IT is its own worst
enemy—Gartner and others have documented that about 80% of all Incidents
occur because of failed change management activities. However, it does
not have to be that way. Change Impact Assessment is well known outside
of IT...
By
Hank Marquis
Anyone familiar with ITIL® has heard of an
“Impact Assessment.” While there are many places in the ITIL that mention
assessing impact, this article focuses on assessing the potential impact of
changes. Change impact assessment is a systematic approach that seeks to
discover possible risks associated with a Request for Change (RFC.)
Failure to discover the risks of a proposed
change is the top reason for that popular dance, the IT 2-step. Everybody
knows it, it goes like this "take one step forward, now two steps back..."
Seriously though, most failed changes are simply the result of not taking
into account current activities, and a failure to communicate anticipated
activities.
This includes any situation which proposes
to alter system configurations, operating practices, policies, or
procedures, and any new or different activities to be performed.
For change impact assessment, or simply
impact assessment, to be effective the practitioner has to explore all of
the differences between normal operations and any conditions that introduce
risks or may contribute to a failed change.
Used effectively, impact assessment can
proactively manage risk. It is a simple idea that can be implemented
quickly using a word processor or spreadsheet -- it requires no complex software. It does
require careful adherence to a formal procedure however, and teamwork.
Following I describe how to reduce the risks
of making changes by introducing a simple, effective, formal change impact
assessment procedure based on best practices.
Impact Assessment 101
Impact assessment is a fairly mature and
formal activity in most organizations outside of IT, the military in
particular has deep experience in planning changes. Yet for some reason most
IT impact assessment consists of emailing around an RFC and waiting for
someone to comment about it, or worse, a lead tech working in isolation on
"one little change." Another popular method I see for impact assessment is a
meeting, usually the day or two before the change is to occur, where
department heads talk about the work upcoming.
However, it is clear that existing impact
assessment does not work. Every year all the big think tanks always
report that IT is its own worst enemy—Gartner and others have
documented that about 80% of all Incidents occur because of failed change
management activities. However, it does not have to be that way. Change
impact assessment is well known outside of IT, and there are models for
performing impact assessment.
While effective and easily implemented,
impact assessment is not a panacea and does not totally replace existing
change procedures, instead, you should strive to improve your existing
process. The procedure for performing an impact assessment consists of the
following five steps:
-
Define the extent of the change
proposed
-
Determine key differences in the
changed state (proposed) from a point of reference or the original
state
-
Focus on the possible effects of
the key differences from step #2
-
Sort and prioritize the possible
effects (#3) from the key differences (#2) based on risk and
possibility
-
Make a decision using the
results
Sounds easy! So lets actually try to use
our new system.
Step #1 -- Define Extent of the Change
Assuming you have in hand an RFC that wants
to "Upgrade Desktops to Vista", prepare a document (word processor is fine)
that describes the extent of the change. Be specific, and list all
boundaries for all systems. Without a clear understanding of the change, you
can't identify risks. Clearly, some of this will be in the RFC already, but
you need to make sure to capture all possible descriptions of the
environment for the change.
Take a tact of tying to identify what will
probably occur or be going on before, during, and after the change. Try to
consider as many environments or angles as possible. For example:
Probable Events With
"Upgrade Desktops to Vista"
-
Scope: Initial
upgrade IT department only, IT non-production
workstations only, no servers, no desktops, laptops
only.
-
Work Environment
-
Platform
Environment
-
Disk space
requires is 300MB
-
Transition of
existing preferences, installations, and
configurations
-
Memory required
is 2GB
-
CPU speed must
be at least 1.2GHZ
|
Table 1.
Probable Events
Your goal here is to describe, as fully as
possible, the environment and situation for the change. The more detail, the
better.
Step #2 -- Determine Key Differences
The basis of the impact assessment is to
compare the proposed state of the infrastructure or organization after the
change with the state before the change. Of course, with something new there
is no existing baseline, in which case you can choose a similar situation,
or even do research about other implementations.
The goal is to identify all of the
differences, large and small, between the pre and post -change states.
Examine previous changes. Check for similar changes in other systems,
departments, technology silos and divisions. Check for failures by reviewing
past change records, incident logs, and problem reports. Review the Internet
for documents describing what others have found and done.
After establishing your reference sources,
in a table list out all the differences between the references (failures and
success.) Don't try to sort, categorize, or judge them. Your goal is as many
types of differences as possible. Common differences include:
-
Hardware and software models,
versions, platforms, etc.
-
Staff, personnel, vendor, and
support persons, etc.
-
Process, procedures,
organization, schedules, time of day, etc.
-
Environment, location,
installation, etc.
This step is ideal for use with a
group—assemble a team, give them the reference data in advance, have them
prepare, then host a meeting where you can lead a brainstorming session to
develop as many potential differences as possible. Record them into a table,
with references, for use in the next step. Assuming the previous probable
events table (like table 1), a possible output (for example purposes only)
might look like this:
Key Differences Between
Normal and "Upgrade Desktops to Vista"
|
Table 2. Key
Differences
Here it is very important to uncover
positive, negative, and neutral trends. This means you should check for
successful changes as well. Don't forget to consider your regulatory
environment, and you can also use theoretical models like the ITIL
and others, that describe a “perfect world” and compare the change to it.
Step #3 -- Focus on Effects
Study each of the differences listed in the
table created in the last step (table 2). Ask if this difference has the
potential to cause unexpected side effects. Consider the reference
materials—has their been a history of fault or failure because of this
difference? Has this difference not had any impact? Be sure to imagine
positive, negative and neutral affects. This step is also ideal for a group.
Impact
Assessment of Upgrade Desktops to Vista |
Differences from Normal |
Potential Effects |
Risk
Mitigation
Prevention Steps |
Monitoring Steps |
|
Some IT workers only have
laptops |
They might not be in the office |
Ensure
accurate listing of all laptops and workers |
Be ready for missing
workers, and walk-in workers |
|
|
They will have work to do |
Make
sure there is a system available to users during the upgrade |
Step up inquires about
which applications are required |
|
Table 3.
Possible Effects of Key Differences
Your goal here is to consider the
differences for each probable event. For each potential effect, try to list
out what you can do to detect it (monitor) and prevent it. This forms the
basis of the impact assessment, for the starting point for risk assessment.
Step #4 -- Sort and Prioritize
Using the risk
analysis system I presented in “10 Steps to Do It Yourself CRAMM”,
make an effort to prioritize potential effect; taking into account the
effort, costs, and probability of its occurrence. Of course, full
probability assessment is beyond the scope of this article, but you can get
pretty close.
Your output is a list of recommendations,
with pros and cons, to submit to the CAB or whomever requires it. Your work
is self-justifying. That is, using plain simple terms, past experience, and
formal planning, you will present what is currently happening, what might
happen and ways to make sure the bad does not occur. This is all it takes
for leaders to make decisions and prepare.
Step #5 -- Make a Decision
Use the results of the sorted, prioritized
risk assessment to identify vulnerabilities and make recommendations to
mitigate the risks. Use the recommendations to show how to avoid, or if too
costly, respond effectively, should something go wrong with the change.
Compare the benefits of following your recommendations to the potential
costs of not following them. Decide whether the potential risk is acceptable
or not.
Summary
A formalized approach to change impact
assessment is inexpensive and can produce high quality decisions. It
requires following a process, and having the time to follow a process. While
the approach presented here is well tested and sound, it does have some
limitations:
-
This method relies heavily on
the references you choose -- choose the wrong reference or miss a
reference and it might not work as well
-
Without a CMDB that incorporates
organizational learning it is hard to find references
-
Your results will only be as
good as the knowledge and skills of the people performing the
assessment
Even with these caveats, this is still
probably the easiest and simplest method to formalize change impact
analysis. Follow these simple steps, and you just might find yourself out of
the “one step forward, two steps back” routine!
Where to go from here:
-
Discuss or comment on this article. Show your support for DITY!
- Subscribe to our newsletter and get
new skills delivered right to your Inbox,
click here.
- Download this article in PDF format
for use at your own convenience, click here.
- Use your favorite
RSS reader to stay up to date,
click here.
Related articles:
|