Information Systems in the Corporate World: Part 4 — Management Support Systems
And so we’ve reached the final instalment of Information Systems in the Corporate World. The aim of this series was to provide an informative backdrop to the different information systems we see in large companies on a daily basis as well as their different applications — which my peers went into in more detail over the last few weeks. The purpose of this final blog post is to look at the top two rungs of the corporate information systems pyramid, (remember this..)
and discuss their day-to-day applications in the corporate world. Decision Support Systems (DSS) and Executive Information Systems (ESS) are typically used in tandem by the upper echelons of management, so I’ve decided to analyse them together rather than give them individual posts. Both of these systems can really be put under the umbrella term of Management Support Systems as they essentially utilise the information from across the company to support high-stakes, strategic decision making by senior management and executives.
Decision Support Systems
We’ve already discussed Transaction Processing Systems (TPS) and how they are essentially the source of all the raw data that can be used by a corporation, then we went on to describe how Management Information Systems (MIS) then collate all this data to provide meaningful information, and we used enterprise resource planning in a manufacturing environment as an example of this. So, a Decision Support System (DSS) is essentially the process of utilising information from MIS to aid strategic decision making for the future. DSS is often conflated with the term ‘Business Intelligence’ as it is here where data-driven models are often implemented to forecast future growth and opportunities in a corporation as well as potential pitfalls (Olavsrud, 2021). This information is often presented in management reports for predicting future trends as well as computer dashboards for real-time day-to-day information.
A brief history
The question at the top of my mind when conducting initial research for this post was ‘what distinguishes DSS from MIS?’ It turns out that by looking at how DSS came about in the 1960s, we can see that it was really built on the shoulders of MIS. Prior to 1965 it was very expensive to build large-scale information systems until MIS was invented, which was focused on providing managers with structured, periodic reports with much of the information coming from accounting and transaction systems, as previously discussed. It was in the late 1960s when a new type of information system became practical: Model-oriented DSS; or Management Decision Systems. The two DSS pioneers , Peter Keen and Charles Stabell, claimed the concept of decision support evolved from “the theoretical studies of organisational decision making done at the Carnegie Institute of Technology during the 1950s and early ’60s and the technical work on interactive computer systems, mainly carried out at the Massachusetts Institute of Technology in the 1960s”. (Power, 2002) Hence, what DSS brought to the table was an interactive information system that allowed for large datasets to be analysed and manipulated to assist decision-making going forwards. This basically marked a shift from gut instinct dictating important business decisions towards more modern data-driven frameworks. Today, we now see many branches of DSS being used extensively like Data Analytics, Machine Learning and intelligent algorithms. I am going to take a closer look at different types of models.
Decision Support Systems are almost always unique to each individual company, hence there are very few specialised models or software packages on the market, although the incumbent Information System providers; Oracle, SAP, IBM etc. do build general software that can be adapted for each company. Most decision support systems and models are really built in-house or with the assistance of consultants who help corporations put frameworks in place to solve problems and meet their goals. Hence, it is difficult to go into depth on a particular company or industry as the problems being solved are generally unstructured, unique to the business, expensive and often not public knowledge. Some types of models do tend to reappear frequently however, such as:
Decision Analysis Models
When corporations are attempting to identify and evaluate different plans of action, each with their respective pros and cons, decision support systems are often designed and implemented. Specifically, they must impose analysis techniques and suitable frameworks in order to fairly compare the benefits and opportunity costs of each viable option and eventually choose one. These systems are rarely carried out just using software as not all pros and cons can necessarily be quantified and put into a software programme. Instead, decision analysis models like the Analytical Hierarchy Process (AHP) are used.
AHP is a multi-criteria decision technique that combines quantitative and qualitative factors when evaluating alternatives. It begins with developing a hierarchical representation of a problem, with the overall objective on top, decision alternatives at the bottom and relevant attributes and selection criteria in between. Relational data is then used to derive the priority of each objective and a consistency ratio which gives management a quantitative overview of what objectives are the most valuable to pursue (Juneja, 2015). Other models like decision trees and influence diagrams can be used in conjunction as a means of providing more graphic and intuitive decision support systems. A version of decision trees are often used in management consulting companies like McKinsey where a MECE (Mutually Exclusive Collectively Exhaustive) approach is taken so that all potential solutions (or ‘hypotheses’) are kept concise and directly-related to the problem at hand (Rasiel, 1999).
Financial & Accounting Models
Budget Financial Models and Ratio Analysis are examples of financial models that are often used by decision-making figures in large companies as well as investors. Often these models are used to build logical relationships between data and the progress of the company and can be designed to provide a snapshot of the business performance at any one time. Other models such as Discounted Cash Flows (DCF) can be used to project future revenue streams, expenses etc. and can prove very useful for future planning.
For purely quantitative problems, forecasting models can often be used to extrapolate further trends. There are many well-known, simple quantitative models in existence today such as Moving Averages, Weighted Averages, Exponential Smoothing and regression statistical analysis that are very quick to implement using computer software programmes or custom scripts using Python and/or other suitable languages. These can then be adjusted for each company’s needs or further developed using intelligent algorithms and time series models (E. Sayed, A. Gabbar, A. Fouadc and M. Ahmedc, 2008). The major caveat of course is that historical data is not always an accurate predictor of the future and as is especially relevant today, black swan events are often unaccounted for and can spring harsh surprises.
Executive Support Systems
This leads onto the final type of information system used in large corporations, Executive Support Systems (ESS). This is arguably the most important information system for determining the long-term direction of a corporation however it is also the least structured due to it’s implementation in the face of uncertainty and dynamic external information. ESS, unlike DSS and MIS must take ‘the larger picture’ into account and therefore utilises information from both external sources (i.e. market movement, competitor analysis, technology advances) and internal sources (typically TPS data aggregated using MIS).
Like DSS, large companies often have different Executive Support Systems from one another as they are very much tailored to each executive suite’s needs, but some common characteristics include (McKinsey, 2010):
• They extract, filter, compress, and track critical data. This essentially allows company leaders to digest information quickly and act promptly.
• Access and integrate a broad range of internal and external data. As mentioned, data sources are widespread for an information system like this. A well designed ESS programme will scrape data from both external sources (stock market, internet etc.) and internal sources (management accounts and reports) to provide accurate, up to date information.
- are user-friendly, and require minimal or no training to use. As ESS requires direct access by executives at all times, it is imperative that the interface is designed to be intuitive and quick.
The biggest problem with ESS today is the potential for critical data to be trapped within inefficient and unnecessary organisational silos within the company. This case of bureaucracy is an ever-growing headache for multi-national corporations especially where data can be lost in transit, translation or even physically. This has actually become such a significant problem for corporations today that some have even elected a Chief Information Officer (CIO) to the C-suite in recent years. The role of the CIO is to ensure the functionality of information systems is fit for purpose. If not, performance suffers. A CIO is responsible for the business intelligence architecture of the company. This is achieved by managing the inflow of data to the executive suite and ensuring it is accurate. According to a report by McKinsey (McKinsey, 2010), three major factors that often hinder success in this regard are:
- Inconsistent, unreliable content
- Poor oversight and system handling
- Inflexible business/IT architecture
Meanwhile, an ESS that avoids these pitfalls and delivers a streamlined approach allows a corporation to remain nimble and exploit market opportunities faster than their competitors.
So, that’s it. From transaction process systems to executive support systems I think I’ve given a brief insight into each of the information systems used at each level of a typical large cap corporation. There is so much more that could be discussed however the reality of information systems is that they are often tailored to the preference of individuals and companies, so it is not an exact science.
When researching these systems, I was often curious about the the level of innovation we are currently seeing in information system technology (or lack of it). Today we are acutely aware of the great strides being made in information technology, and I absolutely agree. For example, transaction process systems are currently seeing a complete revolution in terms of how we utilise data to create value — I talked about Stripe to encapsulate this incredible progress, and it has since become the most valuable private company in Silicon Valley! The revolution of data analytics and machine learning in recent years has also led to highly advanced quantitative models becoming common in DSS, and as a student it is clear that the jobs market for graduates with analytical skills is a lucrative one as companies are trying to leverage their vast databases more effectively.
Meanwhile, I cannot say the same for MIS and ESS. MIS systems such as Enterprise Resource Planning software seem to have provided the same standard features for years. I think this may be due to the longstanding monopoly shared by the 5 or 6 incumbents of MIS software (looking at you, Oracle and SAP) or maybe just a shear lack of imagination on how MIS can be improved, however times may be changing if the recent shift by Boeing to Systems Applications Projects is anything to go by.
ESS seems similar, although granted I’m in less of a position to declare the stagnation of this type of information system as I’m yet to to see it in action. Regardless, looking forward I will be keenly looking out for disruption in the management and executive information systems market. There is no doubt that more efficient and cost-effective information systems in large corporations could provide tremendous value. My own prediction? I suspect that we will see more AI and ‘big data’ powered information systems in the future as automation begins to work up the hierarchy from workers to management level.👀
- Olavsrud, T., 2021. Decision support systems: Sifting data for better business decisions. [online] CIO. Available at: https://www.cio.com/article/3545813/decision-support-systems-sifting-data-for-better-business-decisions.html#:~:text=A%20decision%20support%20system%20(DSS,data%20for%20informing%20business%20decisions.&text=A%20DSS%20leverages%20a%20combination,to%20help%20users%20make%20decisions
2. Power, D., 2002. Decision Support Systems: Concepts and Resources for Managers. Westport, CT: Greenwood Publishing Group.
3. Juneja, P., 2015. Building Model-Driven Decision Support System (MDSS). [online] Managementstudyguide.com. Available at: https://www.managementstudyguide.com/building-model-driven-decision-support-system.htm
4. Rasiel, E., 1999. The McKinsey way. New York: Washington, D.C.
5. E. Sayed, H., A. Gabbar, H., A. Fouadc, S. and M. Ahmedc, K., 2008. A Forecasting Decision Support System. Fourth International Workshop on Computational Intelligence & Applications IEEE SMC Hiroshima Chapter, Hiroshima University, Japan,.
6. McKinsey Digital, 2010. Data to dollars: Supporting top management with next-generation executive information systems. [online] McKinsey. Available at: https://www.mckinsey.com/business-functions/mckinsey-digital/our-insights/data-to-dollars-supporting-top-management-with-next-generation-executive-information-systems.