I am adding distinct order count measure to the cube.
I have to place it in a separate measure group for better performance
When I click on "new measure group" button, I am presented with a dialog box "select a table from the source view" . My Facts_sales table is not on the list. because the table is already used by other measure group "Facts_sales".
How could I create a new measure group for the distinct count measure?
Just create a new measure in the group referencing the fact table of interest. As long as you select Distinct Count as the aggregation function, the new measure will auto-magically be placed in its own separate measure group.
EDIT: http://msdn.microsoft.com/en-us/library/ms166573.aspx
13.In the Measures pane, right-click the Sales Quotas measure group, and then click New Measure. For more information, see Defining
Measures.
The New Measure dialog box opens, containing the available source
columns for a measure with a usage type of Sum.
14.In the New Measure dialog box, select Distinct count in the Usage list, verify that SalesQuotas is selected in the Source table list,
select EmployeeKey in the Source column list, and then click OK.
> Notice that the measure is created in a new measure group named Sales
Quotas 1. Distinct count measures in SQL Server are created in their
own measure groups to maximize processing performance.
Related
I have a couple measures in a Power Pivot data model to analyze activities by a group of consultants:
Total Activities = calculate(DISTINCTCOUNT(ServiceRecords[Activity ID]))
Service Status = if([Distinct Activities]>=5,2,IF([Distinct Activities]=0,0,1)). >> This is to assign a numerical status to customers based on volume of activities
I have a pivot table set up to slice by consultant and show the output of these measure against clients in rows.
I run into trouble when I add a new fact table (Addressable) to the model - this contains Yes/No feedback from sales as to whether a particular account is relevant for a particular consultant. This is linked to the original fact table with two typical index tables classifying consultants and classifying accounts. But the direction of the relationship is inwards from the index tables to this table.
What I'm trying to do is add a second slicer by the Yes/No field in "Addressable" to filter only the records that show Yes or No for the chosen consultant. I thought a CROSSFILTER measure was supposed to do this, but it ends up filtering the data behind calculation of the existing measures instead of just filtering the view. Any idea what I should be doing here?
When grouping by multiple columns using the following function:
groupBy:['column_one','column_two'],
columnCalcs:"Both"
What I am hoping for is a total for each group (and sub group), as well as the grand total at the very bottom of the screen.
Currently, only the last defined group (column two in the above scenario) and the grand total show. I do not receive a total, per group level.
Im afraid that it is not possible to configure the table in that way at present, you can either have calculations in the child groups or the table as a whole but nothing in between
I have a simple SSRS report which has one group and details. The grouping is by employee and the details are performance data on each. After tedious calculation, it just comes down to
select * from table
and I have SSRS do the grouping on the employee column. There are several tasks for each employee, so that is why the grouping in the first place.
My problem is, the user would like to be able to distribute these stats to the employees, and it would be easier if there were some white space between these groups (between each employee).
I've tried adding a blank row inside or outside the group, but I can't find a way to do that so it won't put a row between each task. I tried using a list, but in the end, got the same problem--the group still forced it to behave that way.
I know I can insert a page break between groups, but that would be a huge waste of paper, having each employee on a separate sheet.
Is there a way to essentially have each employee (group), be in a separate "table"--such that I would have maybe a half dozen on a sheet that could easily be guillotined?
EDIT: Here's a screenshot of it as it is now:
Table
I don't know how I could use a rectangle because the results (groups) are all in the same table. The idea would be to insert a space between each group (person).
I need to create several pivot table once and week, and I wonder if there is any efficient method to drag several item into value field. (I tried to left click on the item I need, but it will automatically go to "row" field.)
And also, is there any efficient method to set up "value field setting"?
I want to summarize value field by "sum", but it will automaticall become "count". If i have 20 items in "value field", I need to click value field setting for 20 times...
Given that you have used the PowerPivot tag then I will assume that you are talking about a PowerPivot PivtotTable not a standard Pivot.
The first thing to do is explicitly create measures (called Calculated fields in 2013) for each calculation you wish to make - not only does this mean you can choose to sum not count but also that you can specify the format you want and once it has compiled, the measure will be added more quickly that just dragging the column into the box.
The easiest way to create a measure is to right click on the table and then in the dialogue box add a formula such as:
=SUM('mytable'[mycolumn])
In the same dialogue you can name the measure and choose the number format.
Once you have the measures you need, you can create a 'set' which is simply a collection of measures/dimensions. The easiest way to do this is create the Pivot you want then go to:
PivotTable Tools>Options>Fields, Items & Sets>Create Set based on....
From there you can name the set and make alterations. Then in future you will see that set in the field list and be able to add it to you pivots with a single click.
Jacob
Is there a way to create a SharePoint calculated column that returns a count of the number of entries in a list? So If I have 3 customers in my list with the company "Starbucks" I'd like the field to return "3"
(Edited some wording for clarity per suggestion from dariom).
You may be able to get what you want with another list using a not-so-well-known variation of a lookup column.
Let's say you have a list called Companies with values in the title column like "Starbucks", "Peets", etc. Now you also have the Customers list you refer to, but the "Company" column is a lookup column pointing to the title column in the Companies list.
You can add a count very similar to what you described to your Companies list. Go to your Companies list, add a column of type "Lookup" referring to the Customers list and you'll notice that in the drop-down area where you define the lookup if you point back to the Customers list, you'll have a new option called "Count Related". This is here automatically because it recognizes that the Customers list has a lookup pointing back to this one. Select that Count Related option and now your Companies list will have a column counting how many customers are associated with that company.
No coding, Javascript hacks, or anything. Just hidden SharePoint auto-magic.
No, I don't think there's a way to do this using the out-of-the-box calculated column.
Some other ways you could accomplish this are:
Create a view for your list that with a group by on the company field and include the total count. This is easiest, but might not be exactly what you're looking for.
Create a custom column type that executes a CAML query to find items that you're interested in. There is a learning curve if you've not done it before and if the list that you're adding this custom column to has lots of rows, you'll be executing a query for each row which is inefficient - it'll be OK for a small number of rows.
Use an event handler on the list that updates a column value each time a new item is added or removed from a list. This is easier, but can also be inefficient if you have a large number of items in your list.
As dariom said (damn my slow typing skills, +1!), only the current row can be operated on with calculated columns by default in SharePoint. There are a couple of documented workarounds involving SharePoint Designer or jQuery, though.
You can get a Count of specific list items in an XSLT Data View
To do this you will need SharePoint Designer.
Right click on your SharePoint List view (ensure the list view contains the field you want to filter by) select convert to XSLT Data View. Then in the Data Source Windows select Data Source Tab and drag and drop the field you want to get a total on for the specific items into where you want it displayed in your XSLT Data View. Click on the numerical value that is showing you should get a lightening bolt icon, select the drop down and choose Count, then select again and choose Filter. Select "Click here to add a new clause" then choose your field name again and enter your unique value as Starbucks and click OK, you can repeat this process for other fields you want the totals on. You will now see the total number of Starbucks items in the list.
I got something similar to work in a way similar to Niall. Basically, I:
Based on the source list, created a Data View Web Part (DVWP) on a "test" web
part page.
Added the footer column, which gives a count.
Set the filter for my conditions (i.e., the items I want to count).
In the code, deleted the recurring items row.
I was left with just the footer, which displayed a filtered count for all the list items. I further customized the footer by taking out the shaded background. Finally, I exported this web part and imported it onto the page where I wanted users to see a total of items in the list (which met the criteria).