Add to Count Based on Conditions

[pmstahl]

To provide context to my issue: The goal of the data provided is to track sales through their life cycle filtered by lead source (Facebook, Referral, Yard Sign, etc.)

We have refined our investigation to four key status points:

Door Knock Count: How many doors did our sales team knock?

Set Appt Count: How many appointments did our sales team make after talking to home owners?

SAO Count: How many signed deals did the appointments generate?

JTIs with SAOs: How many SAOs actually became Jobs Turned In 

We pulled data from our Sales Rep's App and gathered this information. There is a problem: There are more SAO and JTI than Appts. If a job has a JTI, it should have an Appt that led to that JTI. Our sales reps were not using the app correctly and missing status updates.

To identify the issue I drilled down into the numbers and found that 5 sales reps used the apps incorrectly. As you can see in rows 1-4 and row 7, the door knock count and set appt count are blank but should be showing a value of 1 (I went to the app and verified that those steps were skipped all together and thus not in the source data). 

The DAX to calculate the various columns looks like this:

What I need is for the Set Appt Count to display a 1 if the JTI count is 1. I took a stab at this but am a beginning PowerBI user (I am in my 2nd week):

My thought process: If the JTI is > 0 and the Set Appt Count is blank, then just copy the JTI number, otherwise, run the calculation for Set Appt Count as normal.  I got what I wanted at a granular level:

In the new Appt Count If Blank column I have a "1" for each row that has a JTI and the correct Appt Count if the JTI is blank. However, in the total row it still only shows 3 below the Appt Count If Blank instead of 8.  When I remove the Sales Rep Name (mostly hidden for confidentiality), I lose the granularity and my Door Knock, Appt Count If Blank, SAO Count, and JTI read 5, 3, 7, 6 instead of 5, 8, 7, 6. 

My two questions are:

1. Why is my DAX equation recognized and accurate on the granular level of Sales Rep but the total row is incorrect? 

2.  Is there any way to write a calculation that provided my desired output filtered across multiple granular levels (by month, fiscal week, lead source, drilled all the way down to sales rep, etc.)?