Call Center Operations Case Interview: Worked Capacity Case
| Case type | Operations |
|---|---|
| Industry | Travel / Customer Service |
| Difficulty | Hard |
| Firm style | BCG |
What does this call center case ask you to diagnose?
Our client is TravelLine, an online travel booking platform. They recently centralized customer service from 12 regional offices into one national call center. The center is overwhelmed — customers face 20-plus minute wait times and satisfaction scores are dropping fast. The CEO needs to understand why and what to do about it.
TravelLine migrated 780,000 active customers from 12 regional offices to a single national call center over 50 weeks. The call center currently has 45 trained agents. Customers are experiencing 20+ minute wait times, and NPS has dropped 25 points. The CEO wants us to diagnose the root cause and recommend solutions.
Which exhibits reveal the capacity gap?
Exhibit 1
Call Center Capacity Driver Tree
Driver tree linking weekly call volume and per-agent capacity to the number of required agents.
Show the data behind Exhibit 1
| Component | Value | Notes |
|---|---|---|
| Average Call Time | 210 seconds | Time on the phone with customer |
| Average Wrap-up Time | 90 seconds | Post-call notes and system updates |
| Total Settlement Time | 300 seconds = 5 min | Call time + wrap-up = 1/12 hour |
| Calls per Customer per Year | 2.5 | Average across all customer types |
| Weeks per Year | 52 | Used to convert annual customer calls to weekly calls |
| Working Hours per Week | 40 hours | Standard full-time schedule |
| Gross/Net Working Ratio | 78% | Net of breaks, meetings, admin |
| Working Time Duration | 55% | % of net time actually on calls |
| Seconds per Hour | 3,600 | Used to convert productive hours into call capacity |
| Seasonal Peak Factor (Summer) | 1.30 | 30% more calls June-August |
Exhibit 2
Cumulative Customers Migrated to National Call Center (thousands)
Cumulative customers migrated to the national call center (thousands), by week (weeks 0-50).
Show the data behind Exhibit 2
| Week | 0 | 5 | 10 | 15 | 20 | 25 | 30 | 35 | 40 | 45 | 50 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Customers Migrated (K) | 0 | 50 | 120 | 200 | 300 | 420 | 550 | 650 | 720 | 760 | 780 |
BCGOperations · hard
TravelLine Call Center Optimization
Travel / Customer Service
How a strong candidate structures it
A strong call center operations framework
Question by question
- 1
Case context
Understand the Case
“Can you summarize the situation and what TravelLine needs from us?”
- 2
Clarifying
Clarifying Questions
“What would you like to know before analyzing the problem?”
- 3Drill the structure
Structure
Framework
“How would you structure your analysis of why the call center is overwhelmed?”
- 4Drill the brainstorming
Analysis
Driver Tree Setup
“Looking at Exhibit 1, walk me through the driver tree for calculating how many agents TravelLine needs. What are the key inputs?”
- 5Drill the math
Math
Calculate Required Agents
“At Week 20, TravelLine had 300,000 customers migrated to the call center. Using the driver tree from Exhibit 1, calculate how many agents are needed at that point. Then calculate how many are needed at Week 50 with 780,000 customers. Compare both to the current 45 agents.”
- 6Drill the brainstorming
Analysis
Gap Analysis and Constraints
“The call center has 45 agents but needs 182. Hiring takes 8 weeks per agent and the budget is limited. What are the key constraints, and what categories of solutions should TravelLine consider?”
- 7Drill the math
Math
Quantify Solution Impact
“If self-service tools deflect 25% of calls and better CRM tools increase each agent's capacity by 10%, how many agents would TravelLine need at Week 50? How many do they still need to hire?”
- 8Drill the synthesis
Synthesis
Final Recommendation
“The CEO just walked in. Present your diagnosis and recommendations.”
The worked path
The numbers that decide it
- The supplied driver tree yields 205.92 weekly calls per agent; confirm whether its 55% productive-time factor includes wrap-up.
- Week 20 demand requires about 70.04 equivalents and Week 50 about 182.11, compared with 45 agents currently, not a verified historical roster.
- Whole-agent workload ceilings are 71 and 183 before any additional service-level planning.
- A 30% seasonal peak implies 236.74 equivalents before improvements, or 237 whole agents.
- The stipulated 25% deflection and 10% capacity increase imply 124.17 equivalents at baseline, approximately 79 additional or 80 under a strict whole-agent ceiling.
- The same improvement assumptions imply 161.42 equivalents at peak, or 162 whole agents.
- Deflection, tool effectiveness and implementation speed are hypotheses; average workload calculations do not guarantee a wait-time target.
Final synthesis
Use the stipulated driver tree consistently: calls per week = customers x 2.5 / 52; capacity per agent = 40 x 0.78 x 0.55 x 3,600 / 300 = 205.92. Clarify whether the on-calls ratio includes wrap-up before operational use. At Week 20, demand is 14,423.08 calls and need is 70.04 equivalents; at Week 50, 37,500 calls require 182.11. Current staffing is 45; historical staffing is not supplied. At the 30% seasonal peak need is 236.74. These are average workloads, with strict whole-agent ceilings of 71, 183 and 237, not guaranteed service-level plans. Under the question's 25% deflection and 10% capacity increase, 28,125 / 226.512 = 124.17 equivalents, approximately 124 and 79 additions; strict whole staffing is 125 and 80 additions. Peak need after the same improvements is 161.42, or 162 whole agents. Start hiring plans alongside measured demand-reduction and workflow pilots and assess qualified overflow. Do not assume tool speed, deflection success or that reducing handling time by 10% is identical to increasing capacity by 10%.
BCGOperations · hard
TravelLine Call Center Optimization
Travel / Customer Service
Why this case
This travel customer-service case tests capacity planning after centralization, a trap generic operations cases often miss. The 20-plus-minute waits are not just a staffing problem: demand rises with customer migration, call time includes a 90-second wrap-up, and summer adds 30% volume. The industry-specific issue is protecting service quality while scaling a seasonal, distributed customer base.
FAQ
- How do I calculate required agents in this call center case?
- Estimate weekly calls from customers, calls per customer, and 52 weeks, then divide demand by weekly calls handled per agent. At Week 50, 780,000 customers create 37,500 weekly calls. With about 206 calls per agent per week, TravelLine needs approximately 182 agents before seasonal or operational improvements.
- Why is hiring alone not the best answer?
- Hiring addresses capacity but leaves demand and productivity unchanged. TravelLine can deflect 20% to 30% of simple calls through IVR, chatbots, and FAQs, while better tools can reduce wrap-up time and raise calls per agent by about 10%. Those levers reduce the number of agents needed and improve response times.
- What are the three operational levers in this case?
- Reduce call volume through self-service, reduce settlement time through better CRM tools and response templates, and increase capacity through hiring, part-time peak coverage, or cross-training. A strong answer ranks the levers by speed and impact, then combines them rather than presenting staffing as the only fix.
- What should the recommendation say at Week 50?
- State that TravelLine needs about 182 baseline agents but has 45, creating a 137-agent gap. Prioritize self-service, reduce wrap-up time, and hire in waves because training takes 8 weeks. With 25% call deflection and 10% productivity improvement, the requirement falls to about 124 agents, before the 30% summer peak.
BCGOperations · hard
TravelLine Call Center Optimization
Travel / Customer Service