SLA Breach Risk Calculator

Service level (SL) is one of the most fundamental contact center KPIs — it measures the percentage of calls answered within a defined threshold time. The most common standard is "80% of calls answered within 20 seconds," often written as 80/20. Service level is measured in real time and reported at interval (typically 15–30 minutes), daily, and monthly. It directly reflects how long customers wait before speaking to an agent and is the primary measure of accessibility.

The 80/20 rule (80% of calls answered within 20 seconds) is the de facto industry standard for service level, originating from early telephone network research by AT&T Bell Labs. It represents a point where the incremental cost of adding more agents to improve service level begins to outweigh the customer experience benefit. However, this standard is not universal: financial services and emergency lines often target 90/10 (90% within 10 seconds), while some back-office queues accept 70/30. The "right" target depends on customer expectations, competitive context, and the cost of a missed call.

SLA breaches almost always stem from one of four causes: (1) Understaffing — insufficient agents scheduled for actual call volume. (2) High shrinkage — agents away from phones for breaks, training, meetings, or unplanned absence. (3) Volume spikes — actual call volume exceeding the forecast, often from product issues, campaigns, or seasonal events. (4) High handle time — calls taking longer than planned, increasing queue buildup. This calculator focuses on causes 1 and 2 (staffing and shrinkage) using the Erlang C model, which is the industry standard for service level prediction.

Shrinkage is the percentage of scheduled time during which agents are unavailable to take calls — including breaks, lunch, meetings, training, coaching, after-call work, and unplanned absence. A typical shrinkage rate of 30% means that only 70% of scheduled agents are actually answering phones at any given time. A center scheduling 20 agents with 30% shrinkage effectively has 14 agents available. Since service level is extremely sensitive to the ratio of available agents to traffic intensity (Erlang C), the difference between 14 and 15 effective agents can swing service level by 10–20 percentage points in high-traffic environments.

Service level (X% within Y seconds) and Average Speed of Answer (ASA) both measure call queue performance but in fundamentally different ways. Service level is a threshold measure — it counts the percentage of calls answered within a specific time. ASA is an average — the mean wait time across all calls, including those answered quickly and those that waited a long time. A queue can have a good ASA (e.g., 15 seconds average) while still failing service level (e.g., only 65% answered within 20 seconds) if call distribution is uneven. Service level is generally considered the more useful operational KPI because it reveals the experience of callers who wait the longest.

Yes — your inputs are encoded in the page URL so you can bookmark or share your calculation by copying the URL from your browser. Results are calculated entirely in your browser; no data is sent to our servers.