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
Generic filters
Exact matches only
Search in title
Search in excerpt
Search in content

The fundamentals of Management Information (4)

When you are making decisions about the direction and design of systems for management information and Business Intelligence, which topics and terms are important to know? Once again, it helps to have knowledge from many different disciplines including information design, IT, business management, psychology, and neurobiology.

Maps

A means of visualization with a real “wow” effect – sometimes, however, with limitations in readability and integrity. Maps are good when you cannot assume that the geography is known, for example, because someone may be interested in a breakdown by city districts. When areas are colored in a map, however, it can lead to distortions because the largest space always appears to be the most important one. It is also notoriously difficult to position columns, bars, circles, etc. in their exact locations because not every geographical unit can offer the necessary space. Maps can be difficult to read when their elements overlap. If you want to show revenues by country, a simple ranking is often the better alternative. As Bella once said, you don’t need to show anyone where England is. With DeltaMaster, you can elegantly handle maps in many different formats. Two highlights include using maps as → small multiples and geographic drill-downs.


One topic that we frequently discuss at our forums is: What factors determine if the time that you invest in management information is worth the effort?

Maximum

The largest value of a series. If the maximum is far from the next value, it is often an → outlier. If so, you will need to decide how you should deal with it. Sometimes, absolute values are better than relative ones. A maximum value may be a sign that you need to clean up data. It can, however, also be a signal of things that were once possible and play a role in motivating staff or defining future targets.

Mean

The average. The mean is susceptible to outliers while the → median is more robust.

Median

The mid value of a series. The more skewed the distribution of values is, the more the median differs from the average. Outliers distort the average; the median, however, is more typical and robust. In DeltaMaster, we try to visualize all values because as the importance of the median and average decrease, the more the values are skewed around them. You can see that immediately with just a glance.

Mobile BI

Displaying or querying data to manage your business on the go. Today, mobile BI has more to do with the output device than the location because more and more companies are working with smartphones, tablets, and other mobile devices. If reports are supposed to be equally attractive on desktops, laptops, tablets, and smartphones, the design has to suit varying screen sizes, typical attention spans, expectations for interaction, reading distances, screen contrasts, and light quality. While DeltaMaster creates reports for tablets in landscape format and with “swiping” functions, it creates ones that you can operate with your thumb in portrait format for smartphones. We have developed a smart “layering” technique to ensure that everything fits together.

Moiré effect

Flickering that occurs in raster patterns. This effect limits the use of shading in charts and → sparklines. DeltaMaster suggests using shaded columns as a → notation for forecast values. We designed them, however, so that they don’t flicker.

Mosaic Display

A chart used to show frequency distribution. These types of charts display frequency as an area. They look interesting but are difficult to understand because you need to compare areas, which is no easy task for your eyes. Bella feels that mosaic displays are worse than radar charts and suggests using → graphic tables instead.

Motivation

Necessary for taking actions. The goal of every good report is to be read, understood, and taken into consideration. We call this a “See. Know. Do.” Process. That isn’t a problem when you are dealing with intrinsic motivation. Most times, however, people only take action when they are directly affected. That is why it is so important for reports to make it past the points of seeing and understanding, and why we feel that every pixel should be taken so seriously. DeltaMaster has formats that work with strong emotional signals and have a much stronger visual force as a result.

Mouse operations

Still the most important but no longer the sole dominating way to operate a computer. You can use programs for management information and Business Intelligence on a desktop, in the Web, on tablets, or smartphones – in other words, using a mouse or your fingers – each with their own limitations. In the Web, you only have simple clicks. There is room for tooltips under your mouse pointer but not under your finger – and you can’t really tout the elegant drag-and-drop or mouse-wheel scroll features. As a general rule, the more platform-independent an application’s design is, the lower the common denominator will be as well. In DeltaMaster, we have taken a middle route and allow differences in usability when the paradigm and platform go hand in hand.

Finding your way through data and reports. Staying oriented becomes harder the further you move away from the starting point. In short, the less you have to move, the better. Reports should optimally display all answers on a page without being a → data graveyard. DeltaMaster has patented navigation methods that preserve the context so that the starting point remains clear as you delve deeper into your data.

“Neurocontrolling”

Our term for a set of business controls that are systematically supported by findings in the field of brain research. Companies set up business controls to align the actions of their employees with corporate objectives. This requires knowledge of human reactions, perceptions, and motivation. Neurobiology currently delivers the most comprehensive approach. We are teaming up with Prof. Gerhard Roth, Germany’s top expert in this field, and are integrating our findings back into DeltaMaster.

Notation

Formal rules for standardizing reports. Our concepts are geared towards analytic reporting because they should work with → graphic tables and automatic → navigation. The primary job of graphical elements in DeltaMaster is primarily to direct the eye to the numbers that deserve a second glance. Rolf Hichert’s concepts go even further and are geared towards annual reports.

Number sense

Intuitive understanding of numbers. People don’t need to count sets of numbers up to 5 – under the right circumstances, even up to 7; they simply capture them. People also have an innate sense for differences in quantities. As the quantities grow larger, these differences need to become larger as well. Since people perceive quantities and numerals on separate channels, the two combined can accelerate understanding. DeltaMaster supports number sense with the necessary cell functions: for example, tally lists.

OLAP

Online Analytical Processing. This type of multidimensional data storage delivers fast answers for typical business questions. OLAP supports different storage concepts, especially with regards to writing back data in → planning scenarios. Typically, data is extracted from an ERP system, processed, structured for analytics, and stored in one or more cubes for later access. Time will tell if → in-memory databases can handle multidimensional data comfortably. DeltaMaster can also run on relational databases but its typical usage scenario is for OLAP analysis, planning, and reporting.

Onepager

Keeping everything important on a single page. One-pagers are highly coveted by managers and casual report consumers alike due to information overload. They often contain → gauges, which we don’t think are appropriate for displaying management information, and are called → dashboards. Implementing them, however, is tricky because having everything on a single page doesn’t mean having everything at a glance. The human eye inevitably compares whatever stands next to each other (see → eyespan). One-pagers place topics and charts that have nothing to do with each other next to each other. That is terribly confusing. You can make one-pagers that are viewed daily a bit more complex because the readers quickly learn where they can find what. For occasional usage, however, one-pagers quickly get too complicated because so many different topics on a page overwhelm your readers. To stay on the safe side in reporting, we recommend focusing on one topic per page. When you change the page, therefore, you change the topic as well, which makes things easier to understand. In DeltaMaster, you can build one-pagers with the Combination Cockpit module – and would be happy to support you with our experience.

Outliers

Always a problem because they skew the chart. Oftentimes they are just accounting errors or, in the case of relative variances, artifacts. Sometimes, however, they represent the information that really should be reported. Reports should display interesting outliers in the proper proportion to the other values and omit those that are not interesting all together. Differentiating these types of outliers, however, is an art of own. → Logarithmic scales, which are still often underestimated, are helpful when displaying outliers. You should avoid references to hidden values because they simply direct the reader’s attention to something you wanted to hide in the first place. In DeltaMaster, we worked hard to find a way to deal with outliers that met our requirements – and have been successful in doing so.