Alma Analytics: Tips & Tricks

Brown Bag Topics

 

Quick links

Below is a collection of videos, and text notes of tip, tricks and techniques learned over the years.  Many were originally recorded in Shockwave, and were been re-recorded in September 2018.  These, and future videos will be delivered through YouTube, and stored locally (for U of M staff only) on the departmental drive at Alma Tips and Tricks.

Known issue with Physical Items subject area

Nichole Chisholm asks:
"When I add the "Fund Ledger"."Fund Ledger Code" field, the "Physical Item Details"."Num of Loans (In House + Not In House)" field goes blank. Is this a known problem?"

This is a known issue and is not fixable (because of the convoluted way that Ex Libris has implemented this. Ex Libris does announce it, though in a slightly unusual way. If you mouse-over (not click) the empty folder named "Fund Information">"-- Fund Instructions --", the message explains the limitations of the whole Fund Information dimension. Particularly, it states 'The measures (yellow fields) from folder 'Physical Item Details' should not be used with (anything) from folder 'Fund Information.' If you try this, you will get incorrect results'

As for workarounds, the most obvious would be to run two reports and then merge them via post processing. A little more tricksy, but also open to possible flaws, is taking advantage of the relatively few funds that are involved, and create separate reports for each fund.  I use that report as a "feeder". Then in a report named "output" I take the barcodes reported from 'feeder' and report out the circulation. It generates four separate reports, but at least you won't have to do post-processing.

Filtering within a measure in Alma Analytics

This video shows how filtering within a measure can provide data in a more readable format.

Using Presentation Variables in Analyses

Presentation Variables provide another way of displaying data that was inputted via a prompt, without having to provide individual reporting lines.  
 

Dates

Date Defaults in Prompts

How to set a default date in a prompt that automatically adjusts.  In other words, how to prompt for "10 days ago."

How to get the most recent complete fiscal year

select case when month (current_date) > 6 then
  concat( 'FY-', cast((year(current_date)+1) as char))
else
  concat( 'FY-', cast(year(current_date) as char))
end
from {Subject Area} (e.g. "Funds Expenditure")
 
How to use string functions in analytics to search for fields where case may not be consistent.
 
How to minimize scrolling by user who wish to export reports from dashboard by adding an Export link at or near the top of the page.

Alma Circulation Statistics
How to begin compiling a report on total circulations within the Fulfillment subject area, focusing on filtering out In-House Loans and where pre-Alma circulation counts are stored. (3 parts)

Using RCOUNT() to get around Ex Libris's line limit on export
65,000 has been almost a curse word among analytics users.  Here is a fairly easy way to (kinda) get around that limit and make your analysis work for larger sets.
 
The default formatting choice for Columns in Analytics is 'Suppressed'.  This makes for a report that's easier to read, visually, but many users wish to manipulate the data further in Excel.  To do that, they need to change the Column Format to "Repeated".  Here's a quick and easy way using "Import Formatting" to apply this without having to manually update multiple columns during design.

Regular Expressions

Testing for specific characters/patterns
Stacie Traill presented me with an interesting question- could she use Regular Expressions in Alma Analytics in order to test some data she knew was erroneous.  The answer is yes, but I also offered her an alternative (and somewhat easier) method using the string functions in OBIEE.  Here are videos showing both methods:
Testing for specific characters

Semi-colons in Regular Expression

In using regular expressions to identify subject headings, Stacie ran into this issue:

  • This  regular expression works:
    EVALUATE('REGEXP_INSTR(%1,''Illegal\saliens.{2}United\sStates--'')',"Bibliographic Details"."Subjects")
  • This throws an error:
    EVALUATE('REGEXP_INSTR(%1,''Illegal\saliens.{2}United\sStates;'')',"Bibliographic Details"."Subjects")
  • So does this:
    EVALUATE('REGEXP_INSTR(%1,''Illegal\saliens.{2}United\sStates\;'')',"Bibliographic Details"."Subjects")

The problem, of course, is that pesky semi-colon, which OBIEE sees as a delimiter, and there's not an easy way to escape it out (as you've found). My solution is to bypass this with:

EVALUATE('REGEXP_INSTR(%1,''Illegal\saliens.{2}United\sStates''||chr(59))',"Bibliographic
Details"."Subjects")

using the CHR() function to pass through the semi-colon.

 
Action Links  
Action Links are an advanced technique in Analytics to allow more interactivity in your reports and dashboards.  This three part series came out of development for the E-Resources Dashboard, based on ideas that Sunshine Carter gave.  The first video is just a demonstration of what Action Links can do in a report or dashboard.  Parts 2 and 3 show how to build the actions demonstrated.
 
 
 
From ELUNA 2018- how to save a column so you don't have to reuse it in future reports.
 
How to calculate hours or minutes for short period loans.

Comparison of print holdings across libraries

Lisa Wheeler asks:
"Is there an easy way in Analytics to compare print holdings across libraries? I've been asked to compare the Duluth Tweed location to other system campuses, having already done it against Martin Library and AILRC
manually via Excel duplicate reports."

The solution is to use the filter "is based on results of another analysis".  The first report, which I call a 'feeder' will identify the MMS IDs for the location (Duluth Tweed here), only.  The second report will use this filter and then can be run against other locations/libraries (either specific to the location, or excluding Tweed to search against all) to identify when a title is held in Tweed and somewhere else.