* The right way to look at business services
There’s been a somewhat arrogant tradition among application vendors to equate a business service with an application. In their view, if an application is performing well, then from an IT perspective the business service is performing well.
While it’s true that certain applications or transactions may reflect outwardly facing business services, the premise that this is a one-to-one equation is fundamentally incorrect. How IT actually can begin to model true business services and what that might mean for the role of IT vis-à-vis the broader business is actually an intriguing discussion that should provoke a bit of soul searching and creative thinking among all of us.
Let’s take the case of a shipping company – call it XYZ Shipping Company. Let’s say it’s going to launch a business service – in this case shipping packages from certain locations in the U.S. to certain locations in Russia with guaranteed two-day delivery. This is a business service, in the true sense of the word. Not an application.
However, let’s look at monitoring and tracking this business service from an IT perspective. First of all, it requires a number of distinct applications, all of which have to function well to support this business service. Being no expert in this vertical, I won’t pretend this analysis is flawless – but for argument’s sake let’s assume that XYZ is progressive in its use of IT applications – for fleet management, accounting, shipping and receiving, inventory and tracking – with special attention to international (customs)-related confirmations and confirmation of delivery (validation that a particular “package transmission” is complete).
So one way to model this business service is to include all of the IT applications as critical to the health of this business service. From a traditional IT perspective, if all of the above applications are performing well, IT’s obligation to this new business service has been met.
But there is another way to monitor this: from a time-sensitive workflow perspective. Doing this you might look at the following:
* Package pickup (fleet management, dispatch).
* Delivery to the warehouse (shipping and receiving, inventory and tracking, accounting).
* Delivery to international transport and customs (subset of inventory and tracking).
* Delivery to international destination and customs (subset of inventory and tracking).
* Dispatch to localized recipient (fleet management, dispatch).
* Confirmation of receipt – validation that package has reached its destination (inventory and tracking).
Now let’s assume that XYZ has been able to “monitor” the progress of this package so that its arrival at different points in its travels can be mapped to expected performance. If expected performance is not met, an alarm would be generated. But in this case the alarm might go to staff at the warehouse to see why the warehouse was causing a delay, or to a trucking company foreman to explain why a certain pickup was slow, or in more dire conditions to XYZ’s interface with the Moscow police to query why a package has apparently disappeared after its pickup.
In other words, IT is monitoring and alerting to conditions beyond the infrastructure, while at the same time capturing the dynamics of a true business service. Each of these conditions has profound implications for the long-term future of IT.
Enterprise Management Associates has long been aware of the potential for IT to monitor non-traditional IT infrastructures – from manufacturing, to utilities, to transportation, to… fill in the blank.
This is not actually new. As long as six years ago a transit system in a major metropolitan area on the West Coast was monitored via SNMP. EMA has also seen research that indicates that there are common processes and requirements to monitor across IT and manufacturing and utilities – with common skill sets and common training requirements. A “global corporate control center” might put a whole new spin on just what a “converged” network is – making the integration of data and telephony merely a foothill in the ascent towards monitoring a truly integrated and complete business service.
On the other hand, IT’s ability to actually model a true business service has significant implications for its role vis-à-vis the line of business it supports. If XYZ has its act together it can provide, through IT, testing and modeling of its new “Two-Nights-to-Russia” service. It can far more efficiently alarm on broken processes (even if the alarm means getting a shipping clerk out of bed at 4 a.m. to backfill a flu-induced staffing shortage). And it can show costs and benefits across the entire business, leveraging the telltale, “informant” nature of the IT infrastructure. I view this as a good thing – see “informant” as a friendly, all-seeing observer, not a black-hooded spy.
It’s fun to think about the potential here for empowerment, relevance, revenue and respect. But it’s not just idle speculation, either. It may foreshadow the real meaning of business service management and the long-term future of IT.
