These projects use cutting edge techniques to turn raw data into interactive insights.  We want to build a culture of data-driven decision making to help drive superior outcomes for all types of stakeholders.

Any stakeholders, business owners or clients who require timely business performance metrics.

E-mail us at contact@matrixbi.com.au

Slicers and filters are tools to refine the data shown in charts. Use them to focus on specific dates, teams, or categories.  Generally, slicers can recalculate entire datasets as they interact with measures, whereas filters generally limit the database from pre-composed columns.  For most of the projects we've chosen to showcase here, we use slicers.

A synced slicer is one where the value of the slicer is synced across multiple pages.  This can make it easy to compare data when navigating across multiple pages of a report.  The different pages may offer different visualisations, but you would like to see those with the same inputs.  A great example of this would be in Home Loan Payments Report.  We have synced the slicers across all pages so that you can define a parameter such as Loan Term or Loan Size a single time, then see the output across the various visualisations. Fine that example here:  Home Loan Payments Visualiser 

Power BI is a business analytics tool that enables visualization and sharing of insights. It was selected for its integration with Microsoft tools and powerful visuals.  It allows for the building of data pipelines, loading & transformation of data & the visualisation of data.  There is also lots of connectivity for downstream distribution of BI content for those with licences.

Not all visual have drill downs.  The most common form of drill-down comes from time aspects of data as BI natively imposes a hierarchy (think years, months, days).  However, when creating data models, we can create additional hierarchies that allow for the drill-down functionality.  Within most businesses this will relate to an Organisational Hierarchy (think Corporate Client, Adviser, Client).

Yes. Click the '...' in the top-right of the visual and select 'Export data'.  Within a Purchased BI licence, additional reports can also be configured from the existing data models

A report page shows a full view of the dashboard; bookmarks store specific filter and visual states for easy navigation.  Bookmarks can also be native to a report where an author has allowed users to toggle the view.  Bookmarks can also be shared with others so that teammates and colleagues can create custom detailed reports.

Click 'Reset to default' in the top menu bar of the report.

Yes, use 'Personal bookmarks' to save and return to a custom view.

If you have permissions, use the Share button in Power BI Service or export to PDF.

Yes, the Power BI mobile app is available on iOS and Android for dashboard access on the go.

Power Query a powerful tool based on SQL that is native to both Excel and Power BI.  Think of it as a way to import data in the form of a relational database.  Many connectors exist to import from exterior data siloes.  It can then be used to clean and shape raw data before it is loaded into Excel or the Power BI data model.   The code which runs Power Query can be very low code allowing for non-coders to effectively load and perform basic transformations on data.  That might involve renaming columns, changing data types, cleaning errors, appending data sets or even merging with another query.  For advanced users, accessing the Advanced Editor code allows for powerful manipulations of data to achieve most any outcome one could imagine.

Power Query steps remove errors, fill blanks, and standardize fields during data import.

Yes. From the PowerQuery Editor, the code which loads and transforms the data is listed in steps on the far right of the Desktop.  Clicking in at any point returns a view of the dataset for where it would be if the query ended there.  As a result, users can step through the code and de-bug it line by line.  This is powerful functionality as it allows the creator to see a visual display of the dataset and error detection is made that much easier.

Splitting columns is just one of many functions that are accessible through the low code Power Query Editor.

A relational database is the standard format for all tables which can be found in PowerQuery, Excel and PowerBI.  They have a few common features.  Each column refers to a different field value, think Account Number or Date.  Each Row value is of the same datatype as every other row in that column.  The datatype is standardised by column.  This set-up allows for very easy manipulations of data whether that is appending datasets or merging them with other relational databases.  A great example of what is not a relational database is actually an income statement for a standardised financial report.  In these reports, each column is often a different month and the first column would have a variety of income and expenses.    A true relational database would have all Dates listed in one column and each line item in its own column.

Data models are used within Excel and Power BI to bring together disparate sources of data that are often relatable by some common field.  An Account number is normally a great example of this common field or Key.  So in a hypothetical data model, we could have trading data with Account numbers and we could have portfolio values with an attached account number.  Those data sources might exist in different locations.  By importing both and uniting the data sources with a common field, we can create a data model.  The takeaways and visualisations from a data model are often far more insightful than those from any single dataset.

A star schema refers to the geometric shape that a data model which unites datasets via a central field is often called.  In our example above, we might have Account number as the central dataset that is common to all fields.  That Account field is then mapped to the Account field of different datasets which surround the central table.  It looks as much like a bicycle hub and spoke as it does like a Star - but it is called Star Schema.

DAX measures perform calculations like sums, averages, or more complex logic on the fly.  

Yes.  Calculated column are the simplest to understand.  They are calculated once at the time of data load and their value would likely be related to some other existing columns in the dataset or derived from a related data set.  Measures are calculated on the fly and as a result can be used to return values for a far wider combination of parameters than would be prudent to load into a datatable.  Measures are also responsive to slicer changes, while calculated columns are fixed once created.

Filters can be locked in the canvas on which reports are created.  Or they can be left open for manipulation at the user level.

Yes, if you have edit rights in Power BI Desktop. Use the DAX formula bar.

Named ranges make formulas easily readable for humans since they can be given intuitive names.  That name can then be applied in all instances where it is required much like an absolute reference.  The use of an intuitive name dramatically simplifies formula auditing.

Tables have many benefits.  First they come in the form of a relational database which means they are optimized for downstream data manipulations.  Within organisations, tables also offer the benefit of tabular formulas which means writing a formula much more like a sentence that can be written, rather than relative referencing which has no obvious context.  Tables are also maintenance free - they exist as a known entity with column names and a table size.  This makes the practice of including a 'buffer zone' to accomodate all possible increases in data size obsolete.  Referencing a table will always refer to every row in that table.  Finally, tables are easily used in Microsoft 365 apps.  That means that storing your data in tables allows for them to be used in the many other Microsoft Apps outside of Excel, which if you are using an Enterprise version of Microsoft.

Data pipelines can be constructed with the assistance of PowerQuery.  This allows for polling of data on regular basis.  Datasets that are static, such as the tables in our Salary Sacrifice report don't really require the regular update of data.  However, markets data might require a once daily or update many time a day of data.  Our Precious Metals Analytics report is a great example of a simple data pipeline.  Here, we use GoogleFinance to regularly update asset values and write a nightly record of asset values into a table.  We then use a Google Sheets Connector within PowerQuery to import that data.  Some additional calculated columns are created at the import stage.  Finally, the relational database is delivered into PowerBI.  BI is set to update nightly so that the data model is always current.

Most datapipelines in these reports are built with Power Query which is highly configurable.  Pro Users can schedule data refreshes up to 48 times daily.