How To Calculate Power BI Embedded Licensing Cost For Quotations

Since you are looking to find information on Power BI Embedded Licensing, we expect that you already know about PowerBI Embedded. However, for starters,  they are capacity based SKUs for ISVs and developers through Azure targetted at ISV/Developer offerings. It allows ISVs to embed stunning visuals and analytics directly within their app or websites, where decisions are made, alongside their business logic.

Also Read: PowerBI Interview Questions

Power BI Embedded Licensing

Just like PowerBI Premium, Microsoft has released a set of SKUs for PowerBI embedded to enable developers to incorporate this functionality into their applications. They are as follows:

The way Power BI Embedded will work is that when it is running, it will charge you x amount of dollars per hour. Therefore if you keep it running for a month, then the cost is equal to 24*30*x = $720x

Now x is the unknown which depends on usage, and every SKU has a different per hour charge.

How is that defined?

It is defined by the highest number of page renders (i.e., interaction with the page) that a particular SKU offers. So let’s say for example there are 2 SKUs. A1 and A2. A1 can handle 300 page renders at the peak hour. A2 can handle 600 page renders.

If your organization has 250 page renders at the peak hours then you will go with A1. If your organization has 450 page renders at the peak hour, and you still go with A1 then the service will fail at the peak hour.

Remember the total number of renders doesn’t matter here. So let now say Org 1 has an average of 100 page renders per hour and is consistent throughout the month. Therefore, in effect, the total number of page renders is 100*24*30 = 72000 page renders.

Another organization, Org 2 has a peak page rendering of 600 for just 2 hours a month and rest of the time it has an average page rendering of 10. By this calculation the total number of page rendering equal 2*600 + 10*(718) = 8380

Because Power BI Embedded doesn’t depend on total usage but rather peak page rendering; Org 1 can still go ahead with A1 SKU, but, even though Org 2 has a much lesser page rendering has to go with A2 SKU.

At this point, you might think that this is unfair as Org 2 has to pay a lot just for having peak rendering for 2 hours. Herein comes the magic of Azure.

Power BI Embedded is a pay as you go service. Therefore when do not use the service, you can just pause it, and you won’t be charged.

So lets now complicate the things a little bit for Org 1 and Org 2.

Usage for Org1 remains the same, and therefore they go with SKU A1 for the entirety of the month.

They will end up paying 24*30*1.008 = 750 dollars which is the cost of the A1 SKU.

1.008 is the price per hour of SKU A1

Since usage for Org2 is sporadic, let us calculate only the hours when it is working. Because we know the peak rendering is 600, therefore, we have to go with SKU A2. If the service is kept running throughout the month, the amount would be equal to around $1500. But instead of that let’s say the service runs only 5 hours every day and not during the weekends. The updated cost would now be

5(number of hours per day) * 20 (working days in a month) * 2.008 = 200 dollars only

2.008 is the price per hour of SKU A2

Now, coming to a scenario as an Example.

You have around an average of 24 users accessing an average of 8 reports at any hour. The peak rendering would be around 192. Therefore this means they should go with the cheapest SKU that is there. The problem, however, is the Monthly dashboard pbix example file. Let’s say that this is accessed by 100 people ten times (a day I presume). This puts the maximum page rendering in 1000 which means that the SKU they would be looking at now would be A3.

I also have a slight problem with the What will be the peak hours for viewing the report question.

You see what we need is the highest number of renderings at the peak hour. What we do need is the actual time you want the reports to be available. For example, if you are using it from “9 AM to 9 PM,” then you keep your service running during this time only. Amount of time doesn’t help us.

A different way to attack the same scenario.

Let us look at the total number of page renderings for your client. That would be the total number of users and how many times they access the reports. The number we get is 636*217 = 138012

Now, I am assuming that your client requires this only on workdays and for 10 hours each day. That would mean average rendering per hour is (138012 / 20 (workdays) / 10 (hours each day) = 690

If the usage is consistent, then this would require SKU A2. However, if say the ratio of peak use and average use is 2:1 then the maximum page rendering becomes 1380, and you have to get SKU A4.

Finally, if you don’t get further information I, would suggest you have a discussion with the client on these terms.

1. During what times do you want the reports running?  A = number of hours running per day
2. Is it required on weekends and holidays? B = number of workdays in a month
3. Rough estimate on the peak hour how many people will be using C
4. Rough estimate on the interaction (which includes everything from slicing, dicing, filtering) the users will be having with a report D

The calculation on the cost would now be this:

Find the value of C*D = E

If E <= 300 then i = 1.008

If E <= 600 then i = 2.008

If E <= 1200 then i= 4.0242

Rough estimate on the prices would be:

A * B * i

Conclusion

I hope this helps you to understand better the various calculations for Power BI Embedded Licensing cost that you need to estimate for customer Quotations.

Suggestions are welcome in the comment box below. Thanks for visiting!

Disclaimer: The Questions and Answers provided on https://www.gigxp.com are for general information purposes only. We make no representations or warranties of any kind, express or implied, about the completeness, accuracy, reliability, suitability or availability with respect to the website or the information, products, services, or related graphics contained on the website for any purpose.