Power BI Embedded, SKU Differences and Cost Breakdowns

Updated: 2021/04/20 – Updated the EM SKU’s to clarify what items you can consume

There still seems to be a bit of confusion about Power BI licensing since the changes made back in May 2017. I get a number of questions about embedding options, were a client wants to share reports internally via SharePoint and/or externally. Now if a customer is a large enterprise and wants to share internally via SharePoint, the best option is Power BI Premium. In Premium you can allocate Power BI work spaces to capacity, so those who log into Power BI with a ‘Free’ license or use SharePoint can see the reports they need. However for smaller organisations paying £3,766 per month is not going to be viable. There are some other options, which is were the confusion creeps in. I’ll be using a very raw measure of Monthly Cost to Power BI Pro Licenses, to show how to get the best bang for your buck/quid/currency denomination of your choice.

There are two types of embedded!

There are two different SKU’s for Power BI Embedded, EM and A. Here’s the break down:

Note: The old version pay-per-click of Power BI Embedded that is going to be deprecated in July 2018, was only licensed for use for external customers, you could not use it to create an internal reporting portal that would replace the Power BI Service. If you did, you would still require a Power BI License for each of the users. The new Embedded service is capacity based, so you are free to use it for internal and external customer, without a 1-2-1 licence for users.

‘A’ SKU’s

These SKU’s are designed for creating your own portal, so you can basically create your own version of the Power BI Service. So you need to create a portal with it’s own sign up/login page, user management etc. and users don’t need a Power BI License to view reports. You’ll need one for creating reports, but for report consumers, nothing is needed. These dedicated capacity SKU’s can be paused so you can reduce the daily/monthly cost as they are billed by the hour (More on this later) and these SKU’s are purchased via the Azure Portal.

‘EM’ SKU’s

Like ‘A’ SKU’s can be used for hosting your own reports, but also can be used for embedding reports into SharePoint allowing your Power BI users to see reports. So you have the best of both worlds to some degree, so you can have your internal users consume reports via SharePoint/Teams, and have a portal for external users to your organisation looking at reports that you produce. However these SKU’s cannot be paused and are purchased via the Office Portal. You are also restricted by what you can consume, so Apps based on a workspace can’t be shared to free users, but Free users do get access to the workspace in PBI if they have a Viewer role:

User without a Pro or Premium Per User (PPU) license can still access a workspace that’s in Power BI Premium capacity, as long as they have a Viewer role.

Source: MS Docs

So you can embed a report in SharePoint or Teams, and Free users can access it, or they can get access to the workspace in the Power BI Service, but not get shared items or Apps. For that you need Premium P SKU’s.

‘A’ or ‘EM’? Which is better?

Now EM and A SKU’s have the same Node types and specification in the background, and EM SKU’s for a single base node is £470 p/m compared to an A SKU at £549 p/m. So you you get a slightly more feature rich EM for less than the A one? Well sort of, If you don’t need embedding in SharePoint/Teams, and you can pause your report platform during some hours in the day, the ‘A’ type can work out cheaper.

So with a rough calculation, if you can pause for 4 hours per day (12am to 4am), you’ll meet the price of EM SKU’s. Say you only need it for business hours (7am to 7pm), then it will cost you £275 p/m about 40% cheaper than the EM SKU. So for A SKU’s you pay a moderate premium for being able to pause it, but if you can, you can save on the running costs.

Price Breakdown

The below chart breaks down the pricing options (As of Jan 2018) for the cost for each service. I’ve added Power BI Premium (P1-3) just for completeness. Also added is a column ‘Monthly Cost vs Power BI Pro Licenses’ so you can see the equivalent number of Pro licenses you would get for your money per month. This shows the cut off points for moving to a capacity service or just buying Power BI Pro licenses.

Capacity Node
Virtual Cores
Memory
Front end Cores
Back end Cores
Peak Renders per Hour
Cost per Month
Monthly Cost vs Power BI Pro Licenses
EM1 1 3GB 0.5 0.5 1-300 £470 63
EM2 2 5GB 1 1 301-600 £935 125
EM3 4 10GB 2 2 601-1,200 £1,881 251
A1 1 3GB 0.5 0.5 1-300 £549 73
A2 2 5GB 1 1 301-600 £1,095 146
A3 4 10GB 2 2 601-1,200 £2,195 293
A4 8 25GB 4 4 1,201-2,400 £4,395 586
A5 16 50GB 8 8 2,401-4,800 £8,795 1,173
A6 32 100GB 16 16 4,801-9,600 £17,594 2,346
P1 8 25GB 4 4 1,201-2,400 £3,766 502
P2 16 50GB 8 8 2,401-4,800 £7,536 1,005
P3 32 100GB 16 16 4,801-9600 £15,077 2,010

Note: EM1, EM2, A1 and A2 are not dedicated hardware, they are still shared infrastructure, anything above this is your dedicated bit of Power BI Service. I’m not sure how pausing works for the ‘A’ SKU’s if they are not dedicated??

Front/Back end cores, indicate how much is applied to the web interface front end, and doing stuff like loading the report, interacting with it is the back end.

If you want, you can mix and match nodes. You can have three P1’s rather than a P3. I have been doing some consulting for a large global company, that wanted 20 Power BI Premium nodes, and they split up the nodes into various P types, regions and departmental areas, to balance loads and make it a bit more manageable in report deployment and compliance.

Don’t forget, you can build you own Power BI Embedded portal or even better get Ricoh BSG to build you one from our base version, that cost will have to be factored into the cost of running the platform, and well as any hosting costs, and other incidentals. For report creators you’ll still need a Power BI Pro license, but for consumers, no charge!

Looking at the pricing options seems to be a price cut off point for the size of your organisation:

  • Less than 65 users, then purchase Power BI Pro licenses
  • More than 65 users but less than 500, then purchase Power BI Premium EM or A SKU’s
  • Over 500 users Power BI Premium is the way to go, if you DON’T need a portal for external clients
  • It will depend on your reporting requirements, and he number of users you expect to interact with them!

Sharing to people external to your organisation is a bit clunky in Power BI. At the moment, you can share dashboard and report to external users, but it is URL based. You can now using Azure Active Directory B2B, allocate external users a Power BI License, so they can log in. I like the Power BI Embedded portal concept as it allows you to control the experience a lot more, make it better looking than the Power BI web interface and allows you to add other visualisation types, such as SSRS, D3.js and any other ones you like into your own portal.

When the change to Power BI Embedded moved from a pay-per-click to a subscription model, the community boards made for some, lets say, heated discussion on the changes. Most didn’t like it and saw it was too expensive. However the old pay per click had one main issue which was raised by a number of potential customers I had encountered, who noted that ‘it was not a ring fenced cost’, so technically it was possible that someone could keep pressing F5 and generate a £0.03 charge every time they did it, and the client would be presented with an unexpected bill at the end of the month. I also thought that the old pay per click workspace model, would have been hard to scale and had the feeling it wasn’t fully fleshed out for production use. Now it’s on the Power BI REST API, it’s been a lot easier to scale up or out, create and importantly maintain all the parts of it.

Any questions? Need a Portal? I’ll reply in the comments.

3 thoughts on “Power BI Embedded, SKU Differences and Cost Breakdowns

  1. Nice post. Thank you for updating at well. I have question about EM versus P capacity. We have purchased 3 P1 SKU with long-term commitment. However, our capacity admin is assigned EM capacity to some projects – EM1 to EM3 depending on the requirements of the project. I have queried this approach because Microsoft documentation and your post suggests that EM SKU are separate from P SKU. When I suggested this to the capacity admin team, they responded by explaining that EM Capacity setting was a subset of the P SKU. It was explained like Physical and Virtual Machine. The P SKU / Capacity was like the Physical machine while using the EM settings for different projects was akin to giving each project a limited share of the P capacity. Also if the project need more resources it would scale to use more of the P1 capacity. Is this correct – the EM capacity and SKU are subset of P SKU?

    Like

    • No they are only a subset of features, not the physical machine itself. P and EM SKU’s are sperate and have to be purchased separately in the O365 portal. It you want to move an EM to a P, you would have to purchase a P node (Or assign one you already have) and assign it to the workspace covered, then remove the EM (or reassign it to some thing else). EM SKU’s do not use a bit of the P in anyway, you can’t separate P into EM SKU’s, The lower levels of the EM SKU’s EM1 and EM2 are shared capacity, not a full physical node like P1 or EM3 and above, may that is what they are thinking of. But you are sharing that capacity (compute node etc) with another Power BI customer. Power BI capacity v2 now has auto load balancing features, this means that for example a EM1/P1 can scale up to the demand, but it doesn’t scale up an EM to a P. Hope that helps

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s