In the 1981 documentary “Vernon, Florida” by Errol Morris, there is the following joke…

Two sailors are looking out at the ocean. First sailor says, “That’s a lot of water out there.” The second sailor responds, “Yeah, and that’s just the top of it.”

That is how I feel about web standards. What you see in the browser, is just the final rendering that is a combination of html, css, and javascript. So much care and detail can go into these elements to produce the final result. Often, as I browse the web and stumble across a nicely designed site, I will turn off styles and/or javascript to enjoy the simplicity of the plain html and then scan the css and javascript to see how it was constructed.

In the previous article in this series, I walked through some of the more significant design changes we made to Tag Builder. In this article, I will walk through the changes we made just below the surface, to the client-side architecture (html, css, and javascript) to create a new user experience.

1. Web Standards

Web standards is nothing new. It is usually the term used to describe a web application or web site that uses basic HTML elements for marking up content, css for presentation and layout, and javascript for interactivity and dynamic content updates. Web standards typically avoids using unnecessary browser plugins to do the heavy lifting. With advances in javascript libraries and browser performance boosts, web standards based web applications are approaching native OS level quality. To build a world class user experience with our web applications, building in web standards is key.

Tag Builder

Our goal for this revision of Tag Builder was to strip down the HTML to its bare essentials. Divs, classes, lists, paragraphs, headlines, and form elements are the basic building blocks we rebuilt it with. All of the layout, styling, and behavior is achieved using CSS and javascript. You can see it for yourself by turning off styles in your browser (use these instructions to turn off styles in Firefox, IE, and/or Safari). By taking the time to separate these layers and let the rendered html, css, and javascript to do the heavy lifting, it frees up the server side code to be much simpler. This dramatically improves performance and reduces server side complexity.

Tag Builder with styles turned off

2. Accessibility

One of the major benefits of working with web standards is accessibility. In my previous experiences with developing web applications, accessibility is often thought of as an additional task like multilanguage support. By staying true to web standards, many accessibility considerations come automatically. Our attention to detail created one of the most unusual challenges we have experienced with web standards. We wanted to ensure that the Tag Builder form could be navigated by keyboard. When the elements tabbed down into the Additional Options, we found that the browser would jump the view to each field as it was highlighted. This seems normal, but what was surprising was that the javascript or CSS was unaware that the page had moved and so we saw that fields would just show up in odd parts of the interface and destroy the multi-slide effect of moving from one section to another. We eventually found a way to trigger a call to javascript just before the field was in focus so that all things were ready when the tab action occurred. We were able to keep the javascript sliding effect and the ability to navigate the form by keyboard.

3. Javascript for Helpful Behavior

There are many javascript libraries available today to add a bit of slickness to web interfaces. These can often be overdone and to the detriment of the user experience. For Tag Builder, we had a few areas in which javascript behavior was considered to be beneficial and we used it sparingly.

Tag Builder is one long form with many options. We wanted to make sure that users could access the many options but not be overwhelmed by them nor did we want to split the form into many separate forms as many users like to hop around the options when configuring their tracking code. The solution we came up with was to collapse the options using javascript and then as the user clicks on the options they are interested in, those options slide into view. As a user fills out the form fields, light gray dots turn dark indicating how much of the form is filled out.

Indicator Dots

Another way in which we leveraged javascript for behavior was to use it to surface help topics.

4. Integrated Help System

In the previous version of Tag Builder, each form field had companion help that explained what the option entailed and exposed the details associated with each option. This was met with praise from users but left the form littered with little question marks and reduced the readability. To solve this challenge, we used javascript and css. If you turn off styles, you’ll see that the quick help is part of the html. When you turn styles and css on, you’ll see that only when you rollover a form field, will the quick help appear. This quick help provides enough information to address the basic implication of each form field. We also added a link at the end of each quick help for more info. When a user clicks that, a simple modal window pops up with the related content, explaining the option in further detail.

Rollover Help

With the focus on web standards, the redesigned Tag Builder is much cleaner, simpler, accessible, friendlier, and ready for future expansion. With the launch of the Webtrends API, we also developed a simple web standards based interface for developers to generate and test various web services.

REST URL Generator

I hope you’ve enjoyed this series but you are probably thinking, “How much can this guy write about a simple one page form web application?” This is the last article. For the 5th and final part in this series, I have something much bigger than Tag Builder or the REST URL Generator to detail. Our team has been very busy over the last couple of months. I’ve been trying to pace myself with this series on our new user experience as we wrapped up development so that I could end it with some exciting news. I’m happy to say that next week is it! Follow us on Twitter to find out as soon as we announce and stay tuned to this blog to learn more following the announcement.

We are proud to announce that the ink has dried on a definitive agreement to acquire Widemile, which adds world-class multivariate testing and site optimization to our arsenal of customer intelligence technologies and services. The combination produces an unmatched analytics and optimization platform giving our customers an even greater advantage in the marketplace. For us, this deal reinforces our company’s momentum under the helm of our CEO Alex Yoder and further solidifies our position as the leading enterprise customer intelligence company.

“Webtrends now offers the most open, elegant and powerful analytics and optimization platform available anywhere.  Our customers want to further leverage the rich data set available with Webtrends Analytics, and Widemile’s capabilities provide a natural path to help our customers achieve better performance and improved ROI,” said Alex Yoder, Webtrends CEO.

Designed by optimization experts, Widemile’s platform provides the capabilities marketers need most, including intuitive wizards, real–time reporting and analysis, advanced visitor segmentation, full service and best practices services and training. Widemile’s third–generation, on–demand multivariate testing and site optimization platform is best known for its industry leading ease of use, algorithms and architecture, and reporting capabilities.

According to Forrester Research, “All too often, multivariate and A/B tests are performed as a means to identify problems and address issues, rather than a method to create better pages from the start.”  Marketers have been hesitant to use and deploy multivariate testing and targeting solutions, which have been traditionally difficult to use, time consuming, and resource intensive.  In response to the needs of the market, Widemile has built a platform that is both comprehensive in it capability and easy to setup and use in its operation.  Adding this functionality to the existing Webtrends platform further extends the core objective that customer behavior must be tested, interpreted, optimized, and measured to provide customer intelligence professionals with the ability make their businesses more successful.

“The Widemile and Webtrends offerings fit naturally to address the macro challenges and opportunities facing our customers in the coming years,” said Robert Bergquist, Widemile CEO and President.  “The combination is a best-in class integrated web analytics and optimization platform allowing users to plan online marketing programs with the assurance that built-in testing, targeting, and optimization solutions will maximize conversion rates.”

We’re on a mission to provide best-in-class enterprise customer intelligence applications capable of testing, monitoring and measuring all aspects of your online visitor’s interactions with your brand. 15 years ago we created the web analytics market and forever changed the marketing world by giving the online space its ultimate trump card: measurement. Fast forward to present day and the customer data landscape has grown significantly, especially over the past decade. We’ve made some big moves under the leadership of our CEO to prepare for the customer data driven future. Forrester recognized our efforts in their latest Wave Report saying,

Webtrends has undergone the biggest transformation of all vendors evaluated in the past year, with the restructuring of its entire management team and refocused efforts on Web analytics and its concept of openness. The result is a company united in delivering leading enterprise-class capabilities, which it does with exemplary metrics, dimensions, and correlations. The company scored high marks because of its ability to create unlimited custom and calculated metrics, as well as the capacity to perform unlimited correlations. Reference clients expressed high levels of satisfaction overall, with extreme satisfaction in responsiveness/support, application reliability, and value. Webtrends continues to develop its extensibility through expansion of data exchange, accessibility options, and a strong partner program.

Nate DiNiro interviews Jeremiah Owyang at the Social Media Club's event at Webtrends.

It was great hosting the Social Media Club with Jeremiah Owyang at our offices this week. Pizza, beer, and beautiful weather for the rooftop patio added to the great discussion facilitated by Nate DiNiro. Owyang spoke about things like the scalability problems with social media, utility of Twitter to small businesses, and how email is still the largest social network. The event was standing room only and several people commented to me how much they enjoyed it and how time seemed to fly by.

Thank you to the Social Media Club for organizing the event. Thanks to Jeremiah Owyang for taking time to join us. Thanks to Widmer for sponsoring the beer. And, thanks to everyone who attended making the night fun!

Recordings of the event

• Amber Case live blogged the discussion.
• Jeffery Bunch has some photos of the event up on Flickr.
• Jeff has also posted video of Owyang’s interview as well.

We’re a week away from the 6th annual Internet Strategy Forum Summit. Webtrends is proud to be a Partner Sponsor for this premier event. This year’s list of keynote speakers includes:

• Jeremiah Owyang, Sr. Analyst Social Media, Forrester
• Katherine Durham, VP of Marketing Imaging&Printing, Hewlett-Packard
• Johan Jervoe, VP of Creative & Digital Marketing, Intel
• Sheila Tolle, VP of Marketing Small Business, Intuit
• Duane Schulz, VP of Interactive Marketing, Xerox
• Lisa Welchman, Founding Partner, WelchmanPierpoint
• Chris Dill, Chief Information Officer, Portland Trail Blazers
• Steve Gehlen, Founder & Executive Director, Internet Strategy Forum

The event will be MCed by Rob Smith, Editor of the Portland Business Journal.

Event Details

Thursday, July 23rd, Check-in starts 8:00am
MAIN PROGRAM from 8:30am – 5:00pm PST
Governor Hotel – Portland, OR
NOTE: This event is open to all.
$195 – Early Bird Registration Fee (association members receive a discount)
$175 – Early Bird Remote Attendance via the Web

Friday, July 24th, 8:00am – 12:30pm PST OPTIONAL: Internet Strategist Symposium
$95 – Early Bird Registration Fee
$75 – Early Bird Remote Attendance via the Web

More details are available on the ISF site along with registration.

Social media is changing the way businesses connect with markets. Conversations take place where the participants choose and are often spread across multiple sources. Pulling the conversation together in order to make informed, engagement decisions is at the heart of marketing’s challenge in this new media landscape. To demonstrate how Webtrends Social Measurement helps with a social media campaign, we advertised on the outside of Portland’s light rail, MAX, to monitor Twitter for responses to our ad. The idea is to publish the results on another MAX ad in October. We forecasted it would take 3 months to get a volume of results that would tell a compelling story, but a compelling story emerged in less than a week and spanned Twitter, blogs, and mainstream news. This is an update on that campaign.

Campaign results

The following are results from our measure of conversation on social media driven by our marketing campaign.

If a source had two or more posts, we combined the results. KATU had the most comments. BikePortland had the most commenters. Also worth noting, after we filtered for duplicates we discovered that the number of comments that KATU reports on their site are nearly double the actual amount. Unique commenters are considered unique usernames.

Sentiment is a hot aspect of social media. We recently wrote a post challenging the use of algorithms to determine sentiment, and this campaign is a great example of how NLP falls short. Sentiment is a complex tapestry of human emotions, not just positive, negative, or neutral. In this case our criteria was:

• Yes answers: People that only said yes.
• No answers: People that only said no.
• Same answers: People that said the amount paid now is fine.
• More answers: People that said pay more; including answers about licensing and registration.
• Less answers: People that said pay less; including answers about rebates or tax credits.
• Other answers: Comments that didn’t explicit state a position; including answers that were sarcastic or answered with a question as we can’t be sure of their inference.

As our results are human scored and could be interpreted differently by other people, we are providing our data so that it can be reinterpreted by anyone who is interested. We showed the top four sources broken out to show what impact sources had on the overall sentiment. Overall, the sentiment was that people favored either the same or less taxes than cyclists currently contribute. The strongest opposition came from Twitter, our original target source for monitoring conversation for this campaign.

Tag cloud for all conversation

Tag cloud for KATU

Tag cloud for BikePortland

Tag cloud for OregonLive

Tag cloud for all Twitter

Reflecting the salient points

In today’s media landscape, it’s challenging to track many viewpoints from many stakeholders. Making sense of a high volume of conversation from disparate sources isn’t easy for anyone. After tracking this conversation with our tools, here are some of the main points we learned:

1.) There is a lot of confusion on the facts about who pays for what currently

The most quoted source was the Federal Highway Administration (FWHA), which reported that 92% of the funds for local roads come from property, income, and sales taxes. It was the lack of general awareness of this fact that we believe causes a lack of empathy with cyclists as it pertains to taxation/registration. We did notice that the mainstream news sites that had the most “yes” answers to our question, many of which were accompanied by messages that indicated people didn’t think cyclists paid taxes currently. Others pointed out that cars, boats, motorcycles, airplanes and more have additional taxes including licensing, registration, and fuel taxes. Respondents pointed out that many cyclists are also motorists and therefore not only pay the unique motor vehicle taxes, but that it would be double taxation. To which people replied that the additional taxes are paid for per vehicle regardless of use, which therefore doesn’t constitute double taxation.

What is clear to us is that more conversation around this point could help clarify for everyone where contribution currently comes from, which is key to having a discussion about where obligations should be moving forward. While the information is out there, it doesn’t appear to be widely distributed, nor agreed upon. We see this as the greatest opportunity for constituents involved in public policy development around transportation.

2.) More taxes on cycling could be bad for everyone

There were two reasons for arguing this point. First is that people pointed out that less traffic, less damage to the environment, and more benefits were good for everyone. Participants reasoned that anything that discouraged cycling, such as taxes, prevents the community from reaping those rewards. As bikes do not use fuel, the only additional per vehicle costs that could apply would be licensing and registration. The only argument we saw against this point was the feeling of it being unfair that road faring cyclists didn’t have to pay these costs as well. Contenders pointed to the fact that state governments and municipalities who have adopted bike registration programs have later abandoned them due to the loss of the revenue to overhead.

3.) Some cyclists want more taxes, but under specific conditions

When people said they’d be willing to pay more taxes, they were only willing to pay more if the money went to bike specific infrastructure (we’ll include numbers in our final report). Nearly every answer for more taxes included the word “if”, often in all caps, and ended with extra exclamation points–indicating that people felt strongly that paying more was critically tied to the conditions that the money went to bike specific infrastructure. People expressed interest in better maintenance of existing bike lanes, more and better bike lanes, new bike-only paths, theft recovery, and more.

Other themes that came up were: Motorists are rude and sometimes dangerous about sharing the road; Road laws are not enforced equally on cyclists; Cyclists aren’t required to have liability insurance; Some people consider “cyclist” a loaded word.

Transparency

As part of our campaign, we are being transparent. That means we share openly about all details of the campaign. It also means our employees are free to discuss their thoughts publicly, even if they are critical. So, let’s talk about how we make money with this campaign. As I mentioned earlier, this campaign is part of a larger campaign called The Open Campaign. The Open Campaign will have it’s own microsite that will openly display the methods and results from various campaigns we’ve run using our tools. The concept is to provide a reference to businesses for what running campaigns with our tools looks like. It is that site that we’ll use to build leads for our sales team. While we do have a landing page for the MAX ad with a form on it, that form is to enter to win a free TriMet pass. Information from that form is *not* given to our sales team. It is strictly a promotional offer to encourage participation.

Want to see the results and make your own reports? Download our data »

Next Steps

Our question surfaced a diversity of viewpoints in much less time than we expected. As a result, we have an opportunity to progress the conversation by asking another question. If you have a suggestion for a follow up question, please share it as us a comment below.

I am very pleased to announce the launch of a market-leading data collection capability for Webtrends. The addition of this API to our capabilities is another example of why Webtrends is the most powerful, open and elegant solution in the industry.

Keeping with our commitment to gather your feedback and deliver the most relevant of tools, we are making this new capability available in Beta for all of our customers and partners to use now.  Please help us help you by getting involved today. Push the limits! Get creative! Then let us know where we can improve on the capabilities. We will continue to evolve the API. Anything that connects to the internet is fair game.

As you, our customers, continue to expand your online presence beyond your web sites, it is our responsibility to grow our own measurement capabilities ahead of your evolving needs. Our data collection API is a critical addition to your arsenal of tools for collecting new data. The primary purpose for the API is to track behavior across various media that access the internet where traditional site tagging isn’t possible or an optimal fit.

Interested in more details? A full overview is available on our Developer Network. Please feel free to reach out to me with any questions, comments, or to explore opportunities to leverage this exciting new capability.

Last Friday we launched an ad campaign on the MAX posing the question “Should cyclists pay a road tax?” We tracked a significant amount of conversation in a short period of time. In fact, it was much more conversation than we had anticipated and agreed with Thom Schoenborn it would be better to get results out sooner. So, in addition to our October rewrap of the MAX, we would also like to provide an update now.

Clarifications

Before we get too far, we’d like to clarify a few points:

• We do not promote the idea that cyclists do not already pay taxes. Our question is “Should cyclists pay a road tax?”, which is not about whether or not cyclists do currently.
• We do not support a dichotomy between bikes and cars. From the beginning we identified multiple stakeholders in this conversation and think narrowing the conversation to bikes and cars is unproductive.
• The volume of conversation around this topic reinforces the fact that it’s an important discussion for the Portland community. As we are cyclists, drivers, and riders we care about the outcome of the discussion and feel that the insights we can provide with our tools will help drive consensus.

The conversation space

Using Webtrends Social Measurement we were able to track this conversation’s 918 comments across 13 sources. As our first update, we wanted to share the list of sources where the conversations have taken place so far:

Sources (sorted by comment volume)

• KATU, Bike ad plastered on MAX train causes a stir (407 comments)
• BikePortland, Editorial: Marketing campaign asks the wrong question (147 comments)
• Twitter (123 tweets)
• OregonLive, New full-train MAX ad asks: Should cyclists pay road tax? (118 comments)
• BikePortland, A look at TriMet’s role in MAX ad campaign flap (53 comments)
• OregonLive, (34 comments)
• TransitSleuth, Should Bicyclists Pay a Road Tax? Measured and Tallied (20 comments)
• Zero Times Any Number, Portland Webtrends and their need for attention (6 comments)
• Portlandize, Should “cyclists” pay road tax? (5 comments)
• PopArt, Portland Bike and Marketing Freak Out (4 comments)
• Blogtown, MAX Stunt Collides with Cycling Community (1 comment)
• BTA, Bicyclists pay taxes too (0 comments)
• I’m in your water, Portland Local Media Roundup! (0 comments)

Reporting

Starting today, we will continue to share updates on this blog. Our final report will contain the following:

• Summary – This will be a roll up of all conversation providing a high-level briefing.
• Timeline – This will show the conversation sources as they unfolded over time.
• Sources – Each source of the conversation will have a unique summary of their contributions. For each source we’re going to provide a tag cloud of that source’s conversation, the talking points from that source listed by volume, number of comments, number of unique commenters, and sentiment of that source.

Tomorrow we will publish an overview of the comments broken out into the following sentiment: those who support the amount of taxes paid now, those who want less, and those who want more.