I was looking through my LiveWriter queue this week, and realized that I had a couple of posts left over from the last several months that I never posted.  This is one of those posts.  I originally wrote it for Microsoft Project 2007, but it should be just as applicable to the 2010 version.  (The screenshots are from 2007.)

This is a question which has come up a couple of times both at work and in the newsgroups over the past couple of weeks.  The challenge has been that one of my users wants to flag all critical tasks in a schedule by using conditional highlighting, i.e. to make the background of the cells turn red.  Sounds easy enough, right?  Other planning software packages can achieve this with no sweat, right?  In Microsoft Project, you simply go to Format > Text Styles, select Critical Tasks in the drop down and enable a red background:

Unfortunately, that doesn’t work for milestones.  Huh?  Some might say that milestones shouldn’t be critical, which is perhaps a religious argument I hope to avoid for now.  From a Microsoft Project standpoint, milestones definitely can be critical.  Check out the following screenshot of a simple path with the Critical field displayed:

Here we can see that the milestones are clearly flagged as critical – yet they’re not showing up with the red highlighting.  The trick is to review the list of text styles available in the Text Styles Dialog Box:

(Note that Project 2010 has the same options with the addition of Inactive Tasks at the bottom of the list.)  As a general rule, the items at the bottom take precedence over the ones at the top.  In our case above, what’s happening is that the Milestone formatting is overriding the formatting we set for Critical tasks.

To fix this, we need to create a custom view.  We’ll call this view “_LOC Gantt Chart.”  Once we’ve created this view, we need to create a custom filter:

Then we go back into the Format > Text Styles dialog box for our new view, and configure Highlighted Cells to appear red.

Now, here’s the feature that a lot of people have overlooked….you can apply the Highlight Filter as part of a custom view.  Simply click on View > More Views to edit the _LOC Gantt Chart view.

Apply the edited view, and that’s how you can get all of your critical tasks to show up with conditionally formatted cells:

Caveats

A couple of caveats:

1) If the user manually changes the formatting of a cell, that cell is no longer subject to conditional formatting.  For this reason, in any view using conditional formatting of text, it is imperative that users are trained not to manually change the formatting of a cell.  The same principle applies to the use of custom macros that may be used to apply conditional formatting – an approach that I’ve seen implemented in a number of circumstances.

2) Change Highlighting will override the highlight filter.  For this reason, you may opt to turn off Show Change Formatting in the View Menu to preclude interference with any conditional formatting.

Enhancements for 2010

Project 2010 appears to work pretty much the same way, with a number of improvements:

1) For those fuschia-lovers out there, you can actually set the text color to fuschia.  As someone in the newsgroup pointed out a week or two ago, there seems to be a bug in Microsoft Project 2007 which precludes the use of fuschia.  Don’t believe me?  Go ahead, try it.  Set critical tasks to appear as fuschia.  Obviously, this is a major step forward in usability.

2) Getting slightly more serious, I can now even choose between multiple shades of fushcia, as is discussed in the Project Team post here: http://blogs.msdn.com/project/archive/2010/02/10/project-2010-introducing-millions-of-colors.aspx

3) And now in Project 2010, we have the option of using the strikethrough text formatting, which is the default setting for inactive tasks.

That little post on Portfolio Analysis in Excel that I threw out into the Interwebs a couple of weeks ago seemed to engender a surprising amount of conversation – both online and off.  As I’ve stated before, I am not a math major, but figuring that as of this writing we had three distinct comments posted – and generally most posts generate zero comments, then by my calculations, if I figure 3/0, we had approximately an infinite% increase in postings (+-10%).

In my books, that puts PPM Simulation postings on par with Earned Value postings as potentially being the third rail of the EPM tool blogging space.  I guess this means that when I develop some sort of controversial content around simulating the application of EVMS to PPM, the kraken will really be let loose.

But seriously, my take is:

1) People do read this blog – and for that I thank you.

2) We need to take a look at what really differentiates Microsoft Project Server 2010 from the other PPM tools out there.

3) What we’re seeing is the fall out from the fact that Project Server 2010 is the first Microsoft EPM Server product to perform its own calculations *on the server*.  (and yes, I know we said the same thing about the 2007 release, keep reading to see what I mean)

The main gist of the comments and offline discussions were around whether or not it was appropriate to document/reproduce some of the specific calculations used in the Project Server 2010 PPM module.  My goal with this current post is not to get into a discussion around that topic, but more to direct the conversation to what I see as the key issue, which is the question of what differentiates the Microsoft offering from the other products out there.

The Report

Interestingly enough, this kerfuffle (if indeed 3 comments constitute a kerfuffle) occurred right around the same time that the latest Gartner Report on Project Portfolio Management tools was released (see here for the announcement and here for the full report). 

A total of 31 vendor packages were evaluated.  According to the report, the key criteria for inclusion was that the products support

1) Core project tracking/management

2) Time/cost management

3) Portfolio level management and reporting

4) Standardized demand collection mechanism

Here’s what Gartner had to say about Microsoft’s EPM package core strengths:

For example, in the last couple of weeks, I have been working with a client to develop a summary reporting structure using a third party tool.  In the process, they found that % Complete was being calculated differently on Summary Tasks than Child Tasks.  That’s when they come to me.  My job is to figure out behind the scenes how Microsoft Project calculates % Complete for Summary and Child Tasks.  Once I’ve created an adequate model of those calculations, I can develop a formula to consistently apply Child Task % Complete calculations to a Summary Task.

As I’ve heard many times in training classes, the role of the consultant in an EPM tool implementation is to ensure that the tool is predictable, and that the end users understand the underlying calculations.  To the best of my knowledge, that’s not a role that only applies to the desktop tool.  It applies to the server product just as well.  The only issue is that until the 2010 Project Server release, we really did not have all that many calculations to explain on the server side.  The ones we did have were all replicated in the desktop scheduling engine.

That’s  an observation that I hadn’t quite realized until I sat down to write this post.  Project Server 2010 is arguably the first version of the Project Server family to feature server-side calculations.  Now wait a second, you’re saying.  Wasn’t 2007 the first version to feature server-side calculations?  Technically, that’s true, but remember that the calculations featured in 2007 are the scheduling engine.  Thus in 2007, whenever a client had a question about how updates were applied to the schedule, I could bring up a copy of Microsoft Project and explain it away.  In 2010, with the portfolio prioritization and optimization engines, we now have a whole set of calculations that have gone to the server.  This now requires a slight shift in mental models, insofar as we now can’t easily draw that distinction between calculations on the Server and calculations on the desktop.

As a consultant, I now have to be able to explain the calculations that are happening on both sides of that fence.

This question came up in the newsgroup a little while ago, and intrigued me enough to play around with it.  The question was how to add Departments to the Issues and Risks reports in Project Server 2010, so that you can report a rollup of how many issues or risks exist on projects by Department.  I played around with it a bit, and the answer turned out to be pretty simple.  Basically, we just have to add the field in the appropriate ODC files within the BI Center.

This answer though is indicative of the ease with which reporting can be modified in Project Server 2010.  If the right field doesn’t occur in the OLAP Cube, we can simply create a join with a table and borrow the field from that table.  In this case, that task is even easier because the sample ODC files included in the default installation already have the right tables joined.  All we have to do then is add a string of text to the ODC file to surface the field from the Project table.  (And for those of you who are not so proficient at writing SQL queries, Microsoft has provided the Report Wizard Solution Starter to generate the ODC code correctly.)

What also intrigued me is that I instinctively went to solve the problem first the same way I would have in Project Server 2007, i.e. trying to get the right field in the right cube.  In this case, that’s not possible using the out of the box interface, and it was actually easier to just bring in multiple tables to Excel and join them there.  Here we have an excellent example of how removing the old Data Analysis views actually force us to use a much more robust reporting methodology in 2010.  In retrospect, it would seem that Data Analysis views were a crutch that held us back from exploring better reporting options in 2007.  And for the record, I see no reason why this solution wouldn’t have worked in 2007 (with the possible exception of WSS not supporting Excel Services).

I suspect that this will be a recurring thread: how to surface specific fields in specific reports…

Verifying the OLAP Cube (Not)

Before talking about how to resolve the issue, let’s talk about how not to resolve the issue.  My first step was to verify whether or not we could simply add the Project Departments to the Issue cube.  To do so, I selected Server Settings, and clicked on the OLAP Database Management option.  I selected the OLAP Database, and Configuration from the toolbar.

I looked for an Issue cube, of which there was none in this screen.  The closest thing I suppose to an Issue cube is the Project cube, and as far as I can tell, Departments are added to that cube by default on install.  So, I confirmed that Project Departments have been added to the Project Cube, and rebuilt the cube.

At the end of the day, this really didn’t do anything for me…..but that’s exactly how I would have attempted to solve the issue in 2007 to get the data in the Data Analysis views.

Modifying the ODC File

So now, how do we solve this in 2010 (and in retrospect how could we have solved this in 2007)?  What we have to do is find a unique identifier in the Issues cube, which turns out to be the Project Unique ID.  We then need to find a table within the Reporting database that contains the Departmental data, and join the two using the Unique ID field.  That’s relatively easy.  It’s made easier by the fact that the Issues ODC file contained in the BI Center on install already is configured to join the Issues cube with the MSP_EpmProject_UserView table.  All we need to do is add a string into the ODC to add one more field from the EpmProject_UserView table and our job is done.

So to do that, we first navigate to the BI Center > Data Connections > English folder, and click on the Project Server – Issues Report.

When you open the file, that should trigger the launch of an Excel file with the ODC stored as a data connection.  (Feel free to check out this earlier post on modifying the ODC file: http://blogs.catapultsystems.com/epm/archive/2010/01/18/modifying-the-default-odc-files-to-filter-on-specific-projects.aspx)

Click on the Connections button in the Data tab to access the ODC file.

Select the Project Server – Issues Data connection, and click on Properties.  Then select the Definition tab.  That should result in the following screen:

Add the following string into the Command Text section…

ProjectOwner.[Project Departments] as [Project Departments], 

(This pulls the Project Departments field out of the dbo.MSP_EpmProject_UserView  table and surfaces it in our Excel file)

Hit OK, and confirm that Project Departments has been added to the Excel workbook (it should be all the way over on the right).

Now export the ODC file from the Data Tab > Connections button > Definition tab:

You may now upload the ODC file over the original one (making sure to keep a copy of that with the “_OLD” suffix) or make a new report based off of the existing Issues sample report.

Coming soon…..how one would accomplish the same thing using the Report Wizard Solution Starter, assuming that I can figure out how to make it work.

One of the schedulers I support was asking me this question the other day, and it seemed like it might make a nice tip and trick.  Hence, you’re seeing it here as a blog post.

The question was how to display both Critical and Near-Critical Path tasks on the Gantt Chart – basically how to flag tasks with Total Slack below a specific predefined threshold.  You’d want to do something like this to visually identify potential risks in your schedule, and it might be used in conjunction with the Resource Critical Path technique identified in Eric Uyttewaal’s book Dynamic Scheduling with Microsoft Project 2003.

You could, of course, simply expose the Total Slack field within the table on the left of the Gantt Chart – and then add indicators and/or grouping to identify tasks with low risk. You could also simply add the slack as a graphical element to the chart in using the Format tab in 2010. This technique would be used in addition to that method in order to add a graphical element. 

So let’s take the following project and see if we can modify it to display the Near Critical tasks graphically.

In Project 2010, we have a couple of options.  First off, we can display the Critical Path tasks by clicking on the Format > Critical Tasks option.

We can also toggle a Slack display by selecting the Slack option on the same tab….

But we’re going to graphically change the display to display the Near Critical Path.

Configuration

The first thing we need to do is to set up a couple of custom flag fields to tell us what is “near critical.”  In this case, I want to create two flags, one for “Imminently Critical” and one for “Near Critical.”  You’ll note that I use the numbers “2401” to represent 5 days.  This is because Project calculates duration in units of minutes.  Therefore 5 days X 8 hours X 60 minutes = 2400.  I add 1 to that total to make 2401, thus simplifying my calculation.  Instead of using “less than or equal to 5 days,” I just use “less than 5 days + 1 minute.”

You’ll want to set the calculation for Summary Tasks to None as summary tasks should not be included on the Critical Path.

Once we confirm the formulas are working correctly, we need to modify the bar styles.  I doubleclick on the Gantt Chart, and add two new bar styles using the Critical Task as a guide.  You’ll note that the Critical Task format only appears in the bar styles when Critical is checked on the Format tab.

…and now my Gantt Chart displays the following (note that you probably want to modify the patterns in a real implementation for your colorblind colleagues).

…and that’s about it.  In future posts, I hope to visit the concept of Ribbon customization and figure out how to add a formatting check box to the Format tab to include the custom graphical elements we’ve just added.

Chris Vandersluis hits another one out of the park with his article on implementing EPM (here).  This is from his From the Trenches series posted to the Project Technet page.

Definitely required reading if you’re contemplating an EPM engagement.

If you’re in Austin on 8/10, you may want to check out this upcoming event by Dan Patterson of Acumen.  Looks to be quite interesting.

And after that, you can check out some Korean tacos at Chi’Lantro.  I read about them last week, and now am looking for any excuse to make it out to Austin (or to go find a competing outfit here in the Houston area.)

Note that registration closes tonight.  Not sure what the walk in policy is.

From the e-mail….

You are invited to a special meeting jointly sponsored by PMO LIG and Tech LIG.

Description: While the Gantt chart has been one of the mainstream schedule reporting components for many years, it still lacks the ability to provide easy visualization and true insight into project quality and performance. This presentation introduces an alternate technique called project ribbons that provides a means of slicing and dicing schedules and subsequently analyzing them using advanced schedule metrics.
Project ribbon analysis provides value throughout the project lifecycle: it improves the quality of the underlying project schedule; it gives a unique insight and control during execution and helps with project forensics during closeout.
Project ribbon analysis can also be used to optimize and accelerate schedules. Through a case study (World Trade Center re-build project), the approach for such acceleration is demonstrated.

About the Speaker: Dr. Dan Patterson PMP is the founder of Acumen and is a recognized thought leader and visionary within the project management industry. Prior to starting Acumen, Dan was Executive VP at Pertmaster leading the Americas business unit from inception through to acquisition by Primavera. Dan is also responsible for developing a now widely accepted integrated qualitative/quantitative approach to risk analysis through facilitated risk workshops. His depth of knowledge extends well beyond risk into schedule analytics and project optimization.
Dan has a proven track record across multiple industries including A&D, government, energy and EPC. As CEO of Acumen, Dan now places focus on Acumen Fuse, the most advanced & comprehensive project analytics tool available for MS Project, Primavera, Deltek, MS Excel and more...
More information can be found at www.projectacumen.com

Registration: Please register on-line by clicking Register Now! or by visiting the event link on the PMI Austin Chapter website listed to the left on this page.

NOTE: If you have difficulties with completing the on-line registration process, please send an email to PMOLIG@austinpmi.org to register for the event. Cash or check payments will be accepted at the meeting; however, credit card payments can not be accommodated. We apologize for any inconvenience this may cause.

I wasn’t planning on pushing anything out today, but a couple of really good blog posts on Project desktop caught my eye, and I wanted to make sure you all saw them:

1) Dale Howard from MS Project Experts on deploying the default 2000-2007 macros to Project 2010.  These macros have always been underused and underappreciated, but the Format_Duration macro has saved my project plans on many occasion.  I’ve also got a post on the Rollup_Formatting macro queued up and almost ready to go out (Stay tuned), so this makes a nice segue.

http://www.projectserverhelp.com/Lists/Posts/Post.aspx?List=f884b43f%2D61db%2D4c4e%2D94a3%2D38f21a8efdad&ID=140

2) Carl Chatfield over at ProjHugger on easily creating custom program views using the Timeline view and Visio.

http://www.projhugger.com/projhugger/2010/07/project-and-visio-friends-for-life.html

Looks like I’ll be speaking on Project Server at the Houston Techfest on October 9, 2010.  (More Information) 

We’ll see if that MVP designation makes everyone think I am smarter – or if it just unreasonably raises their expectations.  The event is free, so I can definitely guarantee that if you attend my presentation, you will at least get what you paid for.

Hope to see some of you there….