Hack 33. Measure Your Intranet or Extranet
Companies collectively spend billions annually on intranet and extranet projects but often fail to take the extra step in justifying the expenditure and ensuring that employees and business partners are actually using available tools. While public web sites get all the limelight, many enterprises have internal and business-customer facing sites that merit measurement as well. These can be intranetssites for use by the organization's members that can be accessed from within the premises of the organization onlyor extranets secure web sites for the organization's members or its affiliates that can be accessed from anywhere on the Internet, but require password authentication. The essence of measuring traffic to your internal sites is to ensure that the right people are using the right resources the right way. Regardless of the type of site, there is significant value in measuring the use of these types of sites, mostly derived from cost justification and usage patterns. 2.22.1. Justifying Return on InvestmentIntranets and extranets often require significant effort and expenditure to create and maintain. One approach you can use to justify these costs is to quantify the savings associated with having people self-serve information from your internal sites. For example, you may wish to quantify the time and effort associated having to take phone calls instead of having your business partners self-serve through your extranet. To do this, you would run web measurement reports to count the basic transactions (find information, change a record, make a purchase, etc.) that your site delivers, assigning a non-web transactional cost to each. The sum of these costs would be the net savings associated with building and maintaining the extranet. Perhaps an easier strategy, one anyone can do, is to determine how much it costs your average human resources person to answer a basic question like "What holidays do we get this year?" Multiply that cost by the number of visits to your corporate intranet's "vacation schedule" page, and you can begin to visualize cost savings. 2.22.2. Measuring Content and Application AdoptionIf you're responsible for content or applications on your internal site, you want to know what is used versus what is not so that you can concentrate your development and maintenance effort on the important areas. Another benefit of deprecating underused content is that internal users will likely find what they are looking for more easily. Also, note that you may be able to determine that users aren't using content and applications they should. In this case, you may want to use this information to drive departmental leaders to reacquaint employees with the corporate intranet. The easiest way to start is with a report covering the most frequently accessed pages and content on your internal site. Sorting the report ascending by visits will show you the least frequently accessed content. The next step is to group content on your web site into logical content groupings so that you can get an overview of which functional groups users are most interested in. You may also want to examine time spent on pages and groups to try and determine whether internal users are spending enough time to reasonably get any value from the content or the application. 2.22.3. Examining Internal SearchesIntranets and extranets that serve as content repositories or portals accumulate so much content over time that local search becomes the most important way visitors try to locate information. Many web measurement tools offer reports on the keywords that visitors enter into the local search box [Hack #64], provided that the keyword can be captured either from the query string in the URL of the search results page or via a page tag. Searches with no or few results can be identified if the aforementioned URL or page tag also includes a parameter with the number of search results that were found, often identifying gaps in the information content your users think you should have available. Abandoned searches can be more difficult to measure, but provide greater insight into how you can improve the quality of your internal applications. In the easiest case, the visit's exit page will be the search results page, telling you that the user just didn't see compelling enough results to bother clicking on links. Usually users conduct additional searches before they resort to navigating the content hierarchy instead of using search. Next-click analysis from the search results page can provide valuable insight in this regard; any "next" click that opens a content folder instead of an individual document may be a click on the site's navigation options instead of the search results. Ideally, the next click after a search is on a documentone that hopefully contains the information the visitor is looking for. 2.22.4. Named User AnalysisAs intranets and extranets are almost always secure sites, a username can be usually captured and associated with the visit. Some sophisticated web measurement tools can even look up the username in your company's user database and translate it into the user's demographic information, such as their name, role, department, branch or company location, or training certification levels (Figure 2-22). Figure 2-22. Tracking named usersIf you want to get really sophisticated with measuring adoption rates, you can combine your organizational database with the web traffic database in suitable web measurement tools. Allowing you to measure beyond visitors, you can then calculate the percentage of persons in each department or branch that have adopted content on your site. Some enterprises also combine their web traffic database with their customer service call center logs to determine which customers are not using of their extranet and are instead calling into the hotline, incurring higher costs of service. 2.22.5. Tying It All TogetherIf you have a web measurement application with cycles to spare, here are five things you should do right away to make sure your intranets and extranets are being used:
The last item is perhaps the most interesting since web measurement reports never really tell you that much about intent and satisfaction. By comparing the information you have available with the requests you get offline, you can hopefully see a trend, helping you identify additional content that needs to be added or made more visible. Akin Arikan and Eric T. Peterson |