Service Level Formula: How to Calculate It, and What 80/20 Really Costs You

Service level is the percentage of contacts answered within a set number of seconds. The standard formula is:

Service level % = (Contacts answered within the threshold ÷ Contacts offered) × 100

If 136 of 170 offered calls are answered within 20 seconds, service level is 136 ÷ 170 × 100 = 80%. That is where the familiar "80/20" target comes from: 80% of calls answered within 20 seconds.

The formula looks simple, but two details change the answer and both are easy to miss: how you treat abandoned calls, and what your reporting screen is actually showing. This guide covers both, then shows what a target such as 80/20, 70/30 or 90/20 does to your headcount, using a worked example you can reproduce with our free Service Level Calculator.

The service level formula, step by step

1. Pick the threshold. The common phone standard is 20 seconds. Some operations use 30 seconds, and chat and email use much longer thresholds.
2. Count contacts answered within the threshold in the reporting period (a half hour interval is the usual unit for planning).
3. Count contacts offered. This is every contact that entered the queue, answered or not.
4. Divide and multiply by 100.

Always calculate by interval first, then roll up. A day-level number hides the bad intervals, which is where the customer pain actually sits.

How abandoned calls change the number

Callers who hang up before an agent answers are the main source of disagreement. Call Centre Helper describes three ways a system can treat them, based on Cisco's methods:

One refinement is widely accepted: very short abandons, such as calls dropped within 5 seconds, can be removed from both the numerator and the denominator, since many are wrong numbers or accidental dials. Many ACD systems can be set up this way. Whatever you choose, write it down and apply it the same way every time, because a change in definition can move the reported number without any change in real performance.

The 80% that is really 64%

Some wallboards and reports show a percentage of intervals or calls that "met target", not the percentage of calls answered within 20 seconds. Call Centre Helper gives a warning example: a displayed 80% can mean only 64% of calls were answered within 20 seconds (80% of 80%). If your dashboard number and your own calculation disagree, check the definition behind the tile before you check the staffing.

What 80/20 costs: a worked example

Here is a standard Erlang C example. It is illustrative, uses no shrinkage and assumes callers never abandon, which makes the model slightly pessimistic about queues.

• 300 calls arrive in a 30 minute interval.
• Average handle time is 300 seconds (5 minutes).
• Workload is 300 × 300 ÷ 1,800 = 50 Erlangs (the number of agents fully busy on average).

How service level responds as you add agents:

And what different targets need at this volume:

Three things stand out:

• One agent can decide the target. At 56 agents the interval reaches about 79% and misses 80/20. At 57 it clears it. Rounding down on a headcount plan is a service level miss.
• Moving from 80/20 to 70/30 saves only 2 agents here (about 3.5%), but pushes occupancy to about 91%, which is a heavy load on people. See The Occupancy and Retention Link.
• Going from 80/20 to 90/20 costs 2 more agents (about 3.5%). For many operations that is a modest price for a visibly better customer wait.

These are model results, not benchmarks. Real centres add shrinkage on top, and real callers abandon, so use the numbers to see the shape of the trade-off, not as a staffing recommendation. You can run your own volume and handle time in our Erlang C Calculator, and for more on the curve see Why Adding Agents Stops Helping.

Is 80/20 the right target?

It is the most quoted phone standard, but common is not the same as correct. The injixo article republished by Call Centre Helper makes a useful point: the target drives staffing cost, so it should be tested against customer outcomes rather than kept because it has always been there. Practical ways to do that:

• Compare CSAT against actual wait times at thresholds such as 30, 60 and 90 seconds. A clear link supports a strict target. A weak link suggests the target may be costing more than it returns.
• Pilot before you change. If you test 70/30 instead of 80/20, watch your key business metrics and only keep the change if nothing gets worse.
• Match the metric to the goal. Sales and revenue teams may care more about the abandon rate, since each abandoned call can be a lost sale.
• Protect the tail. Standard service level ignores how long the unlucky callers waited. A secondary target, for example 99% answered within 300 seconds, caps the long waits.

Abandon rate benchmarks often quoted for phone are in the range of 2% to 5%. Treat them as a starting point, not a rule, because they vary a lot by industry and call type.

Common service level mistakes

1. Planning on a daily average. The day looks fine while three intervals collapse. Plan and report by interval. See Intraday Management: How to Recover Service Level in Real Time.
2. Under-forecasting. A volume miss in the wrong direction hurts far more than over-forecasting. See The Forecast Error Asymmetry.
3. Changing the definition silently. Moving how abandons are counted changes the number without changing the service.
4. Chasing the target and ignoring the people. A target that keeps occupancy above about 85% for long periods has a retention cost that does not show up on the service level report.
5. Forgetting shrinkage. The agents you need on the phones are not the agents you schedule. Gross up properly. See our guide on how to calculate shrinkage in a BPO.

Quick checklist

• Write down your formula, including how abandons and short abandons are treated.
• Check what each dashboard tile really measures.
• Calculate by half hour interval before you roll up to day or week.
• Test any target change against customer outcomes before you commit.
• Keep a secondary target for long waits.
• Round headcount up, not down.

Want a quick view of where your planning is weakest? Try the free WFM Health Check.

FAQ

What is the formula for service level in a call centre? Service level % = (Contacts answered within the threshold ÷ Contacts offered) × 100.

What does 80/20 mean? 80% of calls answered within 20 seconds. It is the most common phone target.

Are abandoned calls included in service level? In the recommended method, yes. They stay in the contacts offered but are not counted as answered in time. Very short abandons, such as under 5 seconds, are often excluded from both sides.

What is the difference between service level and ASA? Service level is the share of contacts answered within a threshold. Average speed of answer (ASA) is the average wait of answered contacts. A good ASA can hide a long tail of slow answers, which is why service level is usually the main target.

Is 80/20 still the right target? Not necessarily. It is common, but the right target depends on your customers, channels and cost. Compare CSAT with wait times, pilot any change, and consider a secondary target for long waits.

How many agents do I need for an 80/20 service level? It depends on your volume and handle time, and you must add shrinkage on top. Use an Erlang C calculation to find the agents on the phones, then gross up for shrinkage.

Sources: Call Centre Helper: How to Calculate Contact Centre Service Level and Call Centre Helper: How to Set the Right Service Level Goal. The worked example uses the standard Erlang C formula calculated by the WFM Helper team.