Power or predictive dialler simulator and abandoned call rate

How to use it

Enter how many agents are talking at the same time, the share of attempts answered by a person, how long the conversation and the wrap-up last, and how much time the agent loses on each unanswered attempt when dialling in power mode. Then set the predictive pace, either as attempts per hour for the whole team or as a factor applied to the attempts that match capacity, and the abandonment threshold above which you want a warning.

The tool compares conversations per hour in both modes, works out how many people would answer with no agent free at that pace, and shows how abandonment changes if the real contact rate moves between 20% and 35%.

What it calculates: a continuous-flow model

It uses the same simplified model as the comparison power dialler or predictive dialler. With p as the contact rate:

• Power-mode cycle per contact = conversation + wrap-up + ((1 − p) / p) × time per unanswered attempt. Conversations per hour = agents × 3,600 / cycle; attempts = conversations / p.
• Team capacity = agents × 3,600 / (conversation + wrap-up). In the ideal predictive case the agent spends time only on contacts.
• Attempts that match capacity = capacity / p. The overdial factor multiplies this figure.
• Answered = attempts × p; connected = the lower of answered and capacity; abandoned = the rest.
• Abandoned call rate = abandoned / (abandoned + connected to an agent), Ofcom's formula.

The model treats calls and agents as constant averages over the hour. It does not count ringing before calls that are answered, nor answering machines.

Worked example

Fictitious data from the comparison, which are the default values: 10 agents, a 25% contact rate, 4 minutes of conversation, 1 minute of wrap-up and 30 seconds per failed attempt. In power mode the cycle is 300 + 3 × 30 = 390 seconds: each agent achieves 9.2 conversations per hour and the team 92, from 369 attempts. In predictive mode capacity is 10 × 3,600 / 300 = 120 conversations, matched by 480 attempts.

At 500 attempts, 5 / (5 + 120) = 4.0%, above the 3% warning threshold preset in the form. The tool's sensitivity table shows the risk of a list that answers more than forecast: at those 500 attempts, a 30% contact rate leaves 30 abandoned out of 150 (20.0%) and 35% leaves 55 out of 175 (31.4%). If only 20% answer, there is no abandonment, but the team handles 100 conversations instead of 120.

How to read the result and its limits

A continuous-flow model gives the average rate, not the rate in each minute. In reality answers arrive irregularly: even when the pace matches capacity, there will be moments with two extra answers and no agent free. With few agents, that effect weighs far more. Nor does it model answering machines, answering-machine detection that mistakes a person for a machine, or the automatic pace adjustment that real diallers perform.

Ofcom's page on silent and abandoned calls links to its 2016 statement, which applies from 1 March 2017. Its Annex 3 calculates the rate per calendar day and per campaign or call centre, and excludes silent calls, unconnected calls and calls to answering machines. Ofcom makes clear that the 3% in its 2010 policy was a criterion for prioritising cases, not a permitted margin. That is why the threshold is yours: an internal warning, not a legal limit.

In Spain, Article 66.1 of Law 11/2022 gives the right not to receive automated calls without human intervention for marketing purposes without prior consent, nor unwanted marketing calls without consent or another basis under Article 6.1 of the GDPR. That law sets no abandonment rate, so do not apply the British percentage as if it were a Spanish rule. A silent call does not identify the caller, and if you plan to play a recorded message, check with your adviser whether it falls under automated calls.

The decision it helps you make

If the rate exceeds your threshold with your own data, or the sensitivity table shows a large jump with only a few more points of contact, stay in power mode or lower the pace. In the example, the matched predictive pace gives 30% more conversations than power mode (120 against 92), but that gain only pays off if someone reviews abandonment per campaign every day. If you are choosing between dialling one line or several at once per agent, also read sequential or parallel dialler. Do not confuse this rate with abandonment in an inbound queue, where the person who hangs up is the one who was waiting.

The values you enter are calculated in your browser and are not sent to CallsIQ. The page measures visits with Google Analytics as described in the privacy policy.