Capstone: redesigning service support for an organization

This capstone brings the course together. You will diagnose the support practices of an invented organization, Juniper Ridge College, and design improvements across the service desk, incident, problem, service request, change enablement, monitoring and event management, and knowledge. The lesson works through the analysis step by step, with numbers you can check, and ends with a strategy for answering practice questions on the Foundation exam. Juniper Ridge College is fictional; any resemblance to a real institution is coincidental.

The organization

Juniper Ridge College is a public community college in Minnesota with 14,000 students and 1,600 staff across a main campus and two smaller campuses. Its IT department of 70 people supports a learning management system, a student information system, email and collaboration tools, campus Wi-Fi, classroom technology, a finance and HR system, and around 6,000 college-owned devices. Student records are protected by the Family Educational Rights and Privacy Act (FERPA), the college must meet accessibility requirements under the ADA and Section 504, and it offers support in Spanish and Somali for many students.

Leadership is unhappy. A student survey rated IT support 2.9 out of 5. Faculty complain that classroom technology fails at the start of lectures. Two outages of the learning management system during exam weeks last year made the local news. The IT director has asked for a redesign of support practices within existing budget, with a business case for any investment.

Current state: the data

A month of data from the service management tool, interviews and a quick review of practices gives the following picture.

Step 1: Understand the vision and value

Following the continual improvement model from ITIL-120, start with the vision. Interviews show what students and staff value: being able to learn and teach without interruption, especially at the start of lectures and during assessments; quick help that does not require a phone call; and clear information when things go wrong. The IT director's goals are to raise satisfaction to at least 4 out of 5, eliminate exam-period outages and reduce cost per contact. These become the outcomes the redesign must serve.

Step 2: Diagnose with the practices

Each finding can be traced to a practice and a cause. The service desk is overloaded at term start partly because of avoidable contacts: 1,250 lockouts a month, most of which self-service could handle. First contact resolution is low because agents have no knowledge base. Nine percent of incidents at P1 suggests priority inflation, which dilutes the attention genuine P1s get. Problem management is reactive and slow, so recurring Wi-Fi and projector issues never get fixed. Requests are manual and unstructured. Change enablement is both slow (everything waits for the weekly CAB) and unsafe (untested emergency changes in exam weeks). Monitoring produces noise without coverage of the most important service. Knowledge is effectively missing.

Step 3: Design the improvements

Improvements are grouped by practice, prioritized by value and effort, and sequenced so that quick wins fund and build support for larger changes.

Step 4: Quantify the case

The program below estimates the effect of the main improvements on monthly contacts, service desk workload and user experience, using the current-state data and assumptions agreed with the team. The assumptions are deliberately conservative and labeled so they can be challenged.

# Juniper Ridge College: before and after estimates (assumptions are illustrative)
contacts = 9000
lockouts = 1250
desk_minutes_per_contact = 9.0
agent_rate = 38.0                 # loaded cost per agent hour, dollars

# assumptions
reset_share_removed = 0.80        # lockouts handled by self-service reset
kcs_deflection = 0.10             # other contacts resolved by self-service articles
fcr_before, fcr_after = 0.48, 0.65
escalation_cost = 30.0            # second-line cost per escalated ticket, dollars

removed = lockouts * reset_share_removed
remaining = contacts - removed
deflected = (remaining) * kcs_deflection
desk_contacts_after = remaining - deflected

def monthly_cost(n, fcr):
    desk = n * desk_minutes_per_contact / 60 * agent_rate
    escalations = n * (1 - fcr) * 0.8      # 80% of unresolved first contacts escalate
    return desk + escalations * escalation_cost, escalations

cost_before, esc_before = monthly_cost(contacts, fcr_before)
cost_after, esc_after = monthly_cost(desk_contacts_after, fcr_after)
print(f"desk contacts per month:   {contacts:,} -> {desk_contacts_after:,.0f}")
print(f"escalations per month:     {esc_before:,.0f} -> {esc_after:,.0f}")
print(f"support cost per month:    ${cost_before:,.0f} -> ${cost_after:,.0f}")
print(f"monthly saving:            ${cost_before - cost_after:,.0f}")
investment = 85000.0               # password reset, portal forms, KCS training, synthetic monitoring
print(f"payback on ${investment:,.0f}: {investment / (cost_before - cost_after):.1f} months")

desk contacts per month:   9,000 -> 7,200
escalations per month:     3,744 -> 2,016
support cost per month:    $163,620 -> $101,520
monthly saving:            $62,100
payback on $85,000: 1.4 months

Even with conservative assumptions, the improvements reduce desk contacts and escalations substantially and, on these assumptions, pay back within a few months. The figures should be presented with their assumptions, and the team should agree how each will be measured after implementation, so that the business case can be checked rather than just believed.

Step 5: Protect exam weeks

The most visible failures were outages during exams, so they deserve a specific design. Change enablement adds a freeze on non-essential changes to the learning management system and student systems for the two weeks before and during exams, published on the change schedule a term in advance. Any change in the freeze must be an emergency change with documented test evidence and approval from the service owner and the dean responsible for assessments. Monitoring and event management adds synthetic checks of login and assignment upload every two minutes, with alerts to on-call staff. Incident management defines an LMS major incident procedure with a communications template for students and faculty, and an agreed decision rule for when the college extends assessment deadlines. Problem management reviews last year's two outages and confirms that their causes are fixed.

Step 6: Govern, measure and sustain

The improvements are governed through a monthly service review chaired by the IT director with student and faculty representatives. A balanced set of measures tracks progress: CSAT and self-service success for experience; first contact resolution, abandonment and fulfillment times for the service desk and requests; P1 share and major incidents per term for incident management; repeat incidents in the top categories for problem management; change failure rate and share of standard changes for change enablement; actionable alert share and share of incidents detected by monitoring for event management. Each measure has an owner and a target for the end of the academic year.

Privacy and accessibility shape the design. Articles and portal forms must be accessible under the ADA and Section 504, and the Department of Justice's 2024 rule for state and local government websites points to WCAG 2.1 AA, with compliance dates that were later extended. Student records in tickets and in any AI tools must be handled in line with FERPA and Minnesota law, and the college should check with its privacy office before sending ticket data to external AI services. Articles in Spanish and Somali should be offered for common student issues.

Exam strategy for practice questions

Practice questions on the Foundation exam test whether you can recognize the purpose of each practice, apply key terms correctly and choose the best action in a short scenario. Several patterns recur.

Read each question for its key words: interruption or reduction in quality points to incidents; cause points to problems; predefined or routine points to service requests; risk and authorization point to change enablement; change of state points to events. When two options both look reasonable, prefer the one that matches the practice's purpose most directly and reflects the guiding principles, especially focus on value and keep it simple and practical. ITIL-160 provides full mock exams and timing advice.

Reflection

The Juniper Ridge redesign shows how the practices in this course work together. No single practice fixes the college's problems. Self-service and knowledge reduce demand on the service desk; better prioritization and a major incident procedure focus attention; problem management removes recurring causes; risk-based change enablement makes routine change faster and protects critical periods; and well-designed monitoring detects failures before users suffer. Measurement ties the improvements to the outcomes stakeholders value. ITIL-141 continues with the delivery practices that complete the picture, such as service level, configuration and supplier management.