There was a time when letting people know about something required calling a land line and hoping the person happened to be near it. Otherwise, you needed to physically track them down. Then voicemail came along, offering an additional way to provide notifications.
The contrast couldn’t be greater between that era and today, when we find ourselves bombarded with notification emails, text messages, social media alerts, and even robot phone calls—some that we choose to receive and many we don’t even recognize.
That shift is driven by an even larger transformation in how we communicate and act. The many means of notification are part and parcel of a world where the speed of information—and reaction—has become paramount. That’s not only true in our personal lives, it’s just as true in the business world.
The role of notifications
Take manufacturing as an example. “Just-in-time” emerged in the 1980s to describe a significant manufacturing transformation: No longer would manufacturers warehouse large inventories of parts and components, waiting until they were needed. Instead, coordination throughout the supply chain would make it possible for suppliers to make precisely timed deliveries of needed parts and components directly to the manufacturing line.
The efficiency and cost savings from this manufacturing transformation led to adoption of similar just-in-time approaches in other industries. Perhaps most famously, Walmart led the way in integrating with its suppliers to the point that those suppliers took over responsibility for monitoring store shelf inventory for their products, delivering supplies exactly when needed for restocking. Today, just-in-time principles have caused a large-scale migration of business operations away from inefficient batch-at-a-time approaches to those driven by immediate reaction to data and events.
Notifications are a key building block in this new event-driven and data-driven paradigm. Notifications provide the ability for services and applications to communicate information in real time, not only to move data but also to trigger actions for users and applications. Whether the technology supporting those notifications is an enterprise service bus, a message queue, or a messaging backbone, notifications act as the glue between different applications for processing and acting on data.
To illustrate, consider the role of notifications in logistics. Examples include:
- Retail companies using marketing notifications to alert customers to new merchandise and offers with the aim of enticing evaluation and purchase. Another example: notifications throughout the retail supply and delivery chain to monitor movement of merchandise from suppliers to warehouses and physical stores, or from distribution centers directly to customers—thus allowing necessary adjustments based on that information.
- Manufacturers rely on notifications to manage supply chains and distribution, allowing them to fine-tune delivery and pickup schedules to minimize inventory. Notifications also ensure smooth operation of manufacturing lines, allowing manufacturers to rapidly detect and correct conditions that could otherwise impair or halt production.
- Freight and delivery providers have long used notifications to give clients information about progress of deliveries and services. Now the expanding deployment of RFID, and even newer IoT (internet of things) sensors and connected devices, has created a whole new class of logistics notifications, triggering a range of processes and actions deep within the supply chain. For example, temperature sensors for perishable goods that generate notifications and re-routing when goods in transit are exposed to potentially harmful temperatures, or the notifications that drive just-in-time updates to delivery routes and schedules, and more.
A common characteristic across these scenarios is the need to trigger and act on notifications as quickly as possible. Delays will lead to inefficiencies and failures, ranging from missed delivery windows to extra warehousing costs to spoiled merchandise and more.
Technology challenges
Achieving that notification and reaction speed isn’t easy; a solid technology foundation is necessary to efficiently generate notifications at scale, as well as process them as quickly as possible for action. That foundation includes these core components:
- Event collection: Notifications are in many cases the result of events that are created by sensors or applications. For example, GPS readings from devices applied to a delivery vehicle. These events need to be collected and processed to identify conditions that trigger a notification, for example that a particular delivery vehicle has entered a specific warehouse.
- Workflow processing: One or more systems are needed that generate, communicate, and track completion of process steps and associated data. For example, the submission of a shipping bill might initiate a workflow of document processing, job scheduling, and route planning that need to be executed in other systems. Although there are some specialized workflow management platforms, workflows are frequently processed using application logic that uses enterprise service bus or message-queuing technologies.
- Delivery: Messages and their accompanying information are used by delivery systems to create, queue, and deliver notifications to recipients, whether those recipients are people or other applications. The communications systems for message delivery may use cellular networks or email, as well as the enterprise service bus. Alerting solutions, messaging solutions, and even message bus systems are examples of common message delivery technologies.
To build the right solution, the event collection, workflow and delivery components need to be evaluated with critical enterprise requirements in mind:
- Elasticity: Both the events that generate notifications, as well as the notifications themselves, do not happen in a consistent, steady flow. Variability can be driven by seasonal fluctuations in consumer demand, traffic, conditions on a manufacturing line, etc. In every case, the volume and velocity vary in both predictable and unpredictable ways. Technology needs to be able to adapt quickly and elastically to these varying workload demands to meet target response times.
- Scalability and performance: Delivering those target response times also requires a technology foundation designed to maintain low latency and high throughput even as demands scale.
- Workflow support: To simplify and standardize development, technology solutions need to support the work queue processing, multistep processing, and dynamic data routing that are necessary to manage application response workflows.
- Data-driven processing: Traditional batch-oriented technologies—whether for data integration, notification scheduling, or analytics processing—are not the right technologies for ensuring that events are processed, and notifications are generated, immediately. Streaming and real-time technologies are necessary to support the data-driven paradigm.
- Durability: Notifications are often used to transmit critical alerts that users and applications rely on. As a result, solutions to support them must be designed with resiliency in mind to ensure that notification processing and delivery is guaranteed.
Case in point: Reverse logistics in retail
To illustrate, consider a scenario my company has seen in its work with customers on implementing event-driven notification platforms. In this particular scenario, notifications play a key role in reverse logistics—the act of processing customer returns. Most people are familiar with this scenario through experience with internet retailers, who have used easy returns to lower barriers to online purchases, but reverse logistics is also important in business-to-business commerce as well.
While making it easy for customers to return goods may increase sales, it also creates significant costs for sellers. Beyond the freight costs associated with returning items to a warehouse, there are also storage and processing costs in inspecting the returned item and, when needed, restoring it to a condition that allows placement back into available inventory. The longer that process, the greater the immediate costs for processing and warehousing, not to mention the opportunity cost of not being able to sell the item quickly.
Real-time notifications are a critical part of reducing those costs. Consider the increased efficiency realized in a notification-driven approach: As soon as a shipping bill or RMA (return merchandise authorization) is received by a business application, notifications and workflows begin for processing that item and preparing its availability for sale. As the item makes its journey back to the seller, events generated by sensors, cameras, devices and more provide ongoing updates regarding not only the location of the item in transit, but even potentially its condition and related indicators of what might be needed to prepare the item for resale. With that information, workflows can be processed to speed purchase availability even before the item has arrived back at the seller’s facilities, reducing the time and cost of storage, as well as accelerating revenue booking from the sale.
That is but a single example of the business impact of event-driven notifications—not only in improving speed and efficiency for existing processes, but even in creating entirely new revenue opportunities. In fact, the most forward-thinking enterprises are building intelligent systems that automate the ability to act on notifications, eliminating the need for human intervention at every actionable step. Yet whether automated or not, the volume and velocity of notifications can quickly spiral out of control (as we’ve all experienced on a personal level). The savvy enterprise will take this into consideration and build new notification services accordingly, not wait until the wheels come spinning off. Because the one certainty with notifications in a modern enterprise is that over time, there will always be more.
This article is published as part of the IDG Contributor Network. Want to Join?
