How Many Calls Can an Agent Make per Hour?
Work out dials, connects and talk time per agent hour for preview, power, progressive and predictive dialing, with the formula and abandon-rate limits.
Resposta rápida
An agent's calls per hour is a formula, not a benchmark: 3,600 seconds divided by the average seconds each dial attempt consumes. That time depends on your connect rate, talk time, wrap-up and ringing time. Dialing modes change the waiting, not the conversations. Predictive dialing removes the most waiting but is capped by abandoned-call limits, such as the US safe harbor of 3 percent of answered calls.
Search for this question and you will find confident single numbers. They rarely say which list, which connect rate or which talk time sits behind them, so they cannot be reused on your campaign. This guide gives you the formula instead, a worked example you can copy into a spreadsheet, and the legal ceiling that stops a predictive dialer from simply dialing faster.
Every number in the worked examples below is an illustrative example input. None of them is a DialerBee result, an industry average or a promise. Use your own campaign reports.
How many calls can an agent make in an hour?
As many as the hour allows after you subtract everything that is not dialing. Start with 3,600 seconds. Each attempt takes some of them: the dial and ring time, the conversation if someone answers, the wrap-up after it, and any wait for the next call. Divide 3,600 by the average cost of one attempt and you have attempts per hour.
Three things follow from that:
- Connect rate matters more than speed. If most attempts go unanswered, the agent's hour is spent on ringing, whatever the mode.
- Talk time is the product. A dialer that raises attempts but lowers the minutes agents spend talking to real people has not helped.
- Modes differ in waiting, not in conversations. A faster mode removes idle seconds. It cannot make a conversation shorter and should not.
The formula: attempts, connect rate, talk time, wrap-up and wait time
Define five inputs for one campaign, ideally from the last few weeks of your own data:
- c, the connect rate: the share of attempts that reach a live person. See O que significa taxa de contato? and the narrower right-party contact.
- u, the seconds an agent spends on an attempt that is not answered (ringing, busy, voicemail).
- t, the average talk time of a connected call, in seconds.
- w, the average wrap-up time after a connected call, in seconds.
- p, the per-attempt overhead the agent spends before dialing (reading the record, clicking dial). It is near zero with auto-dial modes.
The average seconds per attempt for a mode where the agent waits through every call is:
seconds per attempt = p + (1 - c) x u + c x (t + w)
attempts per hour = 3,600 / seconds per attempt
connects per hour = attempts per hour x c
talk minutes = connects per hour x t / 60 Worked example (illustrative inputs)
Take these example inputs: connect rate 20 percent, 25 seconds on each unanswered attempt, 180 seconds of talk time, 30 seconds of wrap-up, and 10 seconds of manual overhead per attempt.
- Unanswered share: 0.8 x 25 = 20 seconds.
- Answered share: 0.2 x (180 + 30) = 42 seconds.
- Overhead: 10 seconds.
- Total: 72 seconds per attempt, so 3,600 / 72 = 50 attempts an hour, 10 connects and 30 talk minutes.
Change one input and watch the result. At a 10 percent connect rate with everything else equal, the unanswered share rises to 22.5 seconds and the answered share falls to 21 seconds, so an attempt costs 53.5 seconds (about 67 attempts), but only about 6.7 connects. More attempts, fewer conversations. That is why attempts per hour is a poor target on its own.
Manual and preview dialing: where the minutes go
In manual dialing the agent finds the record, types or clicks the number and waits. The overhead p in the formula is the cost. In preview dialing the agent sees the contact record first and decides when to dial. That adds preparation time on purpose, which suits high-value or sensitive conversations where the call matters more than the count.
With the example inputs, manual dialing gives 50 attempts and 30 talk minutes per hour. Preview dialing will usually show fewer attempts than that, because the preview window is time you chose to spend. Judge it on outcomes per call, not dials.
Power and progressive dialing: one line per free agent
A discador automático dials the next contact immediately after the agent finishes wrap-up, a 1:1 ratio between agents and calls. A discador progressivo also dials one contact for each available agent. In both, the agent still waits through ringing and unanswered calls, but the manual overhead disappears.
Set p to zero in the example: 62 seconds per attempt, so 3,600 / 62 is about 58 attempts, 11.6 connects and 34.8 talk minutes an hour. That is roughly 8 more attempts and almost 5 more talk minutes than manual dialing, from removing clicks alone.
Because there is never more than one line per free agent, pacing does not create abandoned calls, which makes these the modes to start with when abandon limits worry you. Our full comparison of the three modes covers the differences between power and progressive.
Predictive dialing: more lines than agents, and the abandon-rate ceiling
A discador preditivo places more calls than there are free agents, predicting that many will go unanswered, and connects each answered call to whichever agent frees up. The agent no longer waits through unanswered attempts, only for the short gap before the next connected call.
Using the example inputs and an assumed 5-second average gap between wrap-up and the next connected call: each connected cycle costs 180 + 30 + 5 = 215 seconds, so about 16.7 connects an hour, roughly 50 talk minutes, and at a 20 percent connect rate the dialer places about 84 attempts per agent-hour on the agent's behalf. Compare that with 30 talk minutes manually: the gain is entirely removed waiting.
The catch is that sometimes a prediction is wrong and a person answers when every agent is busy. That is an abandoned call, and the law limits how often it may happen.
The United States: Telemarketing Sales Rule
Under 16 CFR 310.4(b)(1)(iv), an outbound telemarketing call is abandoned if a person answers it and the telemarketer does not connect the call to a sales representative within two seconds of the person's completed greeting. The safe harbor in 310.4(b)(4) shields a seller from liability for abandonment if all of these hold:
| Safe-harbor condition | What the rule says |
|---|---|
| Abandonment rate | Technology that ensures no more than 3 percent of calls answered by a person are abandoned, measured over a single campaign if it runs under 30 days, or separately over each successive 30-day period |
| Ring time | Each unanswered call rings at least 15 seconds or four rings |
| Recorded message | When no representative is available within two seconds, a recorded message promptly states the name and telephone number of the seller |
| Records | Records kept that show compliance with the three conditions above |
(Source: 16 CFR 310.4, Cornell Legal Information Institute, checked 2026-11-19.) The FTC's business guide to the Telemarketing Sales Rule is the plain-English companion.
What 3 percent means in practice
An illustrative example: if a campaign has 1,000 calls answered by a person in a 30-day period, the safe harbor allows at most 30 of them to be abandoned. Abandon rate is calculated against answered calls, not total dials, so a low connect rate does not loosen the limit. Try your own figures in the estimador de taxa de abandono, and read what abandon rate is for the definition.
Rules differ by country. The US rule above applies to telemarketing calls it covers, and other regulators set their own ceilings and measurement periods. Check the rule for the country you call. For pacing settings that keep a campaign inside a ceiling, see the guia de ajuste de discador preditivo.
Which number should you track instead?
Track the outputs that pay for the agent's hour:
- Talk minutes per agent hour. The share of paid time spent in live conversation.
- Right-party contacts per hour. Conversations with the person you meant to reach, not any answer.
- Outcomes per hour. Appointments, payments, qualified leads, whatever the campaign is for.
- Abandon rate against your ceiling. A measure of risk, not productivity.
Dials per hour is a diagnostic, useful for spotting idle time, not a target. If you want to raise contacts rather than dials, Como melhorar as taxas de contato de saída covers lists, timing and caller presentation, and the Calculadora de ROI shows what a change in agent productivity means in money.
Which mode fits which campaign?
| Modo | Agent waits through unanswered calls | Abandon risk from pacing | Suits | Plano DialerBee |
|---|---|---|---|---|
| Manual | Yes, plus dialing overhead | Nenhum | Low volume, one-off calls | Qualquer |
| Pré-visualização | Yes, plus preview time | Nenhum | High-value or sensitive accounts | Starter and up |
| Poder | Sim | None (1:1) | Small teams, steady pace | Starter and up |
| Progressivo | Sim | None (1:1) | Regulated campaigns | Professional and up |
| Preditivo | No | Yes, must be limited | Larger teams, low connect rates | Professional and up |
Nosso predictive dialer page says it works best with teams of 20 or more agents, where there is enough data for accurate predictions. Starter ($69 per seat per month) includes preview and power dialing; Professional ($129 per seat per month) includes all four modes, with annual billing saving 20 percent. Carrier minutes are billed by your own carrier. Details are on the página de preços.
Perguntas frequentes
How many calls can an agent make per hour?
There is no single true number, because it depends on your connect rate, how long conversations last, how long wrap-up takes and whether the agent waits through ringing. The honest method is a formula: 3,600 seconds divided by the average seconds one dial attempt consumes. In our illustrative example (every input is an example, not a benchmark) that gives about 50 attempts an hour with manual dialing and about 84 with predictive pacing. Replace the inputs with your own campaign data.
How many dials per hour can you get with a predictive dialer?
More than with manual dialing, because the agent no longer waits through unanswered calls, but the figure depends on your connect rate and talk time. In the worked example on this page, the same agent who makes 50 attempts an hour manually is covered by about 84 attempts an hour when the dialer places calls ahead of the agent. The legal ceiling on abandoned calls limits how aggressive the pacing can be.
What is the legal abandon rate limit for a predictive dialer?
In the United States, the Telemarketing Sales Rule treats a call as abandoned if a person answers and no sales representative is connected within two seconds of the person's completed greeting. The safe harbor requires technology that keeps abandonment to no more than 3 percent of calls answered by a person, measured over a single campaign if shorter than 30 days, or over each successive 30-day period. Other countries set their own limits, so check your local regulator.
Is a power dialer faster than manual dialing?
Usually yes, because the next call starts automatically after wrap-up, so the agent skips the dialing and lookup steps. In the worked example it adds roughly 8 attempts an hour over manual dialing (58 against 50). The agent still waits through ringing and unanswered calls, so a power dialer gains less than a predictive one on low-connect lists.
How many calls should an outbound agent make per day?
Set the target from talk minutes and right-party contacts, not from dials. Multiply your hourly model by the real productive hours in a shift, after breaks, coaching, meetings and system time. A daily dial target on its own rewards short, low-quality calls, so pair it with a conversation or outcome measure.
Which dialing mode does each DialerBee plan include?
Starter ($69 per seat per month) includes preview and power dialing. Professional ($129 per seat per month) includes all four dialing modes: preview, power, progressive and predictive. Annual billing saves 20 percent, and carrier minutes are billed by your own carrier.
Fontes
- 16 CFR 310.4, Abusive telemarketing acts or practices, paragraphs (b)(1)(iv) and (b)(4). law.cornell.edu/cfr/text/16/310.4 (checked 2026-11-19)
- FTC, Complying with the Telemarketing Sales Rule. ftc.gov business guidance (checked 2026-11-19)
Leituras relacionadas
- Discador de potência vs. Discador preditivo vs. Discador progressivo
- Guia de Ajuste do Discador Preditivo
- Como melhorar as taxas de contato de saída
- Estimador de taxa de abandono
- Calculadora de ROI
Isenção de responsabilidade: This article is general information, not legal advice. DialerBee does not provide legal advice or guarantee regulatory compliance. It summarises 16 CFR 310.4 as published by Cornell Legal Information Institute and checked on 2026-11-19, and the rule may change. Throughput figures are illustrative examples, not performance claims. Confirm your obligations with qualified counsel.
Artigos relacionados
Como fazer testes A/B em campanhas de chamadas ativas
Leitura de 10 minutos
OperaçõesHow to Measure AI Voice Agent Containment
Leitura de 10 minutos
ConformidadeLei 32323 do Peru e o prefixo 500 para ligações de vendas
11 minutos de leitura
IndústriaPhone Survey Response Rates: AAPOR Formulas
12 minutos de leitura
Pronto para ver DialerBee em ação?
Solicite uma demo ao vivo de 15 minutos ou comece um teste grátis de 14 dias — sem slides, sem compromisso. As ligações começam assim que o seu próprio SIP trunk ou PBX estiver conectado e aprovado.
Teste grátis de 14 dias · crédito de IA gratuito · sem cartão · 11 idiomas · BYOC · controles de apoio à conformidade