Comparison report does not recognize similar tasks (Project 2013) RRS feed

  • Question

  • I have two projects that I import from an excel file that I want to compare. These projects are created using the same import map and for control purposes are identical except for one additional task in the second file.  Part of the import map is a unique ID field that is generated and imported so each task has a unique ID (that is the same in both file versions).  When I run the comparison report in Project the report shows all tasks from both files but fails to recognize like tasks and so does not create differences.  I just end up with twice as many tasks in the report.  The  two files cannot be any more similar but the comparison report treats all tasks as unique from one file project to another.  What could I be doing wrong?
    • Edited by roscog Tuesday, August 16, 2016 6:27 PM
    Tuesday, August 16, 2016 6:15 PM

All replies

  • roscog,

    As far as I know the Project Compare utility does use the Unique ID as the indexing reference for its comparison. However, the Unique ID field is not something that the user can control and therefore your import of that field is ignored by Project. I'd take a look at the two files after import and carefully examine the Unique ID field of both. Is it in fact identical for each task in each file? My guess is that it is not and that's why the compare utility is not working as you want.

    If by chance the Unique ID field tracks one-for-one between the two files, then there is more to the compare utility indexing. I do know the utility had to be revamped starting with Project 2010 as it did not work reliably previous to that. Perhaps it now uses a different method of indexing between the two files (GUID field?), I just don't know.

    Question, why are your files in Excel? It might be easier to compare them there.


    Tuesday, August 16, 2016 6:31 PM
  • JOhn, thanks for the reply.  My files are in excel because the tasks are generated from another application on a weekly basis and fed to me in excel format and needs to be imported to Project anyway.  I am interested in seeing the differences over time at a glance and feel I can do this easier in Project.

    The Unique ID field in Project is mapped to an excel field where the identifier is located.  In the import map I designate this as the Merge Key, as I will be merging the new data with the existing file and comparing it to the previosu version.  The Unique ID field in Project does in fact have the imported numbers from my excel file (so whether it accepts them as the true identifier or not, my numbers are presented in the Unique ID field) and are identical in both files.

    Tuesday, August 16, 2016 6:47 PM
  • roscog,

    I did a couple of tests using a very simple file with a few linked tasks (5) and no resources. I saved the file and also exported it to Excel with these fields: Unique ID, Name, Duration, Start, Finish, Predecessors, GUID. In Excel I changed the duration of the second task which would cause a ripple through the other tasks on import.

    For the first test I imported the file back to Project as a new file. I only imported the Name, Duration and Predecessors fields as that is the only basic data needed to replicate the original file. As such the original and imported versions of the file looked identical including the GUID values. Then I ran a compare. It basically said all tasks in both files were unique although there were no values in the differences field of the comparison file. In other words, the compare "failed". Interesting.

    For the second test I opened the original file in Project and then did an import using the Unique ID as the merge key. Again I only imported the basic fields (i.e. Name, Duration and Predecessors). Then I ran a comparison of that file with the saved original. A little bit to my surprise, but the compare faithfully told me which fields were different, Duration, Start and Finish in accordance with the changed duration in the Excel file.

    Based on my simple test I'd say that the comparison should work on data imported from Excel using the Unique ID as a merge key but you need to be very careful what data is imported.

    Hope this helps.


    Wednesday, August 17, 2016 12:10 AM
  • John, based on your efforts I ran some more attempts with my files with the following findings:

    If the two versions of Project files are created as new projects, the GUID is different for each task and will be considered unique tasks in the comparison report.

    This is consistent whether I use my defined Unique ID or let Project create the Unique ID for each task.

    If my second excel file is MERGED to the first file, saved and then compared to the original file, then the comparison report is correct.  Unfortunately for me this creates other problems in my project structure (Added tasks are not retained under the appropriate rollup tasks but added to the bottom of the project). 

    So in summary it appears as you say that the comparison key is the GUID field, and to compare two project files one must be created from the other by a merged import, or Project will not recognize equivalent tasks. 

    Wednesday, August 17, 2016 5:33 PM
  • roscog,

    Interesting because when I imported as a new file, the GUIDs were identical, which surprised me a little, but then I did use a very simple file structure.

    And you are correct about merging new tasks, they are added to the end.

    If the conditions for using the Compare Projects utility in your scenario just won't meet your needs, I do have another option for you if you are interested. Years ago (circa Project 97) I wrote a macro to compare two Project files. That was before Microsoft developed the compare utility for Project. I hadn't used the macro for many years but I did run a test yesterday. The macro uses the Unique ID field as the reference and it worked fine with the imported file (i.e. compare the original in Project to a revised imported version from Excel).

    Are there particular fields that you want/need to compare?


    Wednesday, August 17, 2016 6:09 PM
  • My file has about 400 lines in total.  The project is structured with about 80 level 1 (summary) tasks each having 1-6 tasks underneath (level 2).  Each week there may be differences in these individual tasks such as additional tasks added, some tasks removed, dates shifted, resource names changed, or status changes. these are the differences I (and others using the Project files) am/are looking to identify quickly.  Unfortunately the source files I use also have some extraneous lines that are not pertinent and I have ways of sorting them out but is extra work - so I have created some macros to do this for me.

    Wednesday, August 17, 2016 6:45 PM
  • roscog,

    Okay so it sounds like you only need to compare the basic task fields but I don't quite understand what you mean by "extraneous lines". Which file is the "source file", the original file in Project or the imported file from Excel?

    The macro I referred to, as currently set up, compares Name, Duration, Start, Finish and Text2 but that is easily modified. It uses the Contact field to annotate changes in each file (i.e. whether a field value is different and if so, highlights those fields that are different, and if a task has been added or deleted).

    If you're interested in the macro, you will need to contact me directly at the address below. I will ask some questions.



    (remove obvious redundancies and the 7th character is a letter)

    Wednesday, August 17, 2016 7:51 PM