The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.
Hi all,
I would like to see how many distinct A,B,C buying customers there have been in each month.
Then i want to see how many customer meetings have been booked per A,B,C customer in each month.
Finally i want to calculate:
number of meetings booked / number of distinct buying customers
per country per month and per ABC category.
I have 3 tables
"Meetings" = Table with records of booked meetings with customers (22.000 rows)
"Buying Customers" = Table with records of each item order transaction (1.350.000+ rows)
"SO Profit Category" = It's a table made for sorting purposes to order A,B,C customers.
A,B,C customers are in the field: Profit Category
Business Unit and Country2 is the country name. They are 100% spelled same way in both tables.
But i can't join these two tables as i want. And i don't know what to do to fix it 😞
Maybe i am approaching this problem incorrectly - its not about connecting tables?
The customers don't necessarily have to be the same customers.
Its just number of meetings held with ANY A customer
divided by
number of meetings budget (which is same as number of distinct A buying customers per month per country).
So the specific customer don't matter only whether its an A, B or C customer and their country.
@Anonymous , Bring A,B,C customers from Profit Category to Customer, but that would only be one values
example : correct column names
Maxx(filter('Profit category', 'Profit category'[Business Unit] = customer[Business Unit] && 'Profit category'[Country2] = customer[Country]), [A,B,C customers])