Considering a Daily Time Dimension – Pros and Cons

Posted by Michele Morrill

Sep 25, 2015 2:28:17 PM

Time Dimension Image

There are some business cases where a daily time dimension can be helpful, essential, or even required, but many clients shy away from a daily time dimension for a variety of reasons.  Regardless of the chart of accounts used or the planning, budgeting and forecasting required, a daily time dimension can add breadth to the information and promote more timely responsiveness under the right circumstances.

Whether it is a Financial Reporting application or a Reporting and Financial Consolidation application, an analytical database is designed for analysis (and aggregations.)  Creating a model with a daily time dimension doesn’t prohibit the use of an analytical database or an EPM solution.  A daily time dimension will still facilitate analysis in aggregate, regardless of the database selected.  A relational database is usually better suited for transaction processing, whereas a multi-dimensional database is ideal for analysis, budgeting and forecasting.  A daily time dimension can be used to plan and analyze daily data without overwhelming the system or slowing it to a crawl. 

The primary consideration of a daily time dimension is:  do the business requirements support this need.  All other considerations pale if the business isn’t tightly cyclical, requiring a daily time dimension.  Some industries are ideal here – think hospitals, airlines, maybe transportation – with heavy transaction level details and the ability to influence business decisions quickly.  Other industries don’t make sense – such as smaller businesses or manufacturing organizations with slow start-up times. 

After business requirements, consider the volumes of data required to facilitate daily analysis.  A daily time dimension requires daily data.  This can get voluminous quickly, and some IT organizations or data warehouses just can’t provide the data at that level yet.  Daily data requires space to store, so sizing needs to be appropriate to match the requirements.  A cloud hosting solution can help here too.  As storage gets less expensive, the issue of how to store this ever-growing volume of data becomes less of a concern than simply how to effectively analyze large volumes of data, particularly when considering a daily time dimension.  The complexity of data increases in addition to data volumes.  Design is key at this point, since a good design can make the difference between a tool used to support the business goals or an expensive paperweight masquerading as a server.

Another benefit to a daily time dimension is the flexibility it will provide for analysis.  SAP Business Planning & Consolidation (BPC) provides the flexibility to have days aggregated to a calendar month, different types of fiscal months, weeks that aggregate into months, or even daily time configured to reflect seasonality.  In addition, BPC software supports calculations performed against the data in the database, so average daily calculations can more easily be derived, rather than calculated outside the system and then loaded back for analysis. 

In addition to a well-designed model that is required to support a daily time dimension, alternatives should be considered as well.  An obvious alternate would be a weekly time dimension since it shares many similarities without such a deep level of granularity, but there are other design possibilities as well.  Consider a “days” dimension working in conjunction with a standard monthly time dimension.  Or maybe an alternate hierarchy in the time dimension might make more sense rather than a full-fledged daily time dimension.  Create a small prototype and build some reports to make sure you can get the required aggregations as you consider alternatives. 

While a daily time dimension will not be the answer to all problems, it should be included in the mix of design possibilities and considered by your SAP BPC consultants.

Contact us to learn more.

Contact Us

Topics: Financial Reporting, SAP Budgeting and Planning, Cloud ERP