Grade of Service - GOS: The Complete Guide to Contact Centre Service Levels
Grade of Service (GOS) — also widely known as service levels — is one of the oldest, most commonly used metrics in contact centre management. In simple terms, it measures the percentage of calls (or contacts) answered within a defined time threshold. It is usually expressed as two numbers: for example, 80/30 means 80% of calls answered within 30 seconds.
Despite its age, Grade of Service remains one of the most important operational metrics in any contact centre. It has a direct relationship with staffing levels — the higher the service level target, the more agents you need. And it is the primary output of Erlang C workforce planning calculations, making it central to every resourcing decision a contact centre makes.
Note that Grade of Service is specific to voice and telephony channels. For the broader concept of service level across all contact channels — including email, chat, social media, and SMS — see the ACXPA Service Level guide.
This guide covers the GOS definition, how service level targets work globally, how to calculate the resources needed to achieve them, industry benchmarks, the risks of relying on GOS alone, and how to use the ACXPA tools to model your own scenarios.
What GOS measures
The percentage of contacts answered within a defined time threshold — the fundamental measure of contact centre accessibility and queue performance.
Why it matters
Service level targets directly determine staffing requirements. Higher targets mean more agents and more cost. Getting the target right is one of the most important decisions in contact centre management.
What this guide covers
The GOS definition, how targets vary by industry, how to calculate resources using Erlang, industry benchmarks, limitations of the metric, and practical tools to model your scenarios.
What is Grade of Service?
Grade of Service (GOS) is a contact centre metric that measures the percentage of inbound contacts (typically calls) that are answered within a specified time threshold. It is expressed as two numbers — the target percentage and the target time — such as 80/30 (80% of calls answered within 30 seconds).
Grade of Service is the contact centre industry's primary measure of how accessible and responsive a service is from the customer's perspective. A high GOS means customers rarely wait long; a low GOS means more customers are experiencing significant queue times.
GOS sits at the heart of all contact centre workforce planning. Because it directly drives the number of agents required, the Grade of Service target you set has a significant impact on staffing costs, customer experience, and operational efficiency.
In plain English
Grade of Service answers one question: of all the customers who called in a given period, what percentage had their call answered within your target time? Everything else in contact centre operations flows from this number.
Example: 80/30
The most commonly quoted GOS standard. Of every 100 calls received, 80 must be answered within 30 seconds. The remaining 20 will wait longer — but GOS alone does not tell you how much longer.
Example: 80/20
Currently the most common target in Australian contact centres (2024 data). 80% of calls answered within 20 seconds — a tighter threshold than 80/30, requiring more agents to achieve.
Grade of Service Illustrated
The diagram below illustrates an 80/20 Grade of Service target — 8 out of every 10 callers are answered within 20 seconds, while 2 are still waiting. The GOS target tells you nothing about how long those 2 callers will wait — that is one of its key limitations. The bar chart shows the most common targets in Australian contact centres based on 2024 benchmarking data.
Grade of Service vs Service Levels — What is the Difference?
In practice, Grade of Service and Service Levels are used interchangeably in most contact centre environments. Both refer to the same underlying metric — the percentage of contacts answered within a time threshold. The terminology varies by region, organisation, and industry:
Grade of Service (GOS)
The traditional technical term, more common in older contact centre literature, telecommunications, and some Australian and UK organisations. Abbreviated to GOS.
Service Level (SL)
The more widely used modern term globally — particularly in North America and in WFM software. Most contact centre platforms, benchmarking reports, and industry bodies now use "Service Level" as the primary term.
Service Level Agreement (SLA)
Commonly used in outsourced and BPO environments where the service level target is contractually defined between the client and the contact centre operator. An SLA specifies the GOS target as a contractual commitment.
💡 Related but different: Grade of Service vs Average Speed of Answer
Grade of Service and Average Speed of Answer (ASA) both measure queue performance but differently. GOS measures the percentage of calls answered within a threshold. ASA measures the average wait time across all calls. A contact centre can have a good GOS target and a poor ASA if a small proportion of callers are waiting a very long time — which is precisely why GOS alone can be misleading.
How to Set the Right Service Level Target
There is no universally correct Grade of Service target. The right target for your contact centre depends on your customers' expectations, the nature of your contacts, and what your business can commercially justify. The most important question to ask is: how critical is it for your business if customers have to wait?
Emergency services
Service levels as high as 100/5 — 100% of calls answered within 5 seconds. Every second counts and no caller can be left waiting. Cost is not the primary consideration.
Revenue-generating centres
High service level targets are commercially justified — every unanswered or long-wait call is a lost sale opportunity. Missing a caller means missing revenue.
General customer service
The most common setting. 80/20 or 80/30 are typical benchmarks. Must balance customer experience expectations against staffing budgets.
Government and public services
Where customers have no alternative provider, some public sector centres operate with significantly lower targets — occasionally as low as 70% in 30 minutes. This is not ideal but reflects budget realities in some environments.
Setting the right target — the key factors
When determining your Grade of Service target, weigh up: customer tolerance for waiting (informed by feedback and abandonment data), the commercial impact of missed calls, your competitive environment, available budget, and the nature of the contact (simple transactional vs complex service). The right target is the one that balances all of these — not an arbitrary industry number.
Industry Service Level Benchmarks
While there is no single industry standard, benchmarking data from Australian contact centre surveys reveals the most common targets in practice. The following five targets were the most frequently reported in the 2024 Australian Contact Centre Best Practice Report:
Real wait times — what customers actually experience
Self-reported service level targets tell you what contact centres are aiming for. ACXPA's Australian Call Centre Rankings reveal what customers actually encounter — real wait times measured independently via mystery shopping across multiple industry sectors including Banks, Car Insurance, Councils, Energy Providers, Internet Providers, and TAFEs.
This is one of the only publicly available sources of externally measured wait time data in Australia — and the gap between what centres target and what customers experience is often significant.
Are Australian Contact Centres Hitting Their Service Level Targets?
Setting a service level target and actually achieving it are two very different things. Data presented at ACXPA's April 2026 Call Centre Roundtable — sourced from the 2024 Australian Contact Centre Best Practice Report — revealed a striking picture of how Australian contact centres are tracking against their own GOS targets.
Source: 2024 Australian Contact Centre Best Practice Report, Smaart Recruitment. Presented at ACXPA Call Centre Roundtable, April 2026.
What this tells us
Nearly two thirds of Australian contact centres are either missing their service level targets or only just within 10% of them. This is a significant industry-wide challenge — and one that points to workforce planning, forecasting accuracy, and resourcing as areas where most contact centres have room to improve. Modelling your staffing requirements accurately using Erlang tools is one of the most direct ways to close the gap.
How to Calculate Resources to Achieve Your Grade of Service Target
The mathematical foundation of contact centre staffing is Erlang C — a formula that calculates the number of agents required to handle a given call volume at a specified service level. Erlang C takes three inputs and produces one output:
Call volume
The number of calls (or contacts) arriving in the period you are modelling — typically expressed as calls per hour. Accurate forecasting of call volume is the starting point for all GOS planning.
Average Handling Time (AHT)
The average duration of each contact including talk time, hold time, and after call work. AHT is the most impactful variable — a 30-second reduction in AHT can eliminate the need for several agents.
Target service level
Your Grade of Service target — the percentage of calls you want answered within your defined time threshold. Higher targets require more agents.
Output: agents required
Erlang C calculates the minimum number of agents needed to meet your target. This becomes your base staffing number — before shrinkage, leave, and other adjustments are applied.
Use the ACXPA Erlang C Calculator
ACXPA's Erlang Calculator makes this calculation straightforward — enter your call volume, AHT, and service level target and it calculates the agents required. You can also model the impact of changing any variable — what happens to staffing requirements if AHT drops by 20 seconds? What if call volume increases by 15%? — making it a powerful business case tool.
The Staffing Impact of Grade of Service Targets
One of the most important things to understand about service levels is that the relationship between GOS targets and staffing is not linear. Due to the mathematics of queuing theory, pushing from 80% to 90% answered in the same time threshold requires a disproportionately large increase in agents compared to the original jump from 70% to 80%.
This is why the GOS target decision is so commercially significant — and why modelling the cost implications of different targets before committing to them is critical.
Higher target = more agents, more cost
Moving from 80/30 to 90/30 requires meaningfully more agents than the same incremental improvement at lower levels. The cost curve steepens as you approach 100% — perfection is extremely expensive.
AHT reduction is highly leveraged
A 30-second reduction in average handling time can have a larger positive impact on achievable service levels than adding an agent. Fixing slow systems and inefficient processes is often the most cost-effective route to better GOS.
Shrinkage multiplies the base requirement
Erlang C calculates agents on the phone. Your actual headcount requirement must factor in shrinkage — the percentage of time agents are unavailable (training, breaks, admin, absence). Typically 25–35% in Australian contact centres.
Interval-level planning is essential
A monthly or daily GOS average hides intraday peaks and troughs. A contact centre hitting 80/30 on average may be at 60/30 during the morning peak and 95/30 in the afternoon. Service levels must be modelled and managed at the half-hour interval level.
Low volume amplifies variance
In low-volume environments, the arrival of a single extra call in an interval can cause a large GOS swing. Statistical randomness makes GOS harder to predict and manage with small teams — a factor Erlang models account for.
Channel mix affects overall GOS
Contact centres managing phone, email, chat, and digital channels simultaneously must manage GOS targets per channel — the dynamics are different for each. Phone queuing follows different patterns to email response times.
Risks and Limitations of Grade of Service as a Metric
Grade of Service is an essential metric — but it is also one of the most easily misread and misused in contact centre management. Understanding its limitations is as important as understanding what it measures.
✕ What GOS doesn't tell you
- What happened to calls that didn't meet the threshold
- How long the longest waits were
- How many callers abandoned before being answered
- Whether the calls that were answered were resolved effectively
- The quality of the interactions
- Customer satisfaction or effort levels
✓ Pair GOS with these metrics
- Abandonment Rate — how many gave up before being answered
- Average Speed of Answer (ASA) — average wait across all calls
- Average Handling Time (AHT) — duration of interactions
- CSAT — how satisfied customers were
- First Contact Resolution (FCR) — whether issues were resolved
💡 Watch your measurement interval
GOS reported as a monthly average can mask serious intraday problems. A contact centre hitting 80/30 for the month could be running at 50/30 during the morning peak and 98/30 overnight — very different experiences for customers depending on when they call. Always report GOS at the half-hour interval level for operational management, and use daily/monthly averages only for trend reporting.
Frequently Asked Questions About Grade of Service
What does 80/20 mean in a call centre?
80/20 is a Grade of Service target meaning 80% of incoming calls must be answered within 20 seconds. It is currently the most common service level target in Australian contact centres according to 2024 benchmarking data. The two numbers always represent: [percentage of calls] / [time threshold in seconds].
What is the difference between Grade of Service and Service Level?
In practice, they refer to the same metric — the percentage of calls answered within a defined time threshold. "Grade of Service" (GOS) is the older, more technical term; "Service Level" (SL) is the more commonly used modern term, particularly in North America and in most WFM software platforms. Both are expressed as two numbers (e.g. 80/30) and calculated the same way.
What is the industry standard for Grade of Service?
There is no universal industry standard. The most commonly quoted target historically was 80/30 (80% of calls answered in 30 seconds). However, 2024 Australian benchmarking data shows that 80/20 is now the most common target. The right target for any contact centre depends on customer expectations, the nature of contacts, and commercial considerations — not an arbitrary industry number.
How do you calculate staffing for a given service level?
The standard method is Erlang C — a mathematical formula that calculates the number of agents required to handle a given call volume at a specified service level. You need three inputs: call volume (calls per hour), Average Handling Time (AHT), and your target service level. ACXPA's free Erlang C Calculator automates this calculation and allows you to model multiple scenarios.
Why is Grade of Service dangerous to use alone?
GOS only tells you what percentage of calls were answered within the threshold. It tells you nothing about what happened to the calls that were not — they could have waited slightly over the threshold, or they could have waited for hours. It also says nothing about abandonment rates, customer satisfaction, or whether issues were resolved. Always use GOS alongside abandonment rate, ASA, AHT, CSAT, and FCR for a complete picture of contact centre performance.
What is a good Grade of Service for a contact centre?
The "right" GOS depends entirely on your business context. Emergency services may target 100/5; general customer service typically targets 80/20 to 80/30; some government services operate at lower targets. More important than hitting a specific industry benchmark is setting a target aligned with your customers' expectations, modelling the cost implications accurately, and consistently achieving what you commit to.
How often should Grade of Service be measured?
For operational management, GOS should be measured and reported at the half-hour interval level — this reveals intraday peaks and troughs that daily or monthly averages hide. For trend reporting and management dashboards, daily and monthly aggregates are useful. Be cautious about managing to monthly averages alone — a good monthly average can mask severely poor performance during specific intervals.
What is the relationship between Grade of Service and abandonment rate?
They are closely related but measure different things. GOS measures the percentage of calls answered within a threshold. Abandonment rate measures the percentage of callers who gave up and disconnected before being answered. When GOS deteriorates (fewer calls answered quickly), abandonment rate typically increases — customers who encounter long queues hang up. Tracking both together gives a much clearer picture of queue performance than either alone.
Where to Next
Summary: Grade of Service and Service Levels
Grade of Service is the fundamental measure of contact centre accessibility — and one of the most important metrics in workforce management. Setting the right service level target, calculating the staffing required to achieve it, and managing performance against it at the interval level are core disciplines for any contact centre leader.
What the data tells us is sobering: nearly two thirds of Australian contact centres are either not hitting their service level targets or only within 10% of them. The gap between target and reality is typically a workforce planning problem — inaccurate forecasting, incorrect Erlang modelling, or insufficient headcount to absorb shrinkage and variability.
The tools to close that gap are available. An accurate Erlang C model, interval-level planning, and a service level target genuinely aligned with customer expectations and commercial objectives are the foundations. Use the ACXPA WFM Hub, Erlang Calculator, and Call Centre Hub resources to build that foundation — and use the Call Centre Roundtables to learn from peers who are doing it well.
