I am looking for examples of how other companies are measuring adoption of the WebFocus toolset. We currently have been looking at the Resource Analyzer - User Reports to see historical use changes month to month but wonder if other companies have found better ways to do this?


Thanks

This message has been edited. Last edited by: Tamra,

We wrapped the WebFOCUS WFServlet calls in a Filter class (java) on our Tomcat Application server. That class writes log lines to a database table in one of our RDBMSes.

That table can then be queried for all kinds of statistical info, as long as you store the necessary metrics of course.

I figure Resource Analyzer does something similar, but I'm not familiar with that product.

We do not have access to the Resource Analyzer, it's been turned off (we have separation of IT/Business Intel). So, we created a fex to include in reports we'd like to track and it appends each use to a log table, so we can see who is using it when. We've also been looking at the Repository Folder but haven't had much time to sit down and figure it all out.

There are some benefits we have found so far. For example, if a business unit is requesting additional functionality or reporting, but is currently not utilizing the already built reports (that they requested to be built), we have some concrete proof to defend our CBA and deny or modify their request (fyi, that hasn't happened yet). The opposite is true as well, if a BU is heavily using the reports, then we'd be more likely to complete their request depending on the CBA.

Also, this allows us to identify our power users, who can then bridge the gap with the less experienced users - this frees up some of our time with answering questions, explaining reports, etc.

I do as kevkelly. we have a wee fex in the baseapp that we include in every reporting fex. grabs the userid, the app and fexname, the output format, and a couple of generic fields we can fill however we want, fex-specific.
a random number is the last field, very important, avoids possibility of dup records.
we have a focus db in each app, in each domain. We do as kevkelley, when someone asks for pdf output, or an xl pivot table, we can prove that no one on the planet earth has ever requested such a format from within this domain. Also, we can tell if people are doing their jobs, and running the reports they're supposed to be running, provide reports to their managers

We use our Filter Java class for the same purposes as kevkelly and susannah, but without the requirement to remember to include a fex in every report - The Filter is on or off, server-wide.

It does require someone with a bit of Java knowledge to implement it though. But then again, that hasn't necessarily have to be someone within the company.

We store Server name (A256V), Timestamp (HYYMDs), User name (A50V), Fex name (A100V), Fex parameters (A1024V) and Request duration (P12.3).

As you can see, we can determine Request duration. That's because both the HTTP request to WFServlet and the response from it pass through our Filter class, so we can just subtract the timestamp of the request from the timestamp of the response.

Since all our fex-names are prefixed with a project-code, we can easily relate them to a project. That does take some discipline, of course...
Obviously, anything that doesn't go through WFServlet is outside our visibility, but in our case that is convenient, as it excludes all static content (images, javascript, html files, style sheets - anything that's not a fex).