Organizations find creating a service catalogue difficult
It is a great mystery why service provider organizations that attempt to improve the management of their services often do not have the most fundamental document—a list of its services. How can one even imagine trying to manage services if those services have not even been identified?
Service providers are so often daunted by the prospect of creating a service catalogue for many reasons. They end up spending many months trying to develop a catalogue. Often, the result still leaves much to be desired. Part of the fault is due is the confusion sown by the tools used to manage catalogues. Part of the fault is the poor communication between the service provider and its customers. But most of the fault is due to an extremely complicated approach to an activity that ought to be quite simple.
Don’t confuse your service catalogues
Let us make it clear that we are not talking about that list of services including such things as “restore a password”, “deploy a new workplace”, “restore a backed up file”, and so forth. These are the sorts of services that many software tools know how to manage, but this is not the type of catalogue I am addressing here. Instead, I am referring to the list of the fundamental business and technical services that IT offers.
Technical Services are highly standardized
The fact is that virtually every IT service provider is using essentially the same technology stack. They are all offering essentially the same technical services. There is absolutely no reason for every service provider to reinvent the wheel and create its own list of technical services. There are only three real types of tuning that an organization should perform to complete its technical service catalogue:
- Adapt the terminology to meet the organization’s historical needs
- Remove any services from a standard list that are simply not provided
- Map the services to the products in use, which will vary considerably from organization to organization
What are the technical services?
Data Processing
This is what computers do. They make complex calculations automatically and rapidly, according to the instructions provided by software. Pretty basic stuff. Without this service, IT is out of business. Most organizations will express this service in terms of the specific platforms used to perform that processing. Each variant will have a different level of processing power, perhaps a different operating system, it may be virtual or physical, etc.
Data Transport
Today’s IT is based on networks. The service provided by network technology is data transport. The service is simply to take data from one node and delivering it reliably and in a timely fashion to one or more other nodes. This service includes the first three layers of the technology stack.
Data Management
This service stores, retrieves and otherwise manages structured data. It is generally performed today using database management software which manages, in turn, the databases where the data lives. There are today many variants on this service, although the particular technology used often depends on the application for which the database was conceived.
File and Data Storage
Storage provides the place in which files, databases and other serialized data structures are stored. Its variants generally concern the availability of the data, the read and write performance and the RPO, in case of catastrophic loss of service.
Scheduling
While IT business services are often provided via interactive applications, batch processing remains a mainstay of many organizations, especially when they support businesses with a daily, weekly or monthly business cycle. The scheduling service provides reliable and automated execution of chains of batch tasks.
Monitoring
Monitoring is obviously the service that provides timely information about the status of the infrastructure and events occurring in that infrastructure and the applications running on it.
Data Security
Not to be confused with the discipline described in such frameworks as ISO 27001, ISO 20000 or ITIL, the data security service implements the technical controls used to ensure the access to and the confidentiality and integrity of data. There are many variants to or sub-services of this service, both from an architectural perspective (is the service provided at the periphery, the core, or somewhere inbetween?) and from a technical perspective (is security provided via encryption, access controls, intrusion detection, etc.).
Data center
More often than not, the platforms used for data processing live in a data center, which center provides a series of basic facility services, such as electricity, heating, cooling, fire suppression, cabling, and so forth.
What about the Business Services?
Business services are somewhat more complex than the technical services. However, it is much simpler to define a catalogue of business services than many would think. I have found the simplest approach is to think of three layers of business services.
Layer 1
In the first layer are all the IT services that are provided to virtually all customers. These services tend to be delivered to all departments or entities within a single customer organization. As such, they are often called “corporate”, or sometimes “infrastructure” services. I would avoid using the latter term, as it creates confusion with technical services.
This layer includes such business services as electronic messaging, printing, personal productivity (word processing, drawing, spreadsheets, etc.) and other forms ofIT-based collaboration.
Layer 2
The second layer of business services includes all the IT services that a specific to a business sector. All organizations working within that sector are likely to use those services. Since there are hundreds, or even thousands, of different sectors, there is no way to provide a comprehensive list here.
Even creating a list of sectors can be a challenge. I cite the SIC and NAICS codes as attempts to classify business sectors. The challenge is not only to define the various sectors, but also to identify in which sectors any single organization might be doing business. Not only does a single business often have multiple business lines; the nature of those business lines tends to be fluid and evolves fairly rapidly.
Be that as it may, most organizations do not face the existential issue of figuring out what business they are in. The issue of a comprehensive ontology of sectors is not really pertinent to the problem of creating a service catalogue.
Let’s take as an example the sector of private banking (as opposed to investment banking, for example). Every private bank, or bank with a private banking business unit, will need essentially the same IT services (on the assumption that running the bank without those services might be feasible, but is hardly competitive today). Such a bank will use services in the areas of:
- financing
- wealth management
- credit management
- trading
- cash management
- asset management
- accounting
- controlling
- reporting
- compliance management
The great challenge today is that even though all organizations in the same sector are using essentially the same services, they tend to name them somewhat differently and draw the borders somewhat differently. As a result, the list of services appears to be very different from one organization to the next. Furthermore, each of the high level services will have many sub-services. What one organization considers a sub-service might be considered as a main service by another organization.
As for the technical services, there is a large issue with naming. Each organization needs to adapt the service offering labels to make the more comprehensible to the businesses. This work could be greatly facilitated if service providers with offerings in the same sector were to get together and hammer out a standard catalogue for the sector.
Layer 3
The third layer contains those services that are indeed unique to the service provider. As such, it is impossible to provide examples here. That being said, I doubt that there are many service providers offering such services. This layer is the zone where innovation occurs. Given the fast pace of IT, innovative services quickly become either standard services—part of layer 2—or they prove to be unworkable and are abandoned.
What comes after the list of services?
Once the list of services is defined, the next step is to define the service variants, both in terms of variants of utility, packaging of services and service levels. Once these aspects are defined, the financial aspects of the services may be established. The financial aspects depend on the overall business model for the funding of IT (cost recovery, profit center, cost+, etc.). Once this is determined, the prices of the services may be established.
Is Service Management a Service?
Where does service management fall in this structure of services? Service management is much like general management and administration in an organization. It is a general rubric covering the 40-odd disciplines, ranging from service strategy definition to incident management. In short, service management is the set of services we provide when we say that we deliver a “managed” service. For it serves to remember that we could, conceivably, offer unmanaged services. We could provide all those business and technical services, leaving management of capacity and availability up to the customers; not doing any problem management at all; telling the customers that if they want changes to be under control, they will have to do it themselves. We could do so, but the level of our service delivery would probably be so low that no customer would want our services.
So, the answer is “Yes”, service management is also a service. It is a special service that is provided in parallel with every other service in the catalogue. It is the service that allows us to say, “We will fix things if they break”, “We will ensure the evolution of our services”, “We will proactively manage our capacities to ensure that agreed performance levels are always achieved”, “We will work to improve our services, both identifying and resolving problems, but also constantly raising the bar, keeping our services as competitive as possible at the lowest costs feasible.”
Herve Meslin says
Captivating article! thanks Robert. As a side question, you are making references to a ‘Technical stack’ is it the whole layers underpinning the end to end delivery of a Service? For instance Business process layer, Business service layer, IT service layer, Systems/application layer, hosting layer, O/S layer, storage layer, network (router switches..) layer?
My primary question to you is: Do you have a reference model you would recommend?
My secondary question is have you assessed/evaluated the OBASHI framework? and if so are you in a position to comment, etc…
The main purpose of it is to be satisfy that all IT components regardless at which ever layer they are must be consistent in term of their classification/support etc (as strong as your weakest link!) it also position the relationship/dependency of IT components used for impact assessment. Another purpose would be to support the costing model by doing an aggregation of cost by layers.
Many thanks in anticipation.
Hervé
Robert Falkowitz says
Hervé, when I referred to the technology stack, I was thinking principally of the services that I describe under the rubric “What are the technical services?”
As for Obashi, rather than getting into a discussion here, I would refer instead to the Taking Service Forward initiative ) for efforts to model services and service management. There have been contributions in that context relating to Obashi.
That being said, any framework that provides a manageable and usable means for assessing impact is worth its weight in gold. The main difficulty I see is that such a framework needs a relatively sophisticated means for modeling relationships of components to services. As a simple example, it is easy enough to state that a given service uses a given router. But to say that a change or a failure in that router impacts the service depends on knowing the operational processes of the router and on the architecture of the network and the redundancy of the data transport and routing functions.
As for costing, I could not agree more. Aggregating costs up the stack makes it simply to offer, for example, a set of platforms that application owners can select, without the complexity of costing out all the sub-components and underlying technologies. Furthermore, it promotes business-like behavior on the part of the technical product owners, forcing them to understand who their customers are, what are their customers’ needs and how to ensure that the offering is reasonably priced.