Answering machine detection (AMD) is very much an art more then it is a science. AMD can be very picky and at times very inaccurate depending on line quality and how people answer the phone. If you are seeing a low connect rate, interviewers seem to be waiting extra long and you are using AMD we recommend turning it off.

To get an accurate rate take the total number of calls in the abandoned list and divide that number by the total number of connects. For example if you had 78 abandoned calls and your total connects were 4568 you would take 78/4568 and get a rate of 1.7%, this rate is very good.

In the UK, Ofcom requires that predictive dialers abandon less than 3% of answered calls and must not be called back within 72 hours unless there is a dedicated agent available.

In the US the FCC and FTC require that predictive dialers abandon less than 3% of answered calls (including Answering Machines) calculated over a 30 day period on a single campaign.

In Canada, the maximum abandon rate is 5%, and calls cannot be made to numbers registered with the National Do Not Call Registry, emergency or health care providers.

This is all dependent on the number of interviewers you have on the study and how clean your sample is. For example if you are dialing with 15 interviewers on a study where the connect rate is around 40% and you set your abandon rate on the dialer to 300 you will most likely have an abandon rate around 2.5% however if you take that same study and lower the number of interviewers maintaining the high connect rate and not lowering the abandon rate your rate is likely to jump to 3 or 4%.

We recommend setting your default rate at around 150 for 1-5 interviewers and then adjust that number as your increase interviewers. So 7-10 interviewers could be set at 200 and so on and so fourth.

An average connect rate when dialing on clean sample without other contributing factors can range anywhere from 30 to 50%. When a connect rate is low it can be caused by many factors, the first and most likely culprit is the actual sample however many factors including the weather can contribute to a low connect rate.

For example during hurricane Irene one of our clients saw that they were having a very low connect rate on the east coast, this is not surprising as most areas were being evacuated, to maintain productivity they hid all sample for the effected areas and this increased the connect rate.

At times a connect rate can be due to trouble with the Carrier, some carriers have call volume limits that you may hit when in full productions, for VoIP systems these limits are set by the carrier in there switch, you will want to know what your limits are before dialing. This is something that we have seen very rarely and most carriers notify you immediately.

If you find you have bad sample there are companies that maintain lists of known non-working numbers that you can scrub your sample against.

When you are considering VoIP you have to change your thinking slightly from the standard POTS world. Where in the standard telecom world you consider connections to be a line(POTS) the VoIP world uses the term Channel or Trunk. Depending on the carrier these words at times get used interchangeably. To calculate your call limit you would need to know the number of channels (call legs) that your carrier has set on your account, some carriers are dynamic and don’t have a set limit of concurrent calls but those account are usually billed at a higher rate than accounts that have set contracted limits. You will want to check with your service provider to get the actual call limit set for your account.

Having had many discussions over the years about how many outbound channels/lines you should have for each interviewer that is dialing the consensus seems to be a 2:1 ratio, so 2 channels/lines per every one interviewer that is dialing. Through the years however using different dialers and software we have found that a ratio of 1.5:1 works very well. Many dialer manufactures sold there dialers in this type of configuration and ratio. Prior to VOIP, PRI and ISDN trunks could be very expensive and a 1.5:1 ratio seemed to give the highest productivity for the lowest monthly overhead.

A ratio of 1.5:1 works rather well because at time you will have preview/auto dialed studies that only dial 1:1 this allows the other channels on the system to be used for predictive project allow for a 2:1 or sometimes a 3:1 ratio for the predictive studies on the system.

For example a dialer with 96 agents and 144 outbound channels is at a 1.5:1 ratio when all agents are dialing predictive so roughly for every 2 interviewers waiting 3 numbers are being dialed.

When you lose connection to Survent and the respondent is still on the line you will need to instruct your interviewers to write down the phone number of the respondent if they have it available, then instruct the respondent that they are having technical issues and if they may call back at a later time.  At this point you will need to then diagnose the reason you lost your Survent connection and restore that connection before logging back in and continuing with surveys.

In most cases the system will ask if you would like to kick the other extension off, if this does not happen notify the supervisor so that they can access the Dialer Dashboard and clear the station from the agents tab.

The survey will actually terminate any call to a respondent at the completion of the survey.  If you are ending the shift first have them quit out of the survey. Then, they can just hang up the phone. The Survox dialer will clear the agent once they hangup.

The survey should terminate any call to a respondent when you terminate out of a survey. They then can continue dialing or if you are ending the shift first have them quit out of the survey. Then, they can just hang up the phone. The Survox dialer will clear the agent once they hangup.

Yes and No, this depends on if you have the interviewer hang up the phone when they go on breaks.  We recommend that interviewers just leave the phone off hook from the time they start interviewing till the time they are done.

When and interviewer is between calls and not sitting on the break or quit screen they should remain connected to the dialer and be ready to receive a call.  If they are on the break or quit screen you can have them stay connected to the dialer as mentioned above for breaks or have them hangup and then reconnect as an agent before choosing to continue interviewing.