|
Advanced
How To Guide
Choosing eBusiness vendors and software
2.07 Service Level Agreements
A service level agreement (SLA) outlines the rights
and obligations of both the customer and the service
provider in measurable terms, and it states the consequences
if the eBusiness vendor fails to deliver on its guarantees.
An SLA should clearly define the responsibilities
of the vendor and specify the required quality of
service - in measurable terms. But defining those
terms can be difficult. For example, a provider may
guarantee uptime of 99.999% (the much ballyhooed 5
nines), but they may not count planned downtime from
maintenance or short outages, such as the 10 minutes
it takes to reboot a machine. It is important to make
sure that an SLA guarantees the uptime for the entire
infrastructure involved in the application - hardware,
software and network - and even the
parts that the provider gets from subcontractors.
SLA terms vary according to the type of agreement
you have. There are at least four distinct types of
SLA.
1. Network SLAs
Network SLAs cover the network connection between
the customer and the eBusiness Vendor. The network
service provider agrees upon a suitable service level
for the delivery of IP services. Possible metrics
include availability, network latency, or low packet
loss.
2. Hosting SLAs
Hosting SLAs cover the hosting services provided to
the eBusiness Vendor. An eBusiness Vendor uses this
type of agreement when hardware is hosted or collocated
with a third party. Metrics vary, depending upon the
type of service performance, service-order acknowledgement,
and mean time to respond. As a customer, you should
also see this document to ensure that the eBusiness
Vendor will be able to deliver as promised.
3. Application SLAs
Application SLAs measure application performance.
The eBusiness Vendor agrees to a certain level of
responsibility, different classes of service, performance
parameters, and a manner of calculating both the demanded
performance levels and penalties that result if the
eBusiness Vendor doesn't perform its services as planned.
4. Customer care/help desk SLAs
Help desk SLAs refer to the point of contact for the
customer with the eBusiness Vendor. These SLAs may
specify how quickly a problem will be reported to
the customers after it has been identified and how
quickly an identified problem will be resolved.
2.08 Monitoring the Relationship
Good contracts do not run themselves; they require
a contract management team that keeps the relationship
in good working order. This team should ideally represent
a cross section of the people in your organisation
who are working on, and affected by, the vendors work.
It is important to include business and technical
people on the monitoring team even when the nature
of the project would seem to suggest that it was either
technical or commercial.
The contract management team should meet at regular
intervals - probably at least monthly - and should
have a performance, rather than control, remit. The
sessions should be structured and conducted in a manner
than facilitates development and improvement. In addition
to periodic meetings, more frequent basic reporting
may be required - for example, a weekly run of key
performance statistics.
It is important not to rely exclusively on the formal
monitoring structures you put in place; vendor relationship
management should be a continuous personal process.
A single overall point of contact should be established
between the organisations and information and ideas
should flow freely and informally through this channel.
In addition, specific people within each organisation
should be identified for specific problem resolution
- for example, for network outages "Mary"
should contact "Bob". This fosters ownership
of particular functions and avoids multiple complaints
to multiple inappropriate people - a relationship
wrecker.
Finally, contracts have an endpoint, as well as a
run-time, and some terminate early. Careful consideration
should be given to transition arrangements such as
data transfers and the preservation of historical
data.
|