How to see the workloads we are using on Azure Microsoft for Startups

Question

Hello everyone!

I have been reading this https://learn.microsoft.com/en-us/startups/microsoft-for-startups/getting-started-mfs#credit-milestones-and-requirements which says this:

• The 60-day clock starts once you have the required number of workloads, each with at least $1 in spend.
• Every workload must stay above $1 in spend for all 60 days. If any workload drops below the threshold, you fall under the required number. The count only completes after 60 continuous days at or above the target.
• What counts as a workload? A workload represents an Azure service you actively use in your solution. Each unique service counts toward your workload total, for example:
  • Virtual Machines (compute)
  • AI and machine learning services
  • Databases
  • Application hosting services

Now, I still have one question: where can I see my workloads on Azure?

For example, we are using 4 different AI models, and we don't know if they all count as 4 different workload or if they all together count as 1.

What I exactly need to know is: How can I on 'Cost analysis' know the amount of workloads we are using?"

Should we group by "Service Name"? Should we group by "Meter subcategory"? Should we group by "Product"?

^ This is the important question. What should we pick on the "Group by" dropdown to see our different workloads?

Then, second question: it says: "above $1 in for ALL 60 days", so, is it 1 dollar for the 60 days? (1 cent per day) or is it actually 1 dollar per EACH DAY of ALL THE 60 DAYS?

I suppose it's $1 per day... but... what happens if a Saturday/Sunday (days where my platform is not used by anybody) I don't have a $1 cost? will I lose it? I may have $50 every day but on Sunday I will have $0 for sure on most of the workloads.

Answer 1

Hello @Candelaria Justiniano

Based on the current Microsoft for Startups documentation, there isn't a dedicated view in the Azure portal that displays your workload count for milestone qualification.

If you're trying to estimate the qualifying workloads in Azure Cost Management > Cost Analysis, the closest option is to Group by → Service name. Since the documentation defines a workload as an Azure service (such as Virtual Machines, Azure AI Services/OpenAI, Azure SQL Database, App Service, etc.), grouping by Service name provides the best approximation. Grouping by Meter, Meter subcategory, or Product reflects billing details rather than distinct workloads.

Regarding your example, if you're using multiple Azure OpenAI models, they would typically be considered part of the same Azure AI/OpenAI service, not separate workloads. Four deployed models would generally count as one Azure AI workload, assuming they're all under the same Azure service.

As for the "$1 for 60 days" requirement, the public documentation doesn't specify whether the threshold is evaluated daily or as an aggregate over the 60-day period. Since this implementation detail isn't documented, if your workloads naturally have periods with little or no activity (such as weekends), I'd recommend opening a support request through the Microsoft for Startups portal for an authoritative answer on how the eligibility check is calculated.

Please "Accept the Answer" if this information helped you. This will help us and others in the community as well.