Returning to the subject of the dimensional structure of Management Information, today we will look at the characteristics of the item dimension.
An Item is a specific performance indicator, such as sales volume, average selling price or headcount. Items may be measured in any yardstick ($, £, tonnes, units, %, bps etc.) and over any timescale.
In a rather similar way to units, items are also inter-related via one or more hierarchies. Broad “families” of items can be established, such as Profit & Loss Account, Balance Sheet, Key Ratios, Marketplace Measures, Customer Service Indicators, Employee Metrics and so on, grouping individual reporting lines accordingly.
The item dimension differs fundamentally from unit dimensions because:
- We need only to have one item dimension, rather than the possibilities that exist for multiple unit dimensions within one organisation.
- Units are generally only calculated from other units by the operation of addition (which accountants call consolidation), but some items can be calculated from other items using a multitude of different calculations.
- All items have a specific nature, and understanding this is fundamental to treating them correctly in tables and charts. We will look at this in more detail below.
At the most basic level, an item is either a stock, a flow or a statistic. A stock is an item that can be consolidated across units but not across time. Examples of stocks include fixed assets and headcount. We cannot calculate a cumulative headcount for the year to date, but instead would use the period end headcount or an average for the period we are considering.
A flow is an item that can be consolidated across units and also across time. Examples of flows include sales volume and profit. We can add these items across all our unit dimensions (products, customers etc) and also over time to create totals by month, quarter, year etc. An example of a year to date trend chart for a “flow” type item such as Net Sales is shown below.
A statistic is an item that cannot ordinarily be consolidated across units or across time. Examples include currency exchange rates and percentages such as gross margin. We cannot simply add percentages to create totals by unit or time. However, if the item is a percentage or ratio and we hold the data for the denominator of the ratio, we can use this as a “weighting factor” in calculations, and that does enable us effectively to consolidate the statistic across units or time.
Another aspect of the nature of an item is whether more than budget, last year or target should be regarded as good or bad. We can set this for each item in our data by setting the value for a field that we call “variance indicator”. If we set the variance indicator to be “favourable”, then exceeding target will be regarded as a good thing. We would set our variance indicator to “adverse” in the case of costs that we store as positive values or an item such as accounts receivable days. An example of using variance indicators to correctly colour code the values of the variances of a set of different items on a business driver driver diagram is shown below.
Understanding the fundamental nature of items and setting appropriate field values for these attributes in our data helps enormously to ensure that our treatment of their data in our visualizations and tables is correct.