Callum Green's Blog

Themes in Power BI – Part 2

In September’s Power BI Desktop release, one of Microsoft’s announcements was a new offering of Themes.  My previous blog back in March 2017 explained what a Theme could (or couldn’t!) do, but I am pleased to say Microsoft listened to the community and have improved it greatly.

Originally, we were limited to setting a few page or chart colours - mainly tables and matrices.  Appreciated this was a first release from Microsoft, but it’s safe to say I was left uninspired.  The good news is Themes can now do a lot more, including:

-          Set the font size for specific charts e.g. 14

-          Set visuals to use a specific font type style e.g. Calibri

-          Turn off/on chart data labels

-          Set default property values for a chart

-          “*” functionality that will apply a specific property across all charts

This blog will focus on how to not only use the new features, but how they can be brought together to form a larger, organizational configuration file.

If you would like to follow my examples, please ensure you have downloaded the latest version of Power BI Desktop – found here.


In order to extend the Adatis Theme, a new section called visualStyles must be added to the JSON.  The format is as follows:


The highlighted JSON sections give you the ability to use ‘*’ syntax, which means the Theme format will apply to everything underneath it.  This is useful for things if you want to apply a global font size across the entire report, rather than needing to configure every chart/visual.

Now, let’s pick a simple use case.  Client A requires the following to always be applied to a Pie Chart in Power BI Desktop:

-          Switch on the Pie Chart Legend

-          Set Text Size to 14

-          Set Font Family to Calibri Light.

Default Pie Chart:


-          By design, Legends are turned off.

The configuration file is tweaked to contain the new JSON - highlighted in yellow.


  "name": "Adatis",

  "dataColors": [ "#002C5C", "#006FBA", "#81C341", "#F4AA00", "#003876", "#00448F", "#0051A9", "#007ED4", "#008DED", "#74B238", "#8EC954", "#DB9800", "#FFB60F" ],

  "background": "#FFFFFF",

  "foreground": "#002C5C",

  "tableAccent": "#81C341",

  "visualStyles": {

    "pieChart": {

      "*": {

        "legend": [


            "show": true,

            "fontSize": 14,

            "fontFamily": "Calibri Light"







The final step is to import the modified JSON file, resulting in a completely different looking Pie Chart. 

Applied Theme:


1.       The traditional Adatis chart colours have been applied.

2.       A Legend is visible – in both the properties pane and the visual itself.

3.       Font family and Text Size are updated.

Adatis Theme Extended

I will now add some additional configurations to the Adatis Themes file.

-          Turn On all Chart Legends.

-          Pie Chart Label Style set to Percent of Total.

-          Set Card Label and Category Label Font Size to 12, but all other visuals use 10.

This changes the JSON code significantly, but by following the visualStyles format, this was more straightforward than it seems.


  "name": "Adatis",

  "dataColors": [ "#002C5C", "#006FBA", "#81C341", "#F4AA00", "#003876", "#00448F", "#0051A9", "#007ED4", "#008DED", "#74B238", "#8EC954", "#DB9800", "#FFB60F" ],

  "background": "#FFFFFF",

  "foreground": "#002C5C",

  "tableAccent": "#81C341",

  "visualStyles": {

    "*": {

      "*": {

        "*": [


            "fontSize": 10,

            "fontFamily":  "Calibri"








    "card": {

      "*": {

        "labels": [


            "fontSize": 12



        "categoryLabels": [


            "fontSize": 12





        "pieChart": {

          "*": {

            "labels": [


                "labelStyle": "Percent of total"







The report is significantly different.  I will pinpoint a specific area of the report to illustrate where the configurations occurred.





1.       Global font size is set to 10.

2.       Percent of Total setting applied to Pie Charts.

3.       Category Labels and Labels set to font size 12 for all Cards.

What else can Themes do?

This article is only scratching the surface at what Power BI Themes can now do.  Click here to find the extensive list of properties that can be configured.  It is worth mentioning that if you have limited JSON experience, some of the configurables can be a little tricky to set up.  However, there are quite a few blogs out there that can get anyone up to speed, which I will further provide below. 


There are a couple of limitations, as well some things I have experienced:

-          Debugging JSON (natively within Power BI) is not supported and a syntax error occurs. Power BI will give you a generic error message like below:


-          Data configurations are not supported

·         This is understandable, though.  Every report tab contains different slices or filtering of data, but colours, fonts and chart properties tend to remain the same.

-          Setting Card font and background colours does not appear to be possible yet.  Or at least I have been unable to get it to work or find any supporting documentation from Microsoft to suggest it is.

-          Overriding a Theme

·         Individually changing a chart property will override the Theme JSON file.  I was caught out by this and couldn’t understand why the configuration wouldn’t change on a Pie Chart.

·         To get around this, use Revert to default within a chart property and it will re-apply the Theme.  This option exists in each top-level chart property – see example below.



Further Reading

o   Jason Lengstorf’s JSON Blog –

o   Power BI Desktop September Update -

o   Power BI Report Themes BI Blog -

o   Themes Gallery (containing example JSON files) -

Contact Me

If you would like a copy of the PBIX workbook/JSON Theme file or have any general questions, feel free to leave a comment below.  I am particularly interested to see what Theme files other people have created and if there any other known limitations.   

Twitter:                @DataVizWhizz

Direct Query vs Live Connection in Power BI – Part 2

This instalment of the Power BI blog series focuses on the two other differences between Direct Query and Live Connection in Power BI. 

If you haven’t done so already, be sure to check out Part 1, which concentrated on Quick Measures and Relationships.

Feature Differences

Find below a Power BI Direct Query screen, focusing on the final 2 differences (highlighted green):

o   New Hierarchy

o   Change to Import Mode


New Hierarchy

Hierarchies are very useful with Power BI, especially when wanting to drill up and down within levels of data e.g. Day > Month > Year. Let’s try and create a New Hierarchy in Live Connection mode:


New Hierarchy does not appear in the list of options.

This is because hierarchies are typically created within the OLAP Cube, therefore, it makes sense why the option is not available.  However, I would argue that Direct Query allows you create a hierarchy, so why can’t Live Connection?

Direct Query stores the hierarchies within a smaller Power BI model (Tabular Cube) running on a report developers local machine.  Ironically, Live Connection does exactly the same thing when you create an ad hoc measure in Power BI. It is surely a matter of time before hierarchies are also supported.

Change to Import Mode

Direct Query mode supports the ability to easily switch to Import Mode.  This is a useful option, especially for a self-serve analyst wanting to make transformations and shape the data.  By right clicking the highlighted option below, a simple wizard appears:



When the import completes, the data (from tables in the database) will be stored in a Tabular cube on the local machine where Power BI Desktop is running.  If we navigate to the same area within Live Connection, notice there is no option to change to Import Mode (“click to change”:


Once again, I am not really sure why this feature isn’t support in Live Connection mode.  I can understand why it would be more difficult to convert a Multidimensional Cube (MDX) into a local Tabular Cube (DAX), but if Live Connection already points to Tabular, it’s an exact copy. 

To vote for Import Mode functionality within Power BI Live connection, click here.

Coming Soon

Part 3 is the final instalment of the blog series, specifically focusing on the underlying Power BI Data Models in Direct Query and Live Connection.

Further Reading

Other than the Power BI Blog, there are some other great pages out there too:

o   Power BI Blog -

o   Import Mode vs. Direct Query -

o   Direct Query in Power BI -

o   Live Connection in Power BI –  

Contact Me

If you have any questions or thoughts, please leave a comment below.  My Twitter details are also provided.

Twitter:                @DataVizWhizz

Direct Query vs. Live Connection in Power BI – Part 1

There are lots articles out there that compare Import Mode vs.Direct Query, but people rarely talk about if there are any differences between Direct Query and Live Connection.  “Wait. Aren’t they the same thing?”  Well, not quite.

The first big difference between Direct Query and Live Connection is the type of data source used for the connection.  The former uses a database connection (typically SQL Server), whilst the latter requires an Analysis Services OLAP Cube.

This blog series won’t explain what Direct Query and Live Connection can do (found here and here), but will instead highlight the other subtle differences between the two connections. 

Feature Differences

There aren’t any features I can find that are available in Live Connection and not Direct Query.  However, there are a few the other way around.

I will first show you a Power BI Direct Query screen, focusing on the 2 of the 4 differences (highlighted green):

o   Quick Measures

o   Relationships


Quick Measures

This feature was first released in April 2017 and is available in Import Mode and Direct Query.  It enables a non-technical Business Analyst to create relatively complex DAX aggregations, with the use of a nice Wizard.  To access Quick Measures, right click on a measure or attribute and select Quick Measures.  Let’s try the same thing in Live Connection mode.


You’ll notice that Quick Measures is missing from the list of options.

I find it bizarre that Live Connection doesn’t support Quick Measures, especially when using a Tabular Cube as the connection.  The Power BI DAX language and engine are the same as Tabular, so you would think the two are compatible!

Please vote for this feature to be added into Live Connection -


There are two tabs displayed on the left-hand pane in Direct Query mode.


If you click the highlighted tab, it opens a Relationships page – where you can begin to join datasets (from the database) together.  I created a manual relationship that joined DimEmployee and DimDate together – as shown below.  No relationships are created in the underlying SQL Server database, but instead stored within the Power BI model.


In Live connection, the left-hand pane looks bare:


There is no option to create any form of Relationship against the Live Connection Tabular Cube.  This kind of makes sense because a BI Developer would be the person responsible for creating relationships within an OLAP Cube.  I would argue that if you want the ability to mashup data or create your own relationships, you shouldn’t be connecting to a Cube in the first place.

Coming Soon

Check out Part 2 of my blog series - available here.  The focus of this article are the Add Hierarchy and Change to Import Mode features.

Part 3 will conclude the Trilogy, where I go off-piste slightly and focus on the Data Models in Direct Query and Live Connection.

Further Reading

Other than the Power BI Blog, there are some other great pages out there too:

o  Power BI Blog -

o   Import Mode vs. Direct Query -

o   Direct Query in Power BI -

o   Live Connection in Power BI –  

Contact Me

If you would have any questions or thoughts, please leave a comment below.  My Twitter details are also provided.

Twitter:                @DataVizWhizz

Slicer Properties in Power BI: Header or Title?

I recently shared a Power BI Report with a customer and they reported that the “Clear Selections” option (Eraser icon) was not available when they used the Slicer.  It took me a while to work out why this was.

This blog will illustrate how you can lose the “Clear Selections” functionality, depending on what Format settings are applied to a Slicer.  I will also show how to work around the formatting constraints, which will help prevent you from ever experiencing the issue.

In order to follow the examples, you will need access to Adventure Works 2014 SSAS Tabular and of course, Power BI Desktop.  Follow the appropriate links to download what you need.

Use Case

I have created a very simple report (available on request), using a Product Category Slicer and a Map to display Internet Total Units. 


o   Format property.

o   Select All is off, ensuring multi select is allowed.

o   Header is off.

o   Title is on, used instead of Header and configured to look like the below:


After multi-selecting the attributes in the Slicer, I tried to “Clear Selections” – which is normally available as an option like below:


However, it does not appear in the report I created:


The reason I cannot see the “Clear Selections” option is because I am not using a Header.  When I use this instead of Title, the Slicer contains the required feature.

Header vs Title

This made me wonder what other differences are there between Title and Header but in fact, there aren’t many.  The subtle differences to be aware are displayed and described below:


o   Header can have an Outline, which includes the ability to underline text.

o   Header is constrained to displaying the name of the attribute (“Product Category Name”), whereas a Title can be customised (“Select Category”).  You can rename your source data attribute to get around this, however.

o   Title enables you align the text, but this is not possible with a Header.

o   Header contains the “Clear Selection” option.


There are couple of ways to work around the missing “Clear Selections” issue, which I will demonstrate below.

Option 1

I could simply switch from a Title to a Header, but then we would lose the ability to centre align the description.  Instead, we can set both options to ‘On’.



After some formatting, the Slicer is pictured below.


o   The Header text is set to White, therefore, it not visible. 

o   You can still see the “Clear Selections” option.

o   However, there is white space in between the Title and Header.  This not only wasted space, but also looks a bit strange from a visual perspective.

o   The actual sections themselves (e.g Bike) are a bit squashed and disproportionate to the Slicer border.


Option 2

The other workaround involves a little more work, but gives the impression that only a Title is being used, but with the added functionality of the “Clear Selections” option.  Furthermore, there is no longer the white, empty space.




o   Two objects were used to produce the result:

·         Text Box

·         Slicer

o   The text box has been formatted and labelled with ‘Select Category’, using the Title property.  This allows you to change the font colour.

o   The Slicer is using the Header option, ensuring “Clear Selections” is available. Title is turned off to reduce the empty space.



So why are there two types of properties for Slicers and other objects like Text Boxes?  I am not entirely sure myself, but occasionally, you may want a Slicer to contain a Title with an Attribute Name as the Header underneath. This gives you a Title > Sub-Title concept (illustrated below).  I do understand why the “Clear Selections” feature is specific to the Header setting, as it directly relates to the Slicer attribute.  

There are certainly ways of solving this issue - here are just a couple of suggestions:

1.       Providing continuity across both the Header and Title format settings.  Quick fix, but not necessarily solving the ambiguity around both options.

2.       The Header setting is contained within the Title, meaning both sets of functionality are merged into one.  It would make the usability of a Slicer (especially from a development perspective) a lot better.

Whatever Microsoft decide to do in the future, I really hope they tidy up and fully define the Settings within Visuals, Slicer and Text Boxes.  Some things are confusing, especially to self-serve Business Analysts who rely on intuitive reporting tools.

Further Reading

Power BI Documentation –
Power BI Community Blog -

Contact Me

If you would like a copy of the workbook containing the examples or want to know more about any of the Power BI settings/properties not mentioned, please leave a comment below.  My Twitter details are also provided below.

Twitter:                                @DataVizWhizz

Themes in Power BI

Microsoft recently released Themes to a Preview version of Power BI.  The concept is pretty simple – imagine having standardized colours that can be applied to charts and matrix visuals? This is where Themes come in. The theory is this will ensure Power BI developers adhere to company policies and can do so quickly and efficiently. The reality is Themes are still lacking a lot of key configurables, but I will get onto that later.

Whilst this blog is not going to show you how to import or apply a Theme, it will demonstrate how to create one and what each configuration means. For a great high level walkthrough of the feature, click here. The aim is to not only show how this makes a developers life easier, but where it can actually be improved too.

JSON Configuration File

I have pasted the following code into Notepad++, which incorporates the Adatis branding and colour scheme.  For many developers, you will recognize the coding language – JSON (JavaScript Object Notation). If you would like to know more about JSON, check out Jason Lengstorf’s blog.

Use the below code as a template for your Theme:


  "name": "Adatis",

  "dataColors": [ "#002C5C", "#006FBA", "#81C341", "#F4AA00", "#003876", "#00448F", "#0051A9", "#007ED4","#008DED","#74B238","#8EC954","#DB9800","#FFB60F" ],

  "background": "#FFFFFF",

  "foreground": "#002C5C",

  "tableAccent": "#81C341"


Notice there are a number of ‘#’ within the code.  These are Hex numbers, in which Power BI natively uses to determine a shade of colour.  This is common across many reporting/image editing tools. 

Configurations Explained

I will now explain what each line of code means and how it will effect a simple un-themed report.



1.       The name of the Theme.

2.       The colours that appear in a pre-set chart colour palette and the order of which a visual applies them e.g. Dark Blue first, Light Blue second, Green third, etc.

3.       Font colour for the matrix or table visual.

4.       Primary background colour for a matrix or table visual.  Even though the setting says’ ‘foreground’, it is more of a background colour!

5.       The table accent applies to a matrix or table visual, displaying as the grid outline or contrasting background colour, depending on what type matrix/table is applied.


Applying a Theme

I have created a quick Power BI report, which is using the default colour schemes applied.  The only formatting applied was to the Matrix visual and this was simply changing the style of the grid.  Here is how the report currently looks:


Now let’s import and apply the ‘Adatis’ theme.  You will see some of the visuals have changed, whereas other elements of the report haven’t.  I will explain in greater detail.


1.       The Report Title, Slicers and Titles are not affected by the Adatis theme.  This is because you cannot currently configure fonts in the JSON file.

2.       The matrix visual has updated, using the three configurations form the JSON file:

a.       background (White)

b.      foreground (Navy Blue)

c.       tableAccent (Green)

3.       The Pie and Bar chart have picked up the colours in the order they are specified in the JSON file, using the dataColors property.  However, note the colour of both data points on the Bar Chart – rather than use different colours, it uses the same primary blue colour.  Not ideal if you want to plot one colour for ‘Male’ and another for ‘Female’.

4.       The Waterfall chart hasn’t changed at all.  We would expect the Adatis colours to have been applied, but this visual seems to ignore the configuration.

We can change the Waterfall Chart manually, using the imported Adatis Colour Palette. 


Here is how the chart now looks:


What’s Next?

Whilst the concept of Themes is great, the current preview functionality is extremely limited and very much ‘Version 1’.  I am a little disappointed with how inconsistent the application of a theme is, which were highlighted in my points above.

With the majority of new Power BI features, Microsoft continue to improve the usability and functionality over time, so I am very hopeful more Theme properties will be opened up in the underlying JSON configuration file.  Options such as font colours/styles, consistent theme application (for all visuals) and company logos are all necessary for this to become really powerful.  There were talks of a CSS type of configuration in Power BI, but this has yet to announced or released.  Imagine how powerful and cool that would be?

As a BI Consultant, I am not currently comfortable with demoing Themes to a client, simply because of the clear gaps.  Once the feature is more mature and in GA, I think enterprise companies will really benefit from standardising reports across their business.  For now, Themes will remain a glorified colour palette.

Further Reading

o   Jason Lengstorf’s JSON Blog –

o   Power BI Report Themes BI Blog -

Contact Me

If you would like a copy of the workbook or have any questions about this blog, please leave a comment below.

Twitter:                                @DataVizWhizz

UK Power BI Summit 2017: What’s next for Power BI?

During the keynote at the UK Power BI Summit, Microsoft announced some new features coming into Power BI. Chris Webb also provided a session on Azure Analysis Services, hinting at what may be next and more importantly, how they may impact Power BI. 

Without further ado, I will dive straight in.

Power BI Visuals

Some exciting visuals were demoed by Will Thompson (Program Manager for Power BI). They were shown in an exclusive preview of Power BI, not available to the general public:

-          Themes

o   Finally! Companies can now easily ‘skin’ a suite of reports to use standardised logos, fonts, etc. 

o   The themes will be driven by JSON file.

o   All themes can be uploaded into, using the ‘Themes Gallery’.

-          Slicer Improvements

o   A numeric, sliding scale slicer will be available as a new configuration option.

o   Currently only designed for attributes. 

o   Measure slicers are not  an immediate focus, but Microsoft are giving it thought.

-          Matrix/Table Visuals

o   Whilst they currently exist, a new visual will be added. This includes:

§  Drill down capabilities.

§  Include/Exclude – like what you see in charts currently.

§  Ability to highlight rows and columns, which also cross filters and highlights the selections on accompanying charts.

Analysis Services

All of the features discussed are for Azure Analysis Services, but I am sure they will also be made available on-prem soon after GA release:

-          Live Connect to Analysis Services

o   Add ad hoc measures within the Power BI Model.

o   It does not update your actual Analysis Services model, therefore, adding measures in Power BI need to be done so with caution.

o   Currently Tabular only, but should eventually come into Multidimensional too.

o   Available in both Azure (cloud) and on-prem.

-          Power Query/M Integration

o   This is the missing jigsaw to the Power BI/Analysis Services puzzle.

-          Table Level Security

o   Ability to implement this and row-level security together.

-          Migrating existing Power BI Models

o   This will be awesome! Imagine being able to move your complex Power BI model (which is now performing badly) into an Azure Analysis Services model?

o   At the moment, you have to create the tabular model again.


Whilst no official dates have been given, I hope that March’s release will include the visual features, such as themes. The ones specific to Azure Analysis Services are ideas that have come from many blogs (professional experts) and the Microsoft roadmap. 

If there is a feature that your customer needs or just something you believe should be in Power BI, the Ideas forum is a great place to submit your suggestion.

Further Reading

For up to date Microsoft announcements, I would definitely advise following the below blogs and twitter users:

-          Blogs

o   Power BI Ideas -

o   Power BI Blog -

o   Power BI Blog Announcements -

o   Chris Webb’s blog -

-          Twitter

o   Will Thompson (Power BI Program Manager) - @Will_MI77

o   Riccardo Muti (SSRS Program Manager) @RiccardoMuti

Contact Me

If you have any questions, please comment below or contact me directly on Twitter (@DataVizWhizz ) .

Dual KPI Custom Visual in Power BI

On February 8th, Power BI released a new custom visual called Dual KPI. The purpose of this chart is to visualise two measures over time and show their trend based on a joint timeline. The absolute values may use different scales e.g. Sales and Profit.

This blog will not only show you how to set up the new visual, but also demonstrate how changing some of the settings can enhance a report. Adam Saxton posted a YouTube video that also walks through the Dual KPI.

Pre Requisites

In order to follow my example, you will need a copy of AdventureWorksDW2014 database – found here. You will also need to download the following custom visuals:

o   Hierarchy Slicer –

o   Dual KPI –

NOTE:   This article assumes previous knowledge of downloading and importing Custom Visuals into Power BI Desktop. If this concept is new to you, Scott Murray’s blog gives great step by step instructions. 

Prepare Data

Open Power BI Desktop and Get Data. Point to the new AdventureWorksDW2014 database and drop down Advanced Options. Paste in the following T-SQL:





       ,SUM([TotalProductCost]) AS [TotalProductCost]

       ,SUM([SalesAmount]) AS [SalesAmount]

       ,SUM([SalesAmount]) - SUM([TotalProductCost]) As ProfitAmount


FROM [AdventureWorksDW2014].[dbo].[FactInternetSales] FI


       [dbo].[DimProduct] DP

       ON DP.ProductKey = FI.ProductKey


       [dbo].[DimProductSubcategory] DPS

       ON DPS.ProductSubcategoryKey = DP.ProductSubcategoryKey


       [dbo].[DimProductcategory] DPC

       ON DPS.ProductcategoryKey = DPC.ProductcategoryKey

WHERE ShipDate BETWEEN '2013-01-01' AND '2013-06-30'






When happy, click ‘OK’ to continue. The preview of the data will open.  Click Load, as we do not need to edit any data in the Query Editor.  Apply and changes and rename the query to ‘Internet Sales’ – final output below:


Some measures and attributes need to be formatted within the ‘Modeling’ Tab.

o   ‘ShipDate’ = dd MMMM yyyy

o ProfitAmout’ = Currency

o   ‘SalesAmount’ = Currency

The final formatting step is to create a Product hierarchy, based on the three product attributes.  Navigate to the Data tab, right click on the ‘EnglishProductCategoryName’ attribute and select ‘New Hierarchy’.  Drag the attributes into the hierarchy and name it ‘Products’.  It should look like the following:


Create Report Visual

We need to use both the Slicer and Dual KPI custom visual. To achieve this, follow the steps below:

Select the Hierarchy Slicer in the Visualizations menu and drag the ‘Products’ hierarchy on to the Fields box. The slicer will now appear in the report.


Select the Dual KPI Slicer in the Visualizations menu and drag the following measures to the appropriate chart properties box:

a.       ‘ShipDate’ > Axis

b.      ‘SalesAmount’ > Top values

c.       ‘ProfitAmount’ > Bottom values


The chart is now configured and each metric/visual is explained in more detail below. Only the top KPI (Sales Amount) is shown because both use the same calculations.



1.       This is a fixed growth percentage, comparing the last (06/30/2013) vs. first (01/01/2013) data point on the graph. The metric acts as a static KPI.

2.       The Sales Amount value for the last data point on the graph. Also a static KPI.

3.       The data point currently being hovered over. This dynamically changes when you move along the axes.

4.       The Sales Amount value for the current data point being hovered over. Also dynamic.

5.       % since metric that looks at the Sales Amount for the last data point on the graph and works out the growth based on the current data point being hovered over. To use the example in the screenshot:

-          Sales Amount for 06/30/2013 = 51,596

-          Sales Amount for 05/17/2013 = 18,442

-          % since:  ((51,596 - 18,442) / 18,442) * 100 = 179.7%

Enhancing the Report

As with all custom visual in Power BI, there are lots of settings that you may never use. I have picked out some that enrich the capabilities of the Dual KPI Chart:

o   Fields

o   Warning State

§  Set alerts around data freshness and view warning messages.

o   Top/Bottom % change start date

§  For the fixed +/- % change on the chart, you can add an override start date. The dates could vary by product category and dynamically impact the % in the visual.

o   Format

o   Dual KPI Properties

§  Show abbreviated values, define multiple tooltips and show stale data warnings.

o   Dual KPI Chart Type

§  Choice of either Area or Line charts.

I have applied the Top/Bottom % change start date functionality and also formatted the chart properties. The report now looks a little more professional:


Further Reading

o   Adam Saxton YouTube Video –

o   Power BI Blog -

Contact Me

If you would like a copy of the workbook or have any questions about this blog, please leave a comment below.

Twitter:  @DataVizWhizz

On-Premise Power BI: Part 2 – What’s next?

Part 2 of this blog series focuses on the future of Power BI On-Premise and what direction Microsoft are heading in.  This is a little tricky, as there haven’t been many formal announcements of how both SSRS and Power BI will work together from an architectural and pricing perspective.

I will be using this article as a forum to discuss both On-Premise and Cloud variations of Power BI, potential licensing and whether the new offering is actually just ‘Plugging Power BI reports into SSRS’.  The last statement is a little crude but from a business perspective, decision makers need to know what version of Power BI suits their needs.  We won’t get all of the answers right now, but it would be nice to shed some light on what appears to be a very dark room.

Cloud vs On-Premise

What does this mean for Power BI as a brand and more importantly, as a reporting tool?  The original purpose for Power BI was to offer self-service, ad hoc end user reporting.  However, as the product has matured, prospective clients have wanted it to do so much more.  “How can we incorporate Active Directory security”? “How do we share Dashboards within a specific workspace and limit permissions to it”? “But our business uses ultra-sensitive data and it MUST stay on a server in the UK…..”.  These are the types of comments/questions that get banded around a lot at the moment.

This is why Microsoft are offering both Cloud and On-Premise.  We are not yet in a position where definitive pros and cons can be laid out for each option, but this will soon be possible when Microsoft reveal their ultimate strategy.  There are still a lot on unanswered questions, especially around licensing.  A sensible assumption is that Power BI On-Premise will be covered under the typical Enterprise Edition version of SQL Server.  However, will there be an add-on fee for Power BI or will the general license costs go up?  What impact will this have on the Cloud costs?  There are some complicated pricing models for Office 365 users and the simpler £9.99 a month for a standalone Power BI Pro license.  Will the same pricing strategy exist or will On-Premise force Microsoft into a rethink?  I tend to stay away from speculation in my blogs and stick to facts, but I genuinely am interested to see how Microsoft market the variations of Power BI.

Coming Soon

After looking through various blogs and forums (links provided below), the following features/functionality will soon be available before Power BI On-Premise goes to GA.

  • Short-Term
    • Custom visuals
    • Additional data connectors (besides Analysis Services), cached data, and scheduled data refresh
    • Power BI mobile apps (viewing Power BI reports stored in SSRS)
  • Longer-Term
    • R Visuals
    • Support for integrating previous versions of SQL Server Databases (2008 +)  and Analysis Services (2012 SP1 +) with SSRS 2016
    • Support for all data connectors currently enabled for Power BI Cloud

Microsoft have also listed some Power BI cloud features that are not planned for the On-Premise version:

  • Dashboards – The concept of pinning a report and sharing it on an ad hoc basis
  • Q&A (Natural query language)
  • Quick Insights

Another pertinent question is “When will Power BI On-Premise actually be available in the real world?”.  Microsoft are targeting a production ready release for mid-2017, although nothing is official yet.  One thing is certain – it won’t be coming in a Service Pack or Cumulative Update.  Another big thing to consider is migrating from SSRS 2016 to SSRS with Power BI Reports, which Microsoft are promising will be easy.


At this stage, it is difficult to give any concrete information on where Power BI On-Premise is heading.  All we know is that Power BI and SSRS will be working together a lot more closely and the majority of functionality will be available in both. 

The concept of having Power BI reports shared and deployed to physical, on premise servers will accommodate companies worried about moving their data to the cloud.  As Power BI continues to increase in popularity, the overall security and infrastructure model will be scrutinized.  Cloud storage is often falsely labelled as a security risk, which is where the common corporate misconceptions are born.  It will be hard to change this train of thought, which is where the On-Premise Power BI offering will come in handy. Even more appealing is the natural integration with SharePoint and SRRS, enabling companies to use hybrid approaches, as well as not need to migrate old SSRS reports into Power BI.  Everything is managed in one location, thus reducing security risks and costs. 

Now we all sit tight and wait for Microsoft’s next big announcement.  If anyone has more information around Power BI On-Premise, please comment below.  

Further Reading

o   Power BI Reports in SSRS Release Notes -

o   October 2016 Technical Preview Blog –

o   December 2016 Feedback Review Blog –

o   January 2017 Technical Preview Blog –

On-Premise Power BI: Part 1 - Technical Preview

Back in October 2016 at SQL Pass, Seattle, Microsoft announced that Power BI was going on premise.  Along with the big fanfare, there was also a Technical Preview released – in which a user could download a pre-built VM from Azure Marketplace and get a first look of how SSRS and Power BI could work together. 

Roll on a few months (January 17th 2017 – to be precise) and the new Technical Preview is available.  This time, you can actually download a standalone copy of Power BI Desktop (for SSRS) and a slimmed down SQL Server Reporting Services 2016 configuration tool.  If you follow the official Microsoft blog, getting set up is really easy.  Just a heads up – there are still lots of limitations with this preview, which can be found at the bottom of the January blog.

The blog will be broken into two parts – with Part 1 focusing on my experiences with the January Technical Preview of SSRS and Power BI.

Technical Preview Feedback

Setting up a local SSRS Server (on my own machine) and deploying a Power BI Report was really easy, which is good feedback in its own right.  Annoyingly, only Analysis Services connections are currently supported within the preview Power BI Desktop application and all of my workbooks use online services or SQL Server databases!  This meant I had to download a tabular model for Adventure Works and create a quick test report using the dataset.  Not a major problem, but an inconvenience nonetheless.

All of my other observations are based around differences between the regular SSRS service and this new offering.  In essence, not much has changed – which can only help a user’s experience.  Here are the things to be aware of:

SSRS Configuration Tool

There are two new options – ‘Scale-out Deployment’ and ‘Power BI Integration’.  The latter is the one we care about.  In effect, you need to tie SSRS version of Power BI Desktop (which is available through the Technical Preview install) to the SSRS Configuration.


Power BI Desktop

The application looks pretty much the same, with the only differences highlighted below.  The most important bit is saving the report to the SSRS Server.  I thought it would have been under the ‘Publish’ option, but Microsoft opted to put it in ‘Save As’.  The application itself is called ‘Power Bi Desktop (SQL Server Reporting Services)’ and I imagine it will stay this way, in order to differentiate between On-Premise and Cloud versions.


The final step is to define the SSRS Server.  A grid appears and you can either choose an old URL or define a new one.  As long as you are connected to an Analysis Services model and type in the correct server name, then the report will successfully deploy.


SSRS Report Server

I deployed a test Power BI Report to the local SSRS Server and there are a few specific options now.  By clicking the ellipses, you can open the report as normal, Edit directly in Power BI Desktop or even download the pbix file.  Notice the ‘Power BI’ icon on the tile of the report, which also helps with distinguish the type of reports in the SSRS portal.


My first impressions of Power BI On-Premise are good.  Whilst there are still some clear gaps and limitations, the fact that Microsoft are looking at bring SSRS and Power BI together must be a good thing.  They recognize the need to bring Mobile, standard On-Premise and ‘director pleasing’ reports together in a one stop shop.

I am certainly excited about the next Technical Preview for Power BI On-Premise.  Not only should it contain more data connectors and features, Microsoft should reveal more about the long term visions.  Look out for my next blog where I will discuss licensing implications, architecture and some of the common Power BI questions I get from prospective clients.  

Further Reading

o   Power BI Reports in SSRS Release Notes -

o   October 2016 Technical Preview Blog -

o   December 2016 Feedback Review Blog -

o   January 2017 Technical Preview Blog -