Automating Decision Making With Low-Code Platforms
Automated decision making
Broadly speaking, automated decision making is the set of technologies that automate the hundreds, thousands, and sometimes millions of operational decisions that many organizations make every day. The most representative examples of such organizations are financial, insurance, and healthcare sectors. All these organizations make daily decisions with several combinations of terms and conditions, legal constraints, eligibility criteria, risk levels, and price ranges. In this blog, I explain how low-code platforms changed the way business analysts and ‘citizen developers’ author decisions with data, rules, and apps.
Authoring with rules
Business rules are not new; but until recently they were encoded in the rule syntax as “IF THIS THEN DO THAT” statements. As such, they needed detailed specifications from business analysts and skilled developers to code these business rules into rule syntax. And once the business rules were coded, they were complicated for business analysts to understand or control.
The introduction of modern low-code rule authoring platforms changed the way business rules are used to automate decisions. Rather than continuing the long and complex process of creating business rules with an IT team interviewing business people, the low-code platforms opened up a new path where the business people take control of not only rule creation, but also testing, deploying, and ongoing improvements.
Authoring with data
Gone are the days when business rule creation started with lengthy interviews where IT professionals asked business experts how they made decisions in line with company policies, industry regulations, and market dynamics. Using a low-code rule authoring platform, business analysts write business rules using a use case approach. The loaded sample data provides the context to create, test, and run rules without prior knowledge of a special rule language and syntax.
For cases where advanced logical, mathematical, and symbolic manipulations are required, business analysts can use user-friendly tools to write rules in a natural language format. For other cases where the decisioning projects necessitate formal requirements and decision modeling, some platforms propose authoring tools that adopt the Decision Model and Notation (DMN), a standard developed and managed by the OMG standards development organization. Once decision modeling is done, some platforms provide a direct path to an executable decision.
Managing with apps
As intuitive as a decision management platform can be, it may never meet the needs of a real business person. The bells and whistles that business analysts need can be overwhelming for the credit manager or insurance underwriter who needs access to decision logic. This person is certainly more inclined to exploit decision-making logic than interested in learning how to create it, and even less in training on a rules authoring tool.
For these ‘citizen developers’, low-code platforms provide web forms and point-and-click tools to take control over the configuration, deployment, monitoring, and replacement of business rules. Thanks to low-code platforms, untrained business users, business experts, and all business owners can adapt to industry regulations, company policies, and market dynamics, without IT intervention beyond the first installation.
Takeaways
- Automated decision making has moved from an old way of coding rules in a programming language to the new way of authoring them with data, rules, and apps
- The old way was time-consuming, yielded inconsistent results, and didn’t lend itself towards improvement over time. The new way is flexible to accommodate the ever-changing industry regulations, company policies, and market dynamics
- Low-code platforms implement this new way via authoring tools that business analysts can use to express their decision logic
- Business users need business applications, not authoring business rules or developing machine learning models
- Low-code platforms give these ‘citizen developers’ full control of business apps through web forms and point-clicks
Further reading
Automated Decision Making Comes of Age, MIT Sloan Management Review, July 15, 2005. A high-level article on the technologies involved in decision support in companies. Rather old article but the applications cited are still relevant today.
Use Low-Code Platforms to Develop the Apps Customers Want, CIO Magazine, November 10, 2014. A more recent article on how low-code platforms allow user feedback to be incorporated into applications during development, reducing the time it takes to go back and forth between developers and users.
Decision Making Rendered Simple and Holistic, CIOReview Magazine, August 9th, 2021. A brand new article on how the low-code Sparkling Logic SMARTS platform unifies decision authoring, testing, deployment, management, and maintenance in one tool.
About the author
Hassan Lâasri is a data strategy consultant, now leading marketing for Sparkling Logic, a Bay Area company that helps businesses leverage data, machine learning, and business rules to automate enterprise-level decisions.
Disclaimer
This post is a generalization of a post on the Sparkling Logic blog page. Although based on research, consulting, and implementation work, the views, thoughts, and opinions expressed in this article belong solely to the author, and not to the author’s current or previous clients or employers. The article was not reviewed nor endorsed by any company or organization mentioned. You can send your comments to the author at hlaasri@sparklinglogic.com.
