Next Best Action Back Inside Your Business
“Next best action” is a hot focus area in customer-facing business processes, especially marketing, sales and service. But it has just as great a potential in back-end business processes, and, in fact, ensures that many companies operate smoothly.
Next best action, in the broadest perspective, is an integration pattern for dynamic process optimization. It relies on advanced analytics to drive agile responses within and across your organization's process infrastructure. It should operate continuously both in your customer channels and in the back-end orchestrations, workflows and message interchange that help you deliver on processes you make to the customer. To the extent that we instrument our processes with real-time sensor grids, automated feedback remediation loops, embedded decision automation, self-healing network-computing platforms, and other analytic- and rule-driven systems, we can build a continuously optimized enterprise at every level.
Depending on the back-end function being optimized, the process infrastructure may be entirely embedded within a line-of-business application, such as finance, enterprise resource planning, or supply chain management, human capital, and case management. Or the process platform may consist of one or more stand-alone middleware engines for orchestration, rules, workflow and other key functions.
Regardless of the deployment model or functional scope, the back-office metrics for next best action are usually some blend of speed, efficiency, quality, agility and profitability. These supplement and support customer-centric business objectives such as retention, upsell, satisfaction, response and offer acceptance rates.
To meet these objectives continually while getting next best action to hum smoothly throughout your operations, you should combine the following core analytic infrastructure patterns:
- Decision automation: This pattern relies on programmatic elements, rather than humans exercising judgment, as next-best-action decision agents. It is enabled through "decision engines" of all shapes and sizes, including rules engines, workflow engines, and recommendation engines, and powered by predictive analytics, business rules, orchestration models, and other process content. Decision engines are usually set up to take as many automated decisions as they can in accordance with complex rulebases. They offload the most routine, repetitive, cut-and-dried decisions from human decision agents.
- Decision support: This pattern relies on humans to make operational judgments at various steps in the back-end business process. Under most decision automation scenarios, the process infrastructure must still escalate high-value “exception conditions” to people for manual resolution. Human beings, as “exception handlers,” are still very much in the loop on most automated business processes. What blend of decision automation and decision support is necessary to optimize next best action to the business process(es) of interest? To the extent that decision support is necessary, you should leverage your business intelligence, collaboration, knowledge management and human workflow environments in your next best action environment.
Analytics often power specific business processes through "solution accelerators" packaged into process applications. Generally, solution accelerators consist of application-specific predictive models, rules, metadata, schemas, reports, dashboards, calculations, and other application-specific content. Accelerators spare users from having to reinvent from scratch the process patterns associated with your vertical industry (such as telecom, retailing or financial services) or with your horizontal business process (such as customer service, marketing, sales, finance, HR and logistics). Alternately, you may rely on system integrators and other process partners to customize these accelerators for your needs. As your business process optimization program continues, you may find the need to update these process accelerators, including refreshes to embedded analytics, periodically to keep up with changing best practices.
Analytics will need to be kept fresh to optimize processes in line with the dynamic business variables, both external and internal to your company. The next best action in various business processes depends on your best predictions of such variables as aggregate customer demand, product pricing trends, availability of key factors of production, failures of critical systems and components, inventory levels under various supply chain scenarios, and cash on hand under various financial scenarios.
Clearly, that's the core of what your data scientists do, day in and day out. Stoking your back-end next best action with the best models, maintained by your best experts, is key to the well-oiled business enterprise of the 21st century.