Service level guarantees are increasingly becoming one
of the major issues determining service providers'
survival in today's business climate. As service
providers try to win customer trust in order to retain
existing customers and attract new ones, they put more
emphasis on their service level offerings. A critical
requirement for service providers is the control of
resource availability to their customers -- enabling
service providers to proactively prevent service lapses
and optimize service delivery on a per-customer basis.
Customers want a service provider to deliver on its
promise of service according to a signed and binding
contract.
As customers of service providers are becoming
increasingly savvy, the issue of service level
agreements (SLAs) is quickly becoming the major factor
in selecting a service provider, and rightfully so. The
SLA has essentially become the instrument defining the "product"
of the service provider -- its service.
Several aspects of SLAs should be carefully
considered by a customer before selecting a service
provider. The first question to ask is what is covered
in the SLA? Undoubtedly a service provider who offers a
comprehensive SLA with guarantees on various aspects of
the service beyond the service availability, like the
performance level of the application, the responsiveness
of the help desk or problem notification, will be
regarded more highly than a service provider who
guarantees only the availability of the service.
The guarantees on responsiveness of the service
provider for handling service failures and other service
related issues are overlooked in many cases, but they
are key to the overall service quality. No matter how
much was invested by the service provider in backup and
redundant equipment and how well the service operation
is managed, at some point it will fail, and then the
questions become: How quickly will the service be
restored? And, how will the service provider handle the
overall situation to minimize the effect on the customer's
business?
SCOPE OF THE SLA
A comprehensive SLA should cover various issues like:
How quickly are phone calls to the help desk answered?
How often should the customer be notified on the
progress of handling cases with high severity? How long
does it take to resolve calls of different severity
levels? It should be noted that resolve time should be
defined not only in severe cases, but also with other "lower
priority" incidents. These issues may not necessarily
critically impact the customer's business, but they tend
to be ignored in many cases, leading to customer
frustration. Another issue that should be covered in the
SLA is the escalation policy regarding customer calls --
after how long and in which cases should the severity of
a call be increased?
In today's business climate, in which the stability
of many service providers is questionable, customers
should try to build a protection policy into their SLAs
in the event that their service provider goes out of
business. In cases where the service provider stores and
manages customer operational and business data, which is
the case with most ASPs, the most imminent risk to the
customer is how he can access the data and continue to
run his business if the service provider suddenly goes
belly up. Dealing with this issue is not simple, but it
can be solved with frequent backups being kept at a
location outside of the service provider's control. The
frequency and policy for generating these backups should
be defined in the SLA and the execution carefully
observed by the customer.
The importance of reporting on the actual service
levels delivered is obvious. Yet, many service providers
fail to provide this essential deliverable, in many
cases because they simply do not know the service levels
they provide to each customer, and if they don't know,
then how exactly can they guarantee it?
Customers are increasingly demanding their service
providers include a commitment to reports in the SLA.
This commitment should define the level of details in
these reports, as well as the frequency and media on
which the reports are delivered. Undoubtedly, a service
provider who allows its customers SLA compliance data
and performance level visibility, through an online
Web-based interface, will gain a much higher level of
confidence from its customers than a service provider
who provides only monthly reports.
Service providers that are willing to back the SLA
commitments with penalties, in situations where the
objectives are not met, demonstrate a higher level of
confidence in their ability to keep their promises.
Still, the question remains: Who is responsible for
pointing out SLA violations -- the customer or the
service provider? Unquestionably, a service provider who
notifies its customers about service lapses, and
consequently credits them according to the SLA, will
gain a higher level of confidence from its customers.
Such a process reduces "finger pointing" and is highly
appreciated by customers. Yet, only a service provider,
who proactively manages his service operation using
monitoring and service level management tools, can offer
such a service to its customers.
Different customers demand different things from
their service providers. Needs may relate to different
hours of operation, different performance levels,
help-desk responsiveness or other issues. It seems
obvious that they would like their requirements to be
reflected in the SLA. However, most service providers
offer only one standard SLA, or several standard levels
(for example gold, silver, bronze), and are unwilling or
unable to customize the SLA to customer-specific
requirements.
DELIVERING THE SERVICE
Service providers typically struggle with these SLA-related
issues because they lack the tools to manage their
service operation in an SLA-oriented fashion. Most
service providers use monitoring tools to control their
network, servers, and other service components. Yet,
they are handling their SLAs manually or have tools to
manage only specific aspects of SLAs. What makes it even
tougher to manage is the fact that many service
providers rely on a service supply chain from other
service providers. Their ability to fulfill their
commitment to their customers relies in part on the SLA
they have with their suppliers. Hence, they have to
manage both the commitments they receive from their
suppliers and their guarantees to their customers.
SLA management solutions for service providers are
beginning to emerge. These tools can leverage the
information received from existing monitoring tools,
correlating the delivered service levels with the
objectives set in the SLAs for both the customers and
suppliers. These tools allow the service provider to
tailor the SLA to the specific needs of the customer and
provide the customer with online reports, building
higher levels of confidence. The fierce competition
among service providers will be inflamed by these SLA-related
issues.
Hagai Schaffer is chief technology officer for Oblicore,
Inc. Oblicore has developed a service level
management solution for service providers including
application service providers (ASPs), network service
providers (NSPs), and managed service providers (MSPs).
The company's product offering provides a comprehensive
service level agreement (SLA) solution for managing
service level performance and commitments.
[ Return
To The July/August 2001 Table Of Contents ]
|