Patent Application
20110029853
1. Field of the Invention.
The present application relates to data visualization, and more particularly methods and systems for more effectively presenting analytics information to a user of such information.
2. Description of the Prior Art.
Programs for analyzing traffic on a network server, such as a worldwide web server, are known in the art. One such prior art program is described in U.S. Pat. No. 6,925,442, titled a Method and Apparatus for Evaluating Visitors to a Web Server, which is incorporated herein by reference for all purposes. Another such prior art system is described in U.S. Pat. No. 6,112,238, titled System and Method for Analyzing Remote Traffic Data in a Distributed Computer Environment, which is also incorporated herein by reference for all purposes. Webtrends Corporation owns this application and also owns the present provisional application. In these prior art systems, the program typically runs on the web server that is being monitored. Data is compiled, and reports are generated on demand are delivered from time to time via email—to display information about web server activity, such as the most popular page by number of visits, peak hours of website activity, most popular entry page, etc.
Analyzing activity on a worldwide web server from a different location on a global computer network (“Internet”) is also known in the art. To do so, a provider of remote web-site activity analysis (“service provide”) generates JavaScript code that is distributed to each subscriber to the service. The subscriber copies the code into each web-site page that is to be monitored. When a visitor to the subscriber's web site loads one of the web-site pages into his or her computer, the JavaScript code collects information, including time of day, visitor domain, page visited, etc. The code then calls a server operated by the service provider—also located on the Internet—and transmits the collected information thereto as a URL parameter value. Information is also transmitted in a known manner via a cookie. Each subscriber has a password to access a page on the service provider's server. This page includes a set of tables that summarize, in real time, activity on the customer's web site.
The above-described arrangement for monitoring web server activity by a service provider over the Internet is generally known in the art. Information analyzed in prior art systems consists of what might be thought of as technical data, such as most popular pages, referring URLs, total number of visitors, returning visitors, etc., as well as commercial activity, e.g. products purchased, time of purchase, total amounts, etc.
The amount of information that must be digested by a user of the traffic analytics tool is immense. Typically, such information is presented in graphical form (e.g.
Accordingly, the need remains for visualization techniques that present data in ways that may be more useful to a wider array of people, and that incorporate contextual information within graphical or charted trend data so that the meaning of the trends, in connection with time-sensitive events, may be better understood.
In one aspect of the invention for advanced visualization techniques for conveying analytics data, a method and apparatus is disclosed for embedding the presentation of analytics data within a natural language statement or series of statements. A template, stored in a template database, includes natural language statements with data fields embedded within the statements. The data fields are populated with the appropriate analytics data such that the resulting reporting statement reads like a conversational statement of data and trends.
The invention, also called “story view” is a unique new way to view key metrics data. Instead of visualizing it with a graph or chart, story view embeds the data into a narrative paragraph providing written context for what the data is indicating.
In another advanced visualization technique, an RSS feed is associated with three types of information: article title, the article itself, and the date/time of publication. The time from the RSS feed article is read by a data incorporator and overlay directly on top of the trended key metric at the appropriate timeline location. Key metrics data include such items as page views or time-on-site. Feeds are correlated with the web page or site and simultaneously posted articles are superimposed using a heatmapping (e.g. progressively darker shading) to indicate a density of events.
Other advanced visualization features described in the invention include: (a) comparing profiles and spaces, (b) intelligent type-ahead for meta-data, (c) multi-level pivot navigation, (d) weekend overlay in trend view, and (e) quick stats for individual days.
Comparison of profiles can be done side-by-side on a display, where the current performance is measured against the past and displayed in the same report in different profiles.
The intelligent type-ahead filters allow reports to be filtered by meta-data type occurring within the reports. Typing several letters within a search field begins the process of presenting several possible filters that may be selected. Upon selection, the reports displayed are narrowed so that only those satisfying the particular filter are included.
Pivot navigation allows one to compare other profiles across various levels of a navigation bar. The same report, but different profile, may thus be selected from the menus.
Weekend overlay provides visual indicia in combination with the graph of analytics data so that the data points occurring over weekends may be easily seen and weekends correlated. In a preferred embodiment, the weekends are shown by vertical bars on the chart. Data reporting periods can be artificially limited to 1 week, 4 week, and 13 week periods so that two charts may be overlaid with properly overlapping weekend.
Quick stats associate days of the reporting period with certain pre-defined analytics events—typically data extremes. The occurrence of multiple such events on a single day can thus give indication that such was triggered by a particular event (such as a press release) thus prompting further investigation.
The foregoing and other objects, features and advantages of the invention will become more readily apparent from the following detailed description of a preferred embodiment of the invention that proceeds with reference to the accompanying drawings.
APPENDIX I and APPENDIX II illustrate script that may be incorporated into a web page to gather analytics data from the browser requesting the web page.
Turning now to
As mentioned above, it would be advantageous to the seller to have an understanding about how customers and potential customers use server 12. As also mentioned above, it is known to obtain this understanding by analyzing web-server log files at the server that supports the selling web site. It is also known in the art to collect data over the Internet and generate activity reports at a remote server.
When the owner of server 12 first decides to utilize a remote service provider to generate such reports, he or she uses a computer 16, which is equipped with a web browser, to visit a web server 18 operated by the service provider. On server 18, the subscriber opens an account and creates a format for real-time reporting of activity on server 12.
To generate such reporting, server 18 provides computer 16 with a small piece of code, typically JavaScript code (data mining code). The subscriber simply copies and pastes this code onto each web page maintained on server 12 for which monitoring is desired. When a visitor from computer 14 (client node) loads one of the web pages having the embedded code therein, the code passes predetermined information from computer 14 to a server 20—also operated by the service provider—via the Internet. This information includes, e.g., the page viewed, the time of the view, the length of stay on the page, the visitor's identification, etc. Server 20 in turn transmits this information to an analysis server 22, which is also maintained by the service provider. This server analyzes the raw data collected on server 20 and passes it to a database server 24 that the service provider also operates.
When the subscriber would like to see and print real-time statistics, the subscriber uses computer 16 to access server 18, which in turn is connected to database server 24 at the service provider's location. The owner can then see and print reports, like those available through the webtrendslive.com reporting service operated by the assignee of this application (examples of which are shown in
The data mining code embedded within the web page script operates to gather data about the visitor's computer. Also included within the web page script is a request for a 1×1 pixel image whose source is server 20. The 1×1 pixel image is too small to be viewed on the visitor's computer screen and is simply a method for sending information to server 20, which logs for processing by server 22, all web traffic information.
The data mined from the visitor computer by the data mining code is attached as a code string to the end of the image request sent to the server 20. By setting the source of the image to a variable built by the script (e.g. www.webtrendslive.com/button3.asp? id39786c45629t120145), all the gathered information can be passed to the web server doing the logging. In this case, for instance, the variable script “id39786c45629t120145” is sent to the webtrendslive.com web site and is interpreted by a decoder program built into the data analysis server to mean that a user with ID#39786, loaded client web site #45629 in 4.5 seconds and spent 1:20 minutes there before moving to another web site.
As will now be explained, applicant has developed the ability to analyze commercial data as well, e.g., number of orders, total revenues, etc., generated by server 18, and attach that information to the variable script image request so that commercial activity for a particular site can be tracked.
To this end, applicant has developed a method in which data relating to revenues, products sold, categories of products, etc., is collected, analyzed and displayed in various report formats. An example of code that can be used to implement this method is shown in Appendices I and II. When the subscriber opens an account with the service provider by connecting computer 16 to server 18, as described above, the code in Appendices I and II is transferred from service 18 to computer 16 in a known manner. The subscriber then determines which pages on the server 12 web site he or she would like to track. The subscriber then opens a text editor for each page to be tracked, and the code from Appendix I is pasted into the bottom of the page. Although the code in Appendix I does not provide an image on the page, it should be appreciated that code that includes an image such as a logo or the like, could be included in the Appendix I code. This would consequently both track the page and display an image thereon.
After the Appendix I code is pasted onto each page to be tracked, including an order confirmation page, the code in Appendix II, which defines a variable called ORDER, is also pasted onto the order confirmation page. This variable appears on line 7 of the Appendix I code.
The variable ORDER, among other things, defines the currency that is used to purchase the product. The currency need only be entered once, and in the example is USD for U.S. dollars. There are four other items that are included in the variable for each product ordered. In the order appearing in the variable they are first, the product name; second, the category that the product is in; third, the number of products purchased; and fourth, the unit price for the product. As can be seen in the Appendix II code, each item of information in the ORDER variable is included for each product purchased.
In operation, a site visitor using computer 14 first fills in all the information in the
An example of this process is described as follows. The variable image source constructed by the inserted commercial activity tracking script can be shown as, for instance, www.webtrendslive.com/button3.asp?usd-lawn_chair#1-1445-002-2499, corresponding to price in U.S. dollars, product name: “lawn chair #1”, product category #1445, 2 units sold at a per unit price of $24.99. Decoder software operable within server 22 reverse engineers the order to extract commercial activity data based on the source of the image requests.
When the business owner operating the website on server 12 wants to determine activity on that site, he or she logs onto his or her account on web server 18 via computer 16. After entering the appropriate user name and password, reports that are maintained in real time, as described above, are accessed, viewed, and—if desired—printed by the subscriber. Examples of various reports are shown in
In addition to viewing the reports that are maintained in real time, the account owner can define time periods during which the information can be displayed in the format shown in the enclosed reports. There is also a feature that the account owner can select to cause reports to be periodically mailed to computer 16.
In story view mode, a natural language template [block (2)] is selected from a template database. The template includes fixed natural language statements interspersed with data fields. In the template illustrated in
Preferably, each of the plurality of natural language statements—such as statements 74a and 74b—include at least one data field 76. When the data for the data field is not available, the resulting statement is an incomplete statement. The system is configured to remove an incomplete natural language statement from the template if a data field associated with the incomplete natural language statement is missing so that the missing information does not take away from the narrative.
The highlights field 98 is divided into sections illustrating the different days on which the extreme points of the measured metrics occurred. Trends can then be determined as by: number of extremes within a certain date, and number of extremes in close date proximities. From the highlights field 98 of
Reports generated using aspects of the invention present meta-data or metrics into a visual form and arrangement that enhances comprehension of complex concepts. Several examples discussed above include the natural language presentation of data using a syntactic narrative or conversational language as shown in
Each report is associated with one or more meta-data or metrics. In the example shown in
Web analytics reflect behavior patterns of visitors. The number of web page visits on weekends may be very different than how many visits to the web page occur during regular weekdays. For instance, a website that displays and comments on the current price of certain stocks would be expected to have fewer visitors on the weekends when the markets are closed. Other commercial websites may exhibit similar analytics patterns, having more visits during the week during normal operating hours. Conversely, some other websites such as leisure sites (e.g. Fandango or other movie sites) might have more business during the weekend than the weekday. The end result is that the peaks and valleys that show up on analytics graphs occur with periodic and oftentimes, predictable, frequency. And while such variations may make it obvious when weekends occur, it would be helpful to have an additional visual indicator or weekend overlay on the displayed chart or graph.
Another aspect of the invention is shown in
Selecting the button causes the system to transition to an RSS feed entry mode wherein the feed URL (e.g. http://www.acmecorp.com/pr.ss) is entered by a user of the system as in block (2). The RSS feed is standardized to have an article title field, the article itself, and a date posted field. The data posted for each event in the RSS feed is mapped to the graph in block (3).
Block (4) illustrates a user view of the RSS data superimposed on the graphical trend data. It is observed, for instance, that the last date shown (June 20) includes two RSS fee article publications. Both are posted with a label ‘A’ and ‘B’, respectively, on the ‘20’ portion of the graph. The ‘B’ article is obscured on the graph because it occurs later in time than article ‘A’. Because multiple articles occur on that day, and to distinguish it against times where only a single RSS feed occurs (e.g. flags ‘D’ and ‘C’), the ‘A’ flag is darkened compared to the others to indicate a density of events on that day. The articles, or just titles of summaries of the RSS feeds, are displayed in conjunction with the graph.
The invention can be generalized to any time of data feed, of which an RSS feed is but an example, and is not intended to be limited solely to the examples given.
Other types of data that can be compared within data compared field 164 include: Visits, Visits % Change, Page View per Visit, Page Views per Visit % change, Bounce Rate, Bounce Rate % change, Avg. Time On Site, and Avg. Time On Site % change. Other sorting means selectable within the sort field 166 include: Name ↑, Name ↓, Measure ↑, Measure ↓ (where ↓ means “descending” and ↑ means “ascending”).
Having described and illustrated the principles of the invention in a preferred embodiment thereof, it should be apparent that the invention can be modified in arrangement and detail without departing from such principles. We claim all modifications and variation coming within the spirit and scope of the following claims.
This application claims the benefit from U.S. Provisional Patent Application Nos. 61/230,982, 61/230,984, and 61/230,987 all filed Aug. 3, 2009 whose contents are incorporated herein for all purposes.
| Number | Date | Country | |
|---|---|---|---|
| 61230982 | Aug 2009 | US | |
| 61230984 | Aug 2009 | US | |
| 61230987 | Aug 2009 | US |
