During the previous four parts of this series( intro, paper prototyping, design, and web standards), I focused on the redesign of Tag Builder that we completed in the spring. While authoring these posts, I didn’t have the liberty to disclose the application we launched just last week, Webtrends Analytics 9 Insight. However, these posts were not unrelated as the methodologies and principles that guided the redesign of Tag Builder were also core to the genesis of Analytics 9 Insight. In the final post in this series, I’ll introduce Analytics 9 Insight features that were inspired by working directly with our users as we progressed through designing, building and testing.
Stories
For the past year, we have been collecting stories from our customers. While listening to what is currently working for them, we paid special attention to what isn’t. We found that our pro users spent a majority of their time creating dashboards and reports for other users in the organization, from executives to marketers. These other users greatly outnumbered the pro users and found that the Webtrends interface was difficult to navigate and had too many options. We incorporated this feedback into stories that guided our design and development.

Some of the stories that guided Analytics 9 Insight
Prototyping and User Testing
Working from these stories, we prototyped the new application completely on paper, using a process called paper prototyping. A cross-departmental team made up of designers, developers, writers, testers, and others “built” the application with markers, paper, scissors, and tape. We then took the paper prototype on the road to our customer conference in Las Vegas. There, we had multiple sessions where our customers attempted to accomplish tasks derived from our stories. When they succeeded, we knew we got it right. When they didn’t, we broke out the markers and paper and designed new solutions. In the end, the paper prototype became a blueprint for our design and development efforts.

David working on some paper prototypes before testing with users
Beta
We developed the application in just nine weeks. We wanted to get it in the hands of users as soon as possible, initially releasing it into a private beta so we could gather feedback and address issues before releasing it publicly. It was a great thing to do because it gave us the opportunity to fine tune the performance and remove many issues before launching the application to the entire On Demand user base.

White board used to track key issues during the private beta
Launch
We were all a bit nervous as we launched the product. It was exciting to see the activity on Twitter and the early reviews pour in. Our goal was to make the most intuitive, useful analytics application in the industry, but because the stories that guided our design efforts came directly from our customers, we were also confident. To keep the dialog going, we included a Get Satisfaction link right in the product. We want to continue to hear from our users.

Get Satisfaction feedback built into the product
New Features
The main features of Analytics 9 Insight have been fairly well documented by now. Web Services URLs ready to embed in Excel, Story View, and RSS overlay among others. The following are features that you may not have heard about that come directly from listening to our users.
1. Shareable URLs
The original version of share included an email option. Then we thought about how people communicate in the office today and realized with corporate IM, intranets, message boards, etc, email was just one way in which users would like to share reports. We took a step back, removed the email option, and created a URL system that would allow anyone to copy and send the link to any other user. This makes sharing reports extremely easy.

URLs reflect choices in interface. Great for sharing.
2. Dimension and Measure Auto Suggest and Search
Creating custom reports in Webtrends allows for endless possibilities. Unfortunately, it can also create endless reports. When considering many different options in which a user could categorize reports, we discovered that most users could remember which reports they were looking for by identifying which dimensions and measures that the report is made of. In the profile dashboard, below the fold, is a list of all reports available to you for that profile. If you start typing a measure or dimension in the search box, the auto complete will kick in at three letters. Choosing one of the measures or dimensions will filter the list to only include reports with that measure or dimension.

Auto-suggest for dimensions and measures across reports

Filter by measure or dimension
3. Copy and Pasteable Content
Sometimes, when scanning through analytics data, you run across a bit of data here or there that you need to copy and paste into a spreadsheet or report. We prioritized this ability over almost everything else. Whether you are in the profile dashboard and click on table view or are in a report, you are able to select the data directly from the interface. Very handy.

Trend view

Switch to table view for easy copy and paste
4. Weekend Indicators
Insight uses 7-day, 28-day, and 91-day date range shortcuts. The first is typical but the second two are a bit peculiar. There is a reason why we chose them instead of the standard 30 and 90 day range options—weekends. You see, both 28 and 91 are divisible by 7. This allows us to display data ranges in compare mode where the weekdays align. Interesting patterns emerge when you show one date range compared to another and the days of the week align perfectly. This is extremely useful as most businesses see a difference between weekdays and weekends.

28 day compare view with weekend indicators

91 day compare view with weekend indicators
5. Date Range Options
The default date range options, as mentioned, are 7-day, 28-day, and 91-day ranges. However, if you click on Custom, you are able to select any range of days you want. After you make a selection, the compare default is to choose the same number of days just before the selection. To change your compare range, all you have to do is select the starting day and the application will automatically select the range equal to the same number of days as the main range. There are also other shortcuts available in the custom option. The year, quarter, and month are all selectable as shortcuts. You can also use one of the numbered week selectors, just to the left of each week.

Custom date range allows for day, week (shown), month, quarter, year, and custom ranges
6. True Visitor Metric
If you select one of the calendar shortcuts mentioned above (year, quarter, month, week, or day), the Visitors key metric will appear in the profile dashboard. Because Analytics is tuned to scan across these standard date ranges and produce a true visitor count, we added this when those ranges are selected.

True visitor count when date selection is set at standard report period
7. Pivot
When you drill into a report and have a specific date range selected, the last thing you want to do when switching reports or profiles is set up the date range all over again. We know that many users switch profiles when viewing similar reports. We added a dropdown to the right of the profile and report selection that makes this extremely easy to jump from profile to profile or report to report. This saves quite a bit of time.

Click on pivot dropdowns to change account, profile, or report
8. Adaptive Account Dashboard
Every customer is different. Some have many profiles. Some have just a few. And some just have one. We wanted to make sure that each customer/user had an interface optimized for them. If a user only has one profile, then they never see the account dashboard and jump right into the profile dashboard. This eliminates unnecessary steps after login. If the user has less than 25 profiles, they see the standard view which features a single table of profiles that can be sorted by any of the available measures and embeds a sparkline of page views. If the user has more than 25 profiles available, they are presented with the compact view which features four columns and reduces the visible metric down to one. In this view, the user can choose a different metric to display by clicking on the metric dropdown at the top of the screen. They can also hover over any profile to get a multi metric and sparkline view. Interesting tip. A user can force the standard or compact view by adding a query parameter to the URL. ?mode=standard or ?mode=compact will change the views.

Account dashboard in standard mode

Account dashboard in compact mode
Inspiration
There are many more customer-inspired features in Webtrends Analytics 9 Insight. If you are a current user, we hope you enjoy Analytics 9 Insight. After all, you helped create it. If you are a customer and haven’t been an active user, we encourage you to login and give it a try. When you do, don’t be shy about sending us feedback through the Get Satisfaction widget in the product (hover over Help and click Feedback). We will also be conducting some user testing next week, August 17 – 21. If you are in the Portland area and would like to test some prototypes of what we are designing next, let me know in the comments.