Erlang Call Center Calculator
Calculate optimal staffing levels for your call center using the Erlang C formula. Input your call volume, average handling time, and service level goals to determine the required number of agents.
Calculation Results
Comprehensive Guide to Erlang Call Center Calculator (Excel Implementation)
The Erlang C formula is the mathematical foundation for call center staffing calculations, helping managers determine the optimal number of agents needed to meet service level targets while balancing operational costs. This guide explores the Erlang C formula, its Excel implementation, and practical applications for call center workforce management.
Understanding the Erlang C Formula
The Erlang C formula was developed by Danish mathematician A.K. Erlang to model telephone traffic. It calculates the probability that a call will be delayed (wait in queue) given:
- Call arrival rate (λ – lambda)
- Average handling time (AHT)
- Number of available agents (N)
The formula is:
P(W > 0) = (AN/N!) / [ (AN/N!) + (1 – A/N) * Σ(Ak/k!) from k=0 to N-1 ]
Where:
- A = Traffic intensity (λ × AHT)
- N = Number of agents
- P(W > 0) = Probability of waiting
Key Metrics in Call Center Staffing
| Metric | Definition | Industry Benchmark |
|---|---|---|
| Service Level | Percentage of calls answered within target time | 80/20 (80% in 20 seconds) |
| Average Speed of Answer (ASA) | Average time callers wait in queue | < 28 seconds |
| Agent Occupancy | Percentage of time agents spend on calls | 85-90% |
| Shrinkage | Time agents are unavailable for calls | 30-35% |
| Average Handling Time (AHT) | Average call duration + after-call work | Varies by industry (3-8 minutes) |
Implementing Erlang C in Excel
Creating an Erlang calculator in Excel requires several steps:
-
Input Parameters:
- Calls per hour (λ)
- Average handling time (seconds)
- Target service level (%)
- Target answer time (seconds)
- Shrinkage factor (%)
-
Calculate Traffic Intensity (A):
A = (Calls per hour × AHT in hours) / 3600
-
Erlang C Calculation:
Use Excel’s iterative functions or VBA to calculate the probability of waiting. The formula requires summing a series from 0 to N-1.
-
Determine Required Agents:
Increase N until the calculated service level meets or exceeds the target.
-
Adjust for Shrinkage:
Total staff = Required agents / (1 – shrinkage)
Excel Functions for Erlang Calculations
While Excel doesn’t have a built-in Erlang function, you can implement it using these approaches:
1. Using Excel Formulas (Approximation)
For small call centers (N < 50), you can use Excel’s factorial and summation functions:
= (POWER(A,N)/FACT(N)) / (POWER(A,N)/FACT(N) + (1-A/N)*SUMIF(ROW(INDIRECT("1:"&N-1)),"<="&N-1,POWER(A,ROW(INDIRECT("1:"&N-1)))/FACT(ROW(INDIRECT("1:"&N-1)))))
2. Using VBA for Precise Calculations
For larger call centers, a VBA function provides better accuracy:
Function ErlangC(A As Double, N As Integer) As Double
Dim k As Integer
Dim sum1 As Double, sum2 As Double
sum1 = 0
For k = 0 To N - 1
sum1 = sum1 + (A ^ k) / Application.WorksheetFunction.Fact(k)
Next k
sum2 = (A ^ N) / Application.WorksheetFunction.Fact(N)
ErlangC = sum2 / (sum2 + (1 - A / N) * sum1)
End Function
Practical Applications of Erlang Calculations
The Erlang C formula has several practical applications in call center management:
- Staffing Optimization: Determine the minimum number of agents needed to meet service level agreements (SLAs) without overstaffing.
- Budget Planning: Forecast labor costs by calculating staffing requirements for different call volume scenarios.
- Performance Benchmarking: Compare actual performance against theoretical models to identify operational inefficiencies.
- Seasonal Planning: Adjust staffing levels for predicted call volume fluctuations during holidays or promotional periods.
- Technology Investment: Justify investments in automation or self-service options by quantifying their impact on staffing requirements.
Common Mistakes in Erlang Calculations
| Mistake | Impact | Solution |
|---|---|---|
| Ignoring shrinkage | Underestimates total staff needed by 20-30% | Always include shrinkage factor (typically 30-35%) |
| Using average call volume | Doesn't account for peak periods, leading to poor service | Calculate for peak hours, not daily averages |
| Incorrect AHT measurement | Over/underestimates traffic intensity | Include talk time + hold time + after-call work |
| Not validating with real data | Theoretical models may not match actual performance | Compare Erlang predictions with historical data |
| Assuming perfect agent availability | Overestimates capacity | Account for breaks, training, and meetings |
Advanced Erlang Applications
Beyond basic staffing calculations, Erlang models can be extended for more sophisticated applications:
1. Multi-Skill Staffing
For call centers with specialized agent groups, modified Erlang models can calculate staffing across multiple skill sets, considering:
- Call routing rules
- Agent skill proficiencies
- Cross-training opportunities
2. Blended Environments
Call centers handling both inbound and outbound calls require blended Erlang models that account for:
- Outbound call connectivity rates
- Agent availability for inbound calls
- Priority rules between call types
3. Multi-Channel Contact Centers
Modern contact centers handle emails, chats, and social media alongside calls. Extended Erlang models can incorporate:
- Different handling times by channel
- Channel-specific service level targets
- Agent multi-tasking capabilities
Erlang vs. Other Workforce Management Models
While Erlang C is the most common model for call centers, other approaches exist:
| Model | Best For | Advantages | Limitations |
|---|---|---|---|
| Erlang C | Call centers with queues | Accurate for predictable call patterns | Assumes infinite calling population |
| Erlang B | Systems with blocked calls cleared | Simpler calculation | No queueing - calls are lost if no agents |
| Simulation Models | Complex, variable environments | Handles non-Poisson arrival patterns | Computationally intensive |
| Machine Learning | Centers with historical data | Adapts to changing patterns | Requires large datasets |
Implementing Erlang in Modern Call Centers
While Excel implementations are common for basic calculations, enterprise call centers typically use specialized workforce management (WFM) software that incorporates Erlang principles with additional features:
- Real-time Adherence: Monitors agent compliance with schedules and adjusts intraday forecasts.
- Multi-channel Forecasting: Predicts volume across all contact channels (voice, email, chat, social).
- Scenario Planning: Models the impact of different staffing strategies on service levels and costs.
- Automated Scheduling: Generates optimized schedules that balance business needs with agent preferences.
- Performance Analytics: Provides insights into schedule efficiency and service quality.
Popular WFM solutions that utilize Erlang principles include:
- NICE WFM (formerly IEX)
- Genesys WFM
- Aspect WFM
- Verint Monet WFM
- Calabrio WFM
The Future of Call Center Staffing Models
As call centers evolve, staffing models are incorporating new technologies and approaches:
- AI-Powered Forecasting: Machine learning algorithms analyze historical data to predict call volumes with greater accuracy than traditional time-series models.
- Real-time Optimization: Systems dynamically adjust staffing levels based on actual call patterns rather than fixed schedules.
- Behavioral Analytics: Models incorporate agent performance data to optimize skill-based routing and training programs.
- Omnichannel Integration: Unified models handle all customer interaction channels with channel-specific service level targets.
- Predictive Engagement: Systems anticipate customer needs and proactively offer assistance, reducing inbound call volume.