How to calculate your abandoned call rate

To calculate your abandoned call rate, divide the calls a person answered that no rep picked up in time by all calls a person answered, then multiply by 100. In the US, the FCC and FTC measure it per campaign over 30 days against a 3% limit. Ofcom in the UK measures it per calendar day and sets no safe rate.

What counts as an abandoned call

The definitions decide the numerator, so start there. As of October 2026:

• United States. A call is abandoned if a person answers and it is not connected to a live sales representative within 2 seconds of their completed greeting. The FCC’s rule (47 CFR 64.1200(a)(7)) and the FTC’s Telemarketing Sales Rule (16 CFR 310.4(b)(1)(iv)) use the same 2-second test. Playing the required message does not stop the call counting as abandoned.
• United Kingdom. Ofcom’s persistent misuse policy calls a call abandoned when the person hears a message instead of an agent, or would have heard one but hung up first. A call where the person hears nothing at all is a silent call. Ofcom keeps silent calls out of the rate but treats them as its highest priority.

The arithmetic is the same in both places: abandoned calls divided by all calls a person answered, which is abandoned calls plus calls that reached a rep. What differs is the period, the unit and whether any rate is safe.

The US calculation: per campaign, over 30 days

Abandoned call rate = abandoned calls ÷ calls answered by a person × 100

• Numerator: calls a person answered that did not reach a live sales representative within 2 seconds of the greeting.
• Denominator: all telemarketing calls answered by a person. The FTC’s guide to the Telemarketing Sales Rule says calls answered by machine, calls not answered at all and calls to non-working numbers do not count.
• Period: one calling campaign, measured over 30 days. If the campaign runs longer, the FCC’s rule calculates each 30-day period, or part of one, separately.
• Unit: each campaign on its own. The FTC’s guide says a telemarketer running campaigns at the same time cannot offset a 6% rate on one with a 0% rate on another.
• Limit and records: abandoned calls may not exceed 3% under the FCC’s rule, and 3% is the FTC’s safe harbor (16 CFR 310.4(b)(4)). Both require records that show compliance.

The FTC’s rule exempts most calls to businesses (16 CFR 310.6(b)(7)). Check with your counsel how each rule applies to your lists.

The UK calculation: per calendar day

Ofcom’s policy, annex 3, sets the formula for each calendar day, midnight to midnight:

Abandoned call rate = abandoned calls (x) ÷ (abandoned calls (x) + live calls (y)) × 100

• Live call (y): answered by a person and connected to a live agent.
• Excluded: silent calls, unconnected calls, and calls to answering machines, whether the detection system hung up on them, wrongly passed them to an agent, or abandoned them.
• Unit: per campaign or per call center, and Ofcom may also add up all of an organization’s campaigns.
• When you cannot tell people from machines: Ofcom estimates how many abandoned calls reached machines, using the share of agent-connected calls that turned out to be machines.

Ofcom uses the rate to decide which cases to pursue, not as a permitted level. The current policy no longer refers to a specific rate.

Worked examples

A US campaign over 30 days. All numbers are made up to show the math.

Outcome	Calls	In the calculation?
Dials	12,000	No
No answer, busy or invalid number	9,650	No
Answered by a machine	900	No
Answered by a person	1,450	Yes, the denominator
Of those, not connected to a rep within 2 seconds	40	Yes, the numerator

40 ÷ 1,450 × 100 = 2.76%, under the 3% limit. Two common mistakes make the rate look better than it is. Including machine answers gives 40 ÷ 2,350 = 1.70%. Dividing by all dials gives 40 ÷ 12,000 = 0.33%.

Ofcom’s own example, for one calendar day. A call center makes 1,000 calls: 392 live calls, 8 calls abandoned to a person or a machine, 200 unconnected calls and 400 machine calls passed to agents. Machines were 400 ÷ 792 = 50.5% of agent-connected calls, so Ofcom assumes 50.5% of the 8 abandoned calls (4.04) reached machines. That leaves 3.96 abandoned to people, and 3.96 ÷ (3.96 + 392) × 100 = 1.00%.

One bad day inside a good month. Suppose day 12 of the US campaign above had 65 calls answered by a person and 5 abandoned. That day’s rate is 5 ÷ 65 × 100 = 7.7%, while the 30-day rate stays at 2.76%. A daily view, which is how Ofcom calculates, shows the problem at once.

What pushes the rate up with parallel dialing

A parallel dialer rings 2 or more numbers for one rep, so two people can answer at the same moment. The rate rises when:

1. Lines outnumber free reps. Each extra line per rep raises the chance of a second answer with no one to take it.
2. Lists improve. Better numbers and calls at the right local time raise the answer rate, and with it the chance of 2 answers at once.
3. Teammates are busy. A second answer can go to a free teammate only if one is free. Small teams and long wrap-ups leave fewer free reps.
4. Detection takes too long. If answering machine detection holds a call past 2 seconds after the greeting, the US definition counts it as abandoned even when a rep is free. See how answering machine detection works.
5. The sample is small. With 20 answered calls, one abandoned call is a 5% rate. Judge early numbers with care, but keep counting them.

For a summary of the rules and how lines per rep work, see what a parallel dialer is.

Steps to track it every week

1. Log 3 facts for every answered call: a person or a machine, the seconds from greeting to rep, and whether a message played.
2. Calculate per campaign, daily and over the last 30 days.
3. Never pool campaigns to bring a high one down.
4. Set an internal trigger below the limit, and cut lines per rep when the daily rate reaches it.
5. Add reps to busy sessions instead of adding lines.
6. Review answering machine detection for delay and for people it marked as machines.
7. Keep the records. The FCC’s rule requires them, the FTC’s safe harbor depends on them, and Ofcom expects them kept for at least 6 months.

The same rate belongs in your weekly review next to connect and conversation rates. See SDR call metrics, and for single-line dialing, what a power dialer is.

How DialSavvy handles this

DialSavvy counts a call a person answers with no rep free as abandoned, and measures the rate across each parallel session. The abandonment governor drops the session to 2 lines from 2.5% abandoned calls and to 1 line from 2.8%. Compliance checks run before every call, from versioned rule packs by country. An evidence record for every attempt is coming. See DialSavvy compliance.

Before your next parallel session, add a daily abandoned-call column to your campaign report, and agree with your team on the rate at which lines come down.