy Wi-Fi.  Now I have the same level of comfort while using Google search in potentially unfriendly places.

The data will show the impact of the change in the next few days as it is rolled out to everyone.  Regardless of the scope of the data that has been affected, I hope this post had made a strong argument for not freaking out about it.

This post is to thank and highlight a video from a person that has had a tremendous amount of influence to my career: Avinash Kaushik.  This video was recorded back in 2007, and talks about a topic that is near and dear to my heart:

Culture.

In the wrong context or to the wrong people, talking about “culture” causes people’s eyes to glaze over.  Based on my experience, I am no longer one of those people. My tale of developing a data-driven culture in a large corporation follows:

In 2006, I accepted a position as a Senior Manager of Web Analytics for a large business services firm.  As the product manager of the organization’s enterprise-wide web analytics software and data collection framework, I had my hands full developing a data capture and reporting framework as part of a complete web reboot by the company.  Although implementing an enterprise click stream tool as well as a framework for web data integration into the company’s data warehouse was a technically complex task, it was fairly straightforward once the requirements were determined.

What was not straightforward, however, was how to develop a data-driven culture in regards to how the company used its web data.

The organization had nine different, quasi-independent business units with about thirty-five different web sites.  Nobody from the business units was focused on web analytics, however, each business unit had a web team that was focused on managing the content on the sites.  My goal was to transform those positions from content managers into data driven product managers of their web sites.

So how did I attempt to accomplish this?  I empowered the web managers with their own data. I trained the web teams on both our clickstream and data warehouse tools and gave them the ability to independently develop actionable insight about their clients’ web usage.

Less than a year later, these managers could look at both individual and aggregate customer data and determine how specific web-based activity affected their business units’ bottom line. They had total visibility into all of the company’s marketing data, allowing them to explore the data and develop objective arguments for action.

Looking at my social graph on LinkedIn, I see that, three years later, some have moved into different roles either within or outside of the firm, but at least four of those former web managers have moved on to be web analysts, two with a top-tier web analytics consulting firm.

My approach here was directly influenced by Avinash’s first book, blog, and talks that he was giving at the time.  His guidance was, and continues to be, useful and inspirational for the entire online marketing community.

Creating a Data Driven Culture by Avinash Kaushik

Is there a privacy concern here? No, because the scope of data that is collected is clearly defined – the data collected by testing tools is far less detailed than that collected by click stream tools or transactional data.

Is there a user experience/user sentiment concern here? Possibly.  Although user experience varies widely by platform and site customization features, by the very nature of a test, certain visitors are getting a sub-optimal experience.  Should informing site visitors of the use of these tools, or even the fact that they are participating as a panelist in an experiment, be part of a policy of transparency and potentially cut off any negative sentiments created by the testing?

This all leads to my question:

Do you explicitly inform your sites’ visitors of any A/B or multivariate tests/tools that are deployed on your sites?
	No.
	Yes, we cover all testing generically in our TOS/Privacy Policy.
	Yes, we mention specific tools and/or tests in our TOS/Privacy Policy.
	Yes, pages that are part of test treatments inform visitors to the existence and nature of the test.
Web Polls

I’d love to hear about any specific experiences related to testing and disclosure in the comments!

Although I have years’ of experience implementing and using other web analytics tools, today I use Google Analytics on nearly every site that I manage.  It has become the de-facto standard web analytics tool for content and small e-commerce sites for a reason: it is easy to implement, has enterprise-grade features and a large user base, and it is FREE.  Here are the reasons why I’m not freaking out about a potential loss of visitor data from this tool:

It Isn’t about the Individual Visit

The power of clickstream analytics tools, like Google Analytics, comes from deriving actionable insights by exploring aggregate site traffic across discrete time periods and specific traffic segments.  You simply aren’t going to get very much actionable insight delivered by looking at one person’s visit to your site or even by tracking one person’s visits over a longer time period. In fact, Google Analytics’ terms of service explicitly forbids implementing it in a way that can uniquely identify individual visitors.

Aggregate Data is More about Precision than Accuracy

Here is a thought exercise: what if Google Analytics or some other clickstream analytics tool is delivering actionable insights that boost your site’s conversion rate but is only collecting data from about 95-99% of your site’s visitors?  That extra 1-5% isn’t a big deal as you can safely assume that the missing 1-5% is acting like the other 95-99% of your visitors.

Due to Javascript not loading, the mechanics of the Javascript not triggering the call back to Google fast enough, or visitors’ current use of ad-blocking and privacy tools, I generally assume that I am CURRENTLY missing about 1-5% of my sites’ pageviews.  Moreover, if you have a site with a large amount of traffic (millions of pageviews per month), Google Analytics suggests that you estimate traffic data based on sampling your site’s traffic to speed up the processing of your reports.

Back to the thought exercise: I expect adoption of this plug-in to be somewhere in the neighborhood of 1-5% of all users.  Is your traffic data fatally flawed if you are missing 2-10% of your pageviews? What about 20%?

Unless the people who install the plug-in are going to behave differently (as a group) than those that do not, Google Analytics will become somewhat less accurate with no loss in precision.  In the context of most sites’ objectives, there is not going to be a reason to question the validity of the conclusions that are drawn from Google Analytics unless there is widespread adoption of the plug-in.  This is because actionable site optimization metrics are based on rates (conversion rate, funnel exit rate), rather than on absolute numbers.

Clickstream is Only Part of the Puzzle

There is an ever-increasing amount of information that is being generated by people interacting with your brand online.  On your site, there is the potential to collect transactional data, direct voice-of-customer data, site testing data, contact us form data, etc., that is typically integrated with, but discrete from, Google Analytics.  Off site, there are interactions with your brand on social media, email marketing activities, and any offline interactions that may also be generating data.  It isn’t that your clickstream data isn’t important – it is just that there are other sources of data that may prompt action on the part of the analyst.

Allowing the Opt-out Is the Right Thing to Do

As site owners, we should never lose sight of our objectives.  There is a reason why our sites exist (sell something, provide information, display advertising) that is fundamentally more important than how we measure and improve our sites’ ability to achieve those goals.  Perhaps unfairly, some peoples’ concerns over privacy will cause them to block a tool that is likely being used to understand and improve their experiences, but we should respect their wishes and accept this as a new browsing paradigm in an environment with many other evolving browsing paradigms.

There Are Alternatives to Google Analytics

Of course, there are other web analytics packages out there if Google Analytics is no longer getting the job done.  It is pretty standard for “enterprise-level” web analytics solutions to include a clickstream tool, a CRM tool, a data warehousing tool, a testing and optimization tool, a social media monitoring and engagement tool, etc., along with their “enterprise level” cost and implementation difficulty.  There are other free tools out there with the features that you would expect with a free tool.  Google also sells Urchin, which doesn’t rely on Javascript to collect data, but instead uses server data logs as its primary data source.

In summary, I don’t think that there will be widespread adoption of the Google Analytics opt-out.  Even if there is, it won’t totally strip away the value of the tool and there are other clickstream analytics tools out there (as well as other sources of web analytics data).

The main point of the speech was to highlight the city’s efforts to bring accessibility to city services using internet-based technologies, including making the city’s famously successful 311 service available through Skype and Twitter.  According to the Mayor, “the more accessible we make it [city government], the more accountable we make it.”

Of note to the Web Analytics community, the Mayor also mentioned that the city is going to be taking to look at the traffic to the site for the purpose of changing the “site architecture” to improve the online experience for city residents.

The Mayor also announced that the city is going to partner with Google to study “anonymous usage data” to “allow us to optimize the content on the web site based on what people are most often searching for.”  Could it be that NYC.gov is implementing Google Analytics?

It is great to hear that New York is committed to both using cutting-edge technology to connect to its residents (Twitter, Skype, whatever Google technology they are using) as well as optimizing user experiences so that New Yorkers have a more positive and effective experience dealing with their government online.

Given the amount of social media participation with the 2008 election, and the subsequent appointment of Vivek Kundra as our country’s first CIO, I expect that there will be increasing pressure on government agencies to not only move services online (for example, see DC’s API’s), but do so in a way that is most useful to those agencies’ constituencies.

An excellent, and timely, resource that examines the state of Web Analytics in the public sector is the recently-released Web Analytics Asoociation’s Public Sector Survey Analysis.  The study describes and analyzes survey data from web managers within government, academic, foundation, non-profit, and other public-sector focused organizations.  The report describes the web analyst roles, metrics, and optimization strategies of these organizations.

The Web Analytics Association is a great resource to organizations and people in the public sector seeking to make their sites more useful to their constituents and (perhaps more importantly) developing and fostering a culture of analytics within their organizations’ web management.

As an aside, I am currently attending the PDF conference.  If you read this post and are at the conference, I would love to have an opportunity to meet you.  Email me tom*at*tomsanalytics*dot*com or DM me on Twitter @tomsanalytics.

The title of my session is “Social Media Analytics: Monitor, Measure and Manage Your Reputation on the Wild Wild Web of Social Media.”  I will be sitting on a panel with the following people:

Kate Kaye, Clickz (Moderator)
Peter Corbett, iStrategyLabs
Stan Magniant, Linkfluence
Joe Mansour, David All Group

I am extremely excited about our potential discussion since we are all hands-on practitioners with different perspectives and experiences related to Social Media and politics.

My session runs from 2:00pm – 3:00pm on Tuesday, April 21st.  The conference runs from Monday, April 20 through the 21st in the Reagan Building – hopefully I will see you there!

Last week, Devaney was before Congress touting the demand for this information.  His testimony was picked up by several news outlets, including these stories at CNN, Federal Computer Week, and the L.A. Times.

In his testimony, he described traffic levels at Recovery.gov to be about “4,000 hits per second” (emphasis mine).  This means that in a day, the site would get about 350 million hits per day (4,000 * 60 seconds/minute * 60 minutes/hour * 24 hours per day).  To most people, this would imply a traffic level equivalent to every internet user in the United States visiting the site more than one time, every single day.  However, this statistic is misleading.

A “hit”, commony confused with a “page view” or a “visit”, is a single HTTP request from a web server to a browser for a specific resource.  HTML pages, images, flash objects, CSS sheets, external scripts, and other other files requested by a browser to render a page each generate a single hit upon their request.  The net effect is that single pages can generate several hits for the display of a single page to an individual user.

In the case of Recovery.gov, the home page loads 30 files to fully render.  If the 4,000 hits statistic is accurate, then there are really about 11.5 million page views per day.  Assuming that the average visitor views 5 pages on the site (which is a big assumption), the true number of visits to the site is more like 2.3 million per day, or about 27 new visits per second.

27 is a lot less impressive than 4,000, but it may be a much better picture of the true nature of traffic to the site.

Why am I lucky enough to have an eMetrics in my city: Washington, DC!

I love DC.  I am lucky enough to live in one of the greatest neighborhoods in one of the greatest cities in the world.  However, the point of this article isn’t to discuss life in DC, it’s to make the argument that DC is the best possible location for the eMetrics Marketing Optimization Summit.  The reasons why:

It is easy to get here. And not just for me.  There is great transportation infrastructure in this city.  You can take a train from New York or Boston, a direct flight into National Airport from nearly every regional airport in the States, and direct flights from the Americas and every major European city into one of our three airports – National, Dulles, or BWI.

It is easy to get around once you are here. The Metro trains that service the city and close suburbs are clean and inexpensive.  We have approximately two taxis per resident in the city (I made that up, but we do have more taxis per capita than any other major American city).

We have tons of hotel options. Since the summit is in late October this year, you‘ll have a wide variety of options.  If you have the means, stay at the Willard Intercontinental – it is worth every penny.

We have tons of conference hosting options (that aren’t the Hilton Alexandria Mark Center). DC hosts lots of events and hosts them well.  From the smallest trade show to a multi-million attendee Inauguration party, we’ve got events covered.

We have fantastic opportunities for extra-Summit activities. Most of my best conference experiences have come outside of conference venues.  We have great bars and world-class restaurants.  The Smithsonian museums and the National Zoo are free.  If you come to DC and don’t take advantage of these “touristy” activities, you are truly missing out.

But most importantly, there are some great people that attend eMetrics in DC. DC is a city of networkers.  Attending eMetrics in DC, you will have the opportunity to meet people that work for trade associations, lobbying firms, political organizations, interactive agencies, e-commerce companies, NGO’s, non-profits, social networks, government agencies (just don’t ask them about third-party cookies!) and web analytics vendors and consultants.  Meeting this wide variety of people at eMetrics who are enthusiastic practitioners of their particular craft and are willing to share their experiences in a collaborative manner makes the summit a great experience, year after year.

It’s true: a big reason why the conference is so good is because it is here.

I agree with Avinash in that Bounce Rate is one of the most important metrics for any type of site. A high bounce rate negatively affects Time Spent on Site in ways that may not be immediately apparent; have a look at my article on Life Tables and Time Spent on Site to see how. However, we need to take a closer look at our data collection methodology before we assign too much importance to this metric.

Bounce Rate Defined

The classic definition of a Bounce is a visitor that views a single page and then immediately leaves without viewing any other pages. A site’s Bounce Rate is the ratio of single page visits by the total number of visits. Anil Batra defines it (in his post “Bounce Rate Demystified“) this way as well, but with an added definition adding people that visit a single page, but exceed some predefined time limit on that page.

Problems with Bounce Rate

There are potential problems with Bounce Rate in that it is not consistently defined/recorded by analytics packages, it is not applicable to certain type of sites, and there is potential for the metric to be over-emphasized, particularly by those who are not aware of the shortcomings of their own analytics methods.

Bounce Rate in an E-commerce Site

As an example, if I were optimizing an e-commerce site, bounce rate would be very important to me on a number of levels. Presumably, I ensure that both my site’s homepage and sub-pages (perhaps individual products) would be indexed by search engines and other product search aggregators. I would also market my site through keywords purchases and banner advertising. I would want to closely examine individual page bounce rates to see which sites are most inviting to people to remain on the site and presumably stay on a path to a purchase. I would also closely look at the bounce rate of visits generated by each of my keywords; a high bounce rate indicates that I may want to look at spending my money on keywords that are more in-line with my site’s offerings.

Bounce Rate in a Content Site

If I were optimizing a site with a “walled garden”, a subscription-based content site, I would be much less concerned about a high bounce rate as it is defined now. Bounces to login pages would be expected, these could be attributed to visitors without access. Bounces from individual article pages could be attributed to subscription users that have individual articles bookmarked or users that email articles to another subscriber.

If I were optimizing an open content site, such as this blog, I would not have the tools at my disposal to make any conclusions based on Bounce Rate. Visitors to individual article pages come from search engines, RSS readers, social networking sites, and linkage from other blogs. A bounce from any of these sources may not be a bad thing – I want my visitors to read a single article if that is what they are seeking. The problem with our classic definition of a bounce is that there is no difference between a visitor that spends a minute to read an article and a visitor that clicks into the blog and leaves within five seconds.

A Solution for Google Analytics Users

One (messy) workaround for Google Analytics users is to simply call the urchinTracker function after a certain delay (perhaps 15 seconds). This will count another pageview for the visit, but no longer count that visit as a bounce.

The Future of Bounce

Unless we change our methodology for measuring bounces, bounce rate will cease to be a “sexy” (or even relevant) metric for most sites due to increasing popularity of social networking, RSS, and alternate browsers such as mobile browsers, televisions, and other types of syndicated content. For example, I regularly read “newsy” blogs such as TechCrunch and Engadget. However, my path to TechCrunch is my Google Homepage >> Techcrunch and my path to Engadget is Google Homepage >> Digg >> Engadget. My personal overall bounce rate to all three of these sites is probably 95%. Deep-linking and social news will continue to drive Bounce Rates up (and the overall relevance of Bounce Rate) unless we change the way we collect and interpret bounce data.

Blogs Cited in this Post

I highly recommend Avinash’s blog, Occam’s Razor and Anil’s blog, Web Analysis, Behavioral Targeting and Advertising. Check them out.