Select a location

This selection will switch the site from presenting information primarily about Tanzania to information primarily about . If you would like to switch back, you may use location selection options at the top of the page.

Insights

Decision Support Systems: Using technology to enhance decision-making in the African legal industry

Africa Connected: Issue 6

By Jamie Theron and Hugo Meyer van den Berg

Whether drafting legal documents, litigating, or managing a law firm, legal professionals are expected to predict, with a relatively high degree of certainty, the possible consequences of every step they take and every word they communicate. Then, having considered all the possible consequences, they must decide on the best course of action. Any such decision typically involves having several alternatives, comparing them and evaluating their outcomes.1 This sounds simple enough when the alternatives are relatively few, do not include too many moving parts, and there are no time constraints on the decision-maker. However, as any legal professional will attest, most decisions taken in practice require consideration of large amounts and varying types of information, involve a complex array of criteria for evaluation, and often allow very little time to process the volumes of information. In these circumstances, trade-offs must be made between competing goals and, many times, it is not uncommon for quality to be sacrificed for speed.

With the increased adoption of digital technologies and digital business processes in Africa, accelerated by the COVID-19 pandemic, businesses have gradually shifted to digital interaction. Legal professionals have had to follow suit to stay competitive in both the local and global markets. This has led to an exponential increase in electronically stored information and, consequently, an increase in the information legal professionals must consider when making decisions.

To meet the difficulties of decision-making in the digital information age, the global legal industry is gradually transitioning to automated systems that use predictive algorithms to support and enhance the decision-making process, generally referred to as decision support systems (DSSs). Adopting these systems can have significant benefits for legal professionals, including cost-cutting, increased accuracy and depth when reviewing information, reduced time and resource expenditure and more time and flexibility for the legal professional to focus on applying the information to solve problems. If the African legal industry is to progress and compete globally, legal professionals in Africa should consider adopting these systems.

Decision Support Systems

A DSS is traditionally used to support managerial decision-making in businesses, but their use has expanded to various industries, including the medical, agricultural and (more recently) legal sectors. A DSS is an approach or methodology for supporting decision-making that uses an interactive, flexible and adaptable computer-based information system specially developed for supporting the solution to a problem.2

A DSS leverages electronically stored information to support all phases of the decision-making process. It does this by extracting the information from a database (intelligence phase), analyzing the information, identifying patterns and relationships, and constructing a model used to evaluate alternative courses of action based on set criteria (design phase), presenting the best alternatives to the end-user (choice phase) and simulating the possible outcomes of those alternatives (implementation phase). The essential components of a DSS include:

  • a centralized database such as that used in an enterprise resource planning system or a data warehouse;
  • software tools for analyzing information and constructing appropriate models; and
  • an interactive user interface that allows decision-makers to input queries and access reports, among other things.

A key feature in a modern-day DSS is the use of machine learning techniques that allow the DSS to carry out each phase of the decision-making process automatically by training the system on a sample set of relevant data and / or information. In this regard, an important distinction must be made between data and information. Data refers to individual values (facts, figures, etc), whereas information describes the relationship between several data points (ie information puts data into context). During training, the machine learning algorithm develops rules based on the relationships between data points that it will use to identify new information similar to the information contained in the training set. Thus, the algorithm predicts what information is relevant and what may be discarded as well as how that information may be applied to solve a particular problem.

Use of DSS by legal professionals in Africa

Legal professionals are making use of DSSs in varying forms. For example, DSSs are used in legal citation systems and systems used for legal research to identify and analyze relevant case law. Legal research services that make use of such systems are available to legal professionals in Namibia and South Africa, among others. DSSs are also used in practice management systems where the information from various departments in a firm is centralized and machine learning techniques are used to generate reports that help legal professionals make decisions about the business of their firms. For example, practice management systems use analytics and machine learning to determine the most efficient way for a firm to allocate its resources by, for example, tracking how work is divided in the firm and identifying who has the capacity and skills to take on new instructions, or by tracking estimated fees and actual time spent on instructions to generate more accurate fee quotes. Such systems are likewise available to legal professionals in Namibia, South Africa and other jurisdictions.

Similarly, DSSs are used in the drafting of contracts and other legal documents. This usually involves a database of contract precedents with relevant clauses and a system that identifies those clauses based on information provided by a client. Complete contracts are then generated in seconds and often require very little amendment to meet their specific needs.

Furthermore, DSSs are being used in litigation to support e-discovery,3 generally referred to as technology assisted review.4 These systems use machine learning techniques to classify and predict which of the many electronic documents subject to discovery in litigation should be withheld, due to their privileged or confidential nature, or produced to respond to the opposing side.5

Adopting Decision Support Systems: Benefits and drawbacks

At the forefront of adopting a DSS is the cost-cutting benefit. The tasks of document review and information organization are “delegated” to a DSS, leaving the legal professional free to focus on higher level tasks and reducing the costs associated with those tasks.6 A further benefit is that automated systems are often less biased, more consistent and more predictable than human beings.7 Thus, the relative accuracy of a DSS, and the depths it may achieve, far exceeds that of a human being and such a system can do the same amount of work in a fraction of the time.

One of the major drawbacks of a DSS is that it may produce unintended and varying results depending on the specific algorithms and training data used. As each system is designed to solve a specific problem or problems based on defined criteria, the system may not produce the intended results if the criteria are poorly defined and the algorithm and training data do not align with the purpose of the system. The interconnectivity of a DSS means that a failure in one part of the system, such as analyzing information based on incorrect criteria, could corrupt the entire system.

Challenges in adopting a DSS

The initial capital investment of developing and implementing a fully operational DSS can be prohibitively high for certain legal departments and smaller law firms. The benefits of adopting such a system are not always measurable with traditional business metrics and as such the investment in the system may be difficult to justify, especially to those unfamiliar with how the system works. The systematic or phased adoption of a DSS may help reduce the initial costs of developing and implementing a DSS. The starting point would be to set up a database where information relevant to a particular problem is electronically stored. Many legal professionals are already doing this by uploading documents to cloud-based storage services or to on-premises storage facilities. Gathering and centralizing information is the first step in developing and implementing a DSS as this will allow easy access to relevant information for analyses. Gradually, legal professionals may move towards adopting analytical tools to analyze and interpret their stored information. This may mean having custom analytics software developed or purchasing an off-the-shelf product. Such software will usually include a user interface and options for how information will be presented to the end user (ie what reports will be generated to support decision-making).

The technical aspects of a DSS can present another challenge to adopting the system. People generally, and legal professionals specifically, find it difficult to trust what they do not understand. Thus, the adoption of a DSS may be dismissed before it is even brought to the table. Those who do understand the systems are often the young legal professionals who cannot afford or do not have the authority to implement the system. Some legal practitioners may feel that automated systems will “take-over,” making human input redundant. However, DSS are not intended to replace human beings. Human judgment is still necessary to evaluate the results produced by the DSS and physically implement the suggested course of action. Furthermore, the design principle of contestability has been suggested as a means to address the issue of human involvement in the DSS process.8 Contestability refers to mechanisms integrated into the DSS that allow users to understand, construct, shape and challenge model predictions. The goal of this is for the DSS, and particularly the predictive algorithms that drive them, to enhance and support human reasoning in the decision-making process. Such interactive, contestable systems can also improve user understanding and allow the user to provide deep and useful feedback to improve algorithms.9

The challenges presented above are those typically experienced by legal professionals in the adoption of a DSS. In the African legal industry, these challenges are magnified by a lack of investment in technology by legal professionals and slow adoption of technical systems. Furthermore, technical proficiency is generally lacking among legal professionals and the adoption of technical systems may not be seen as a priority. However, this may be due in part to the lack of understanding of the benefits that such systems can bring.

Footnotes
1Eilon S. 1969. What is a decision? Management Science, Dec 1969, Vol. 16, No. 4. Application Series. B-172.
2Turban et al. 2005. Decision Support Systems and Intelligent Systems 7th Ed. p 103.
3Ediscovery in South Africa and the Challenges it Faces
4Klutz, D.N et al. 2019. Automated Decision Support Technologies and the Legal Profession. p 853.
5Ibid: 853.
6Ibid: 872.
7Ibid: 875.
8Ibid: 887.
9Ibid: 888.

Authors