One of the strengths of Business Intelligence is that it quickly provides great insight into the development of data over time. With Business Intelligence, for example, you can instantly see what this year’s sales are, and how they compare to last year’s sales. This kind of insight is possible because transactional data has a date such as an invoice or ledger date.
However, there is often static data too, which cannot be directly related to a date. Think of master data such as item master data or monitoring data such as the current status of a machine. All these kinds of insights are called ‘As Is’. A good example of As Is data, is the current status of outstanding invoices. You often request As Is data at some point, just to briefly check the status. But often you also need insights in a development over time. Are there more or fewer items than last year? Are machines experiencing more or less breakdowns? How is the payment behavior of my customers now and how was it last year? These kinds of insights can be called ‘As Of’. With As Of, you look at the state of affairs at a certain point in the past by diving into the history of the data.
Tableau makes it child’s play to get these kinds of insights. In the world of data warehousing, this is often referred to as snapshots (pictures). Where the ETL process pulls the data from the source systems, it is possible to create these kinds of snapshots. It is important to think carefully about how often this is done, the time period over which these snapshots should be available, and the volume of data.
We once received a request from a customer for insight into the turnover rate of their inventory. To gain insight into the turnover rate, a daily inventory count over a two-year period was needed. However, there were 160,000 inventory items, so the dataset consisted of nearly 120 million records. A hefty server can obviously handle this quite well, but it is questionable whether the benefits outweigh the costs. The same insight could probably be obtained with one picture per month. You also need to think about data cleansing, because not all historic data is relevant anymore.
These snapshots make it very easy to visualize the state of affairs over time in Tableau. You can also compare the current state with any snapshot from the past. This allows you to infer trends as well as take advantage of all the other features of Tableau such as setting up notifications and much more.
See below an example of a Tableau dashboard in which we have combined both ‘As Is’ and ‘As Of’ functionality. This dashboard provides an overview of accounts receivable. We created this dashboard so that you can quickly see the current state of affairs (as is). In fact, when you open the dashboard, you will automatically see the state of affairs up to the most recent data. Would you like to see the state of affairs at a specific time in the past? Then you can set the ‘As of’ parameter in this dashboard to a desired date in the past.
Try out the interactive dashboard for yourself below!
As a Senior BI Consultant at Cadran Analytics, Rick is highly specialized in Tableau.
Floris is Analytics Consultant and Data Scientist at Cadran Analytics.
