We use cookies. You have options. Cookies help us keep the site running smoothly and inform some of our advertising, but if you’d like to make adjustments, you can visit our Cookie Notice page for more information.
We’d like to use cookies on your device. Cookies help us keep the site running smoothly and inform some of our advertising, but how we use them is entirely up to you. Accept our recommended settings or customise them to your wishes.

Enabling personalised interactions through decision modelling

Enabling personalised interactions across all channels (both online and offline) is the service we provide to add great value to organisations. This boost in value is generated by addressing customer needs and concerns at a one-to-on level; reducing churn and increasing customer satisfaction, thus delivering a major uplift in revenue.

If the concept of personalisation works for customers, then it should work for solutions built for organisations. Solution functionality should be fully tailored and “personalised” towards the organisation to enable world class service. Being fully tailored and “personalised" means more than applying different exclusion eligibility rules or a few values to a predefined contact pressure strategy or prioritisation context weightings. It means a personalised decision framework fully modelled on the organisation's vision.

The fundamental notion that effective customer interactions can only be achieved with a personalised decision framework will be revisited throughout this blog series. Indeed, this is the fundamental motivation behind the use of Decision Modelling and Notation.

The Decision Modelling and Notation (DMN) standard was defined by Object Management Group (OMG). A summary of its roles includes:

  • helping all stakeholders understand a complex domain of decision-making using easily readable diagrams
  • provide a natural basis for discussion and agreement on the scope and nature of business decision-making
  • reduce the effort and risk of decision automation projects through graphical decomposition of requirements
  • Simplifying the development of decisioning systems using a standard specification
  • provide a structured context for the development and management of predictive analytic models
  • enabling the development of a library of reusable decision-making components.

Selling the concept of next-best-action marketing is fairly straightforward for companies that:

  • Want one-to-one communications to their customers across all their touchpoints
  • Multi-staged joined up messaging
  • Real customer journey interactions

The tricky part comes in translating requirements into reality. If you're lucky, a large requirements document exists for you to interpret. If not, you’ll need to determine the roadmap and requirements - while building the solution – a truly agile approach.

The lead business architect and his/her respective team of analysts will attempt to translate the requirements into an NBA catalogue, high-level solution documents and many PowerPoints with arrows pointing to boxes and people. Bridging the requirements gap with the Lead Decisioning Architect works for the simplest requirements, but critical requirements can get missed where complex concepts are introduced. The impact of this can range from lost time due to rework to losing a project deal. A standard way of expressing decisioning requirements would prove valuable in bridging this requirements gap. In fact, such a standard could be used in the RFP process and further refined as the pedal hits the metal. This standard is DMN.

Decision Requirements Models (DMN) are a common language for building decision management systems. This language cuts across business, IT and analytic organizations and so improves collaboration, increasing reuse, and easing implementation.

This three-part blog series will describe why the act of modelling decisions is effective and necessary for next best action marketing projects and walks through the steps involved in identifying, describing, modelling and refining decision requirements for marketing models.

Modelling decisions gets to the heart of the next best action business issues that need to be solved using all the decision-related technologies at our disposal.

Decision models are an effective means of describing decision logic in the most business-friendly and vendor-agnostic way possible.

Currently DMN is BPMN (Business Process Management Notation)-centric, and as such does not contain all the rule types and components required to describe a full marketing-centric next best action decision model. However, its open nature means that models can be widely shared and implemented across multiple technology platforms. This is DMN major advantage. In summary, it acts a great springboard along the way to design a Next-Best-Action decisions model. From a BPMN perspective DMN is also viewed as a service which drives processes. From a Next Best Action Marketing perspective DMN drives channel interactions.


Decision modelling as a means of driving clarity

A further reason for decision modelling is clarity. It's easier to engage a subject matter expert (e.g. the marketing team) by using terminology or structures which are not tailored towards a particular software platform (such as Pega or SAS). Discussing subjects such as eligibility and relevancy constructs without using an obscure language syntax as a reference should boost understanding and ultimately productivity. 

The DMN notation enables a project’s participants to get fully involved in the iterative decisioning modelling specification activities. “Iterative” is the key here, as solid platforms for NBA allow for continuous improvement and indeed are “built for change”. This change should be mostly a function of business requirements. DMN then provides that interface for the business stakeholders to be involved in shaping these changes – continuously.

The aim of using decisioning model notation is to make the final decision model design more business-driven and thus increase its acceptance, maintainability and performance (from both a business and technical perspective).


Avoiding Rule Grouping without context

In decision marketing projects there is a strong temptation to assemble all rules into a big group of rules based on the headings of either eligibility or relevancy. What the DMN forces or at least encourages the designer to do is to structure these rules into their business context and bring the rules to life based on why they are used.


Decision Frameworks

Predefined decision frameworks make a great starting point, if the requirements closely align with that of the framework. That’s the central point of most frameworks – increased productivity via pre-built assets. As the organisation matures in its use of decisioning, it will invariably extend the framework to suit its advanced needs. For example, the business may wish to apply treatment level champion-challenger rules or offline analytics as part of their scoring model.

However, decisions are very much an organisation’s secret sauce – at all levels (more on decision levels in the next section). And as such the decision framework should be specifically tailored towards an organisation's needs. Positive outcomes are more likely to exceed those of a standard framework. In much the same way a tailored and personalised message is more effective than a standard one.

If you need help with your next steps in decisioning, do reach out and contact us. Check back in December for the next part in this blog series too, when I’ll discuss how an organisation decides which decision models to progress.