Network Effectiveness Ratio (NER) is a call quality metric defined in ITU-T Recommendation E.425. It measures the percentage of call attempts that the network successfully delivered to the destination terminal — whether or not the called party answered.
NER = 100 × (Answered + User Busy + Ring No Answer + Terminal Reject) ÷ Total Calls Attempted
NER is one of the most important route quality metrics in wholesale VoIP, yet it is regularly confused with ASR (Answer-Seizure Ratio). The two measure fundamentally different things — and mixing them up leads to misdiagnosed route problems, missed SLA violations, and wasted troubleshooting effort. This article explains exactly what NER measures, how its formula works, why certain outcomes count as "effective," and how to use NER alongside ASR to accurately interpret route health.
What NER measures — and what it does not
NER answers one specific question: did the network do its job? Its job is to deliver the call to the destination terminal — the actual phone or SIP endpoint of the person being called. If the network gets the call there, NER counts that as success regardless of what happens next. The called party may be busy, may not answer, or may actively reject the call from their handset. None of that is the network's fault, so NER does not penalize those outcomes.
What NER excludes is equally important. If a call fails because of a routing error, a congested trunk (a 5xx SIP response class), a signaling failure, or any other network-level fault — those are not effective deliveries. They count against NER. This makes NER a clean indicator of infrastructure quality, separated from the noise of user behavior.
Breaking down each component of the formula
Each term in the NER numerator represents a call outcome where the network successfully reached the destination:
- Answered — the call connected and the called party picked up. The network delivered it, and the human (or machine) accepted it.
- User Busy (486 Busy Here) — the destination terminal was active on another call and returned a busy signal. The network reached the terminal; the terminal reported that the user was unavailable. Network delivery: successful.
- Ring No Answer — the call reached the destination and the phone rang, but the called party did not pick up before the call timed out. The network delivered the call; the user chose not to answer. Network delivery: successful.
- Terminal Reject — the destination device rejected the call for a terminal-side reason such as Do Not Disturb, call rejection, or a similar handset-level rule. Again, the network got the call there; the terminal made a local decision. Network delivery: successful.
The denominator is simply total calls attempted — every SIP INVITE the originating side sent, regardless of outcome. NER divides effective deliveries by total attempts and multiplies by 100 to express the result as a percentage.
Why busy and no-answer count as effective delivery
This is the part that trips people up. It seems counterintuitive: the call didn't connect, so why count it as a network success?
The logic follows from what NER is designed to measure. The network's responsibility ends at the destination terminal. A busy signal (SIP 486) means the carrier chain successfully routed the call all the way through, the destination switch responded, and the called party's line status was accurately reported back. That is precisely what a well-functioning network should do. The fact that the user was busy is information about the user's state, not a signal of network failure.
The same reasoning applies to ring no answer and terminal reject. In both cases, the network completed its delivery task. What happens at the endpoint — whether a human picks up, ignores the call, or sends it to DND — is outside the network's control and outside NER's scope.
Outcomes that do not count as effective delivery include SIP 5xx responses (server errors, service unavailable), 4xx failures that indicate routing problems, unroutable calls, and any call that never reaches a destination carrier. Those represent genuine network-layer failures.
NER vs ASR: the critical distinction
ASR — Answer-Seizure Ratio — is the simpler and more widely cited metric. It counts only answered calls as successful:
ASR = 100 × Answered ÷ Total Calls Attempted
The difference in what each metric counts as success creates a significant divergence in what low values mean:
| Metric | Counts as success | What a low value indicates |
|---|---|---|
| NER | Answered + Busy + Ring No Answer + Terminal Reject | Network-level delivery failure: routing errors, congestion, SIP signaling faults |
| ASR | Answered only | Low answer rate: list quality issues, bad calling time, unanswered campaigns, or network failure |
ASR conflates network quality with user behavior, which makes it harder to diagnose problems. NER separates them. A carrier SLA based on NER holds the network accountable for what the network actually controls.
For a deeper look at how ACD (Average Call Duration) fits alongside these metrics in route quality monitoring, that article covers the third major dimension of route health.
NER thresholds: what the numbers mean
NER is most useful with reference points. Industry practice for high-quality wholesale routes generally sets these expectations:
- 90% or above — considered good NER for high-quality routes. The network is reliably delivering calls to destination terminals. Any failures that do occur are within acceptable range for real-world routing complexity.
- 85%–90% — a gray zone worth monitoring. May reflect route congestion, destination-side issues, or a route that is performing marginally. Worth investigating before it degrades further.
- Below 85% — indicates network-level delivery problems. At this level, the route is failing to deliver roughly one in seven calls or more before they even reach the destination. That is a routing, signaling, or congestion problem — not a user behavior issue.
These thresholds apply to route-level NER aggregated over meaningful call volume. Short measurement windows on low call counts can produce misleading NER figures due to statistical noise.
How to interpret NER and ASR together
Reading NER and ASR in combination is where the real diagnostic value emerges. There are two primary patterns to understand:
High NER + low ASR
Example: NER = 95%, ASR = 25%. The network is delivering calls effectively — 95% of attempts reach the destination terminal. But only 25% of total attempts result in an answer. The gap between NER and ASR is accounted for by busy signals, no-answers, and terminal rejects. This is normal for call center outbound traffic on a healthy route. The low ASR reflects list quality, campaign targeting, or calling time strategy — not a network problem. Improving ASR here requires changes to the dialing program, not the route.
For call center traffic, ASR of 30–60% is typical depending on campaign type and list quality. A well-functioning CC route with good list hygiene can sit anywhere in that range without indicating route failure.
Low NER + any ASR
Example: NER = 70%, ASR = 20%. The network is failing to deliver 30% of calls at the network layer. Some of those failures may generate answered calls by chance, but the route is fundamentally broken. This is a carrier or routing problem — congestion, misconfigured routes, or signaling failures. The fix is a route change, not a dialing strategy adjustment.
Low NER means calls are disappearing before they reach the destination. No amount of list cleaning or campaign optimization will fix a NER problem, because the issue is upstream of the destination entirely.
Understanding how CC routes work from origin to destination provides useful context for where in the call path NER failures originate.
NER in carrier SLAs
Because NER isolates network quality from user behavior, it is the preferred metric for carrier-to-carrier SLAs in the wholesale voice market. An SLA written around ASR would be unfair to the carrier — they cannot control whether called parties answer. An SLA written around NER holds the carrier accountable only for what they actually control: delivering the call to the destination terminal.
Typical SLA structures in wholesale voice set a minimum NER floor (often 90% or higher for premium routes) with measurement intervals defined by time period and minimum call volume. When NER drops below the contracted threshold, the SLA triggers remediation obligations — route switching, rate adjustments, or service credits depending on the contract.
When evaluating a route provider, ask specifically how NER is measured (time window, call volume minimum, per-destination or aggregate), what the SLA threshold is, and what remediation looks like when it is breached. Those details matter more than the headline NER figure. The CC routes provider evaluation guide covers what to ask and how to assess the answers.
Practical example: reading a route quality report
Suppose a weekly route quality summary shows:
- Total attempts: 500,000
- Answered: 130,000
- User Busy: 40,000
- Ring No Answer: 295,000
- Terminal Reject: 10,000
- Network failures (5xx, routing errors): 25,000
NER = 100 × (130,000 + 40,000 + 295,000 + 10,000) ÷ 500,000 = 100 × 475,000 ÷ 500,000 = 95%
ASR = 100 × 130,000 ÷ 500,000 = 26%
Interpretation: the route is healthy (NER 95%). The 26% ASR reflects normal call center outbound dynamics — most attempts ring without answer, which is typical for predictive dialer campaigns. The 25,000 network failures (5%) represent the route's irreducible failure floor; anything above 10% failures would signal a problem requiring escalation.
Frequently asked questions
For more on the metrics that define CC route quality, see what is ACD in VoIP (average call duration and how it interacts with route monitoring) and how CC routes work (the full call path). For a framework on evaluating providers based on these metrics, see how to choose a CC routes provider. Related: False Answer Supervision (FAS) distorts ASR and can affect NER interpretation — understanding both metrics together is essential for route vetting. For a step-by-step route quality testing process, see how to test wholesale VoIP route quality.
