There have been many incidents where I have been involved in discussions around Answering Machine Detection(AMD). Lets first try to understand what is AMD and why it deserves a article in this blog. Most of the international Outbound Call Centers (Dialling to US UK etc) face a issue that around 40% – 60% of the numbers usually go to Answering Machines. This can waste a lot of time for the agents hence Predictive dialer usually do something called “Answering Machine Detection”. Objective is simple to filter out the answering machins and give only Human contacts to the agents. How is it done?

Well the dialer actually listens to the call when connected and the AMD algorithm identifies whether its a human being or an answering machine. If its a answering machine it drops the call (or leaves a message) otherwise it transfer the call to the agent.

All this happens in a few seconds (and if it does not the customer hangs up the call and you get a CALL DROP).

Each vendor talks about a detection rate and interestingly there is no definition of answering machine detection %.

Lets assume that during a shift Dialer detected 20 Answering machines and 20 Answering machines went to agents. Most people would jump to the conclusion that Answering machine detection is 50%. I am afraid that is not correct. Apart from the above two numbers it is important to see how many CONNECTED calls where there in total.

Lets take two scenarios.

Scenario A:

(CC) Total Connected Calls (At dialer level) 100

(DD) Calls disposed as AMD by dialer 20

(AD)Calls disposed by Agents as AMD 20

Scenario B:

(CC) Total Connected Calls (At dialer level) 50

(DD) Calls disposed as Answering Machines by dialer 20

(AD)Calls disposed by Agents as Answering Machines 20

By our previous calculation the detection would be the same but actually in Scenario A dialer has done detection on 100 calls and in scenario B it was on 50 calls. Remember all calls that were actually humans and went to agents are also correct detection.

Hence the % mistake that dialer has done is = AD/CC

Scenario A 20%

Scenario B 40%

So Correct Answering Machine detection for both scenarios is:

Scenario A 80% detection

Scenario B 60% detection

One interesting fact is that this does not consider all humans that were detected as answering machines by the dialer (and there is no way to establish that apart from testing it manually for 100s of known numbers)

So next time when a vendor talks about the detction rate do calculate it this way to ascertain.

The crude and correct way to calculate the Answering machine Detection algorithm is

(1-(AD/CC))%

I would like to hear from vendors on this observation and see if we can arrive at a better formula. Any Call Centers who want to share there numbers are also welcome.

Here is the result link:

http://results.beijing2008.cn/WRM/ENG/INF/BX/C73/BXM054303.shtml#BXM054303

All the best to other two!

So lets do this; 

All solutions are allowed to post their sales pitch and comparative analysis. Let it be informative and have the contact informations so that people can reach you through this forum. 

so go ahead; Lets see who has to say what.

- And the War begins

I am hoping all Sales people can use this brainstorming discussion to take it back to product dev teams. The following is the format we will be using. I will start entering some of the basic features.

FEATURE:

DESCRIPTION:

IMPORTANCE:

ABOUT YOU (OPTIONAL)

FEATURE:    Predictive Dialling

DESCRIPTION: This is a automatic dialling functionality wherein dialer would pick up numbers automatically from a list and dial those numbers. Idea is to keep agents talking to customers for maximum amount of time. The dialer should consider various parameters like talk time, ringing time, avg setup time etc and then predict how many calls should be thrown at any moment.

IMPORTANCE: HIGH in INTERNATIONAL CENTERS, MEDIUM in DOMESTIC CENTERS

ABOUT YOU (OPTIONAL): Owner of the Blog.

Your IT team plays a very imp role in operations of a call center and care must be taken if they will do the job. Lets try to break-up the skills into technical and soft skills. (Both are equally imp)

Technical Skills:

- Excellent understanding of networking, LAN, Network cards, configuration of TCP / IP (In windows and linux), Using troubleshooting techniques like trace route, ping

- Good understanding of Databases (ability to write SQL queries)

- Very good knowledge of MS Excel

- Basic knowledge of UNIX (Most solutions are on Unix)

- understanding of Call Center Scenario

Soft skills

- Excellent communication skills because poor fellow would be co-ordinating with your vendors and operations people.

- Stress/Pressure handling skills

Happy Debugging!

Now may be touchstar philippines is just looking to give standalone Asterisk PBX or may be they have already integrated Asterisk onto their Call Center Solution.

This is another shot in the arm for Asterisk in Call Center Industry (Not vicidial puhleez!)

What are Cube, Telesynergy thinking? Any takes if these guys have got a SIP version of there solution?

Is anyone exploring there own soft-switch?

http://www.callcentersindia.com/showall-orig.php?value1=36553.

Very interesting discussion and here are my 2 cents. I have worked with Asterisk in various scenarios.

Fact of the matter is its a excellent tool as a IP-pbx and a switch (mind the scale) but to use it as a appication platform is nto a wise decision. Most of the BIG and Serious providers use Asterisk only as a soft switch. vicidial does not. It still uses Queues provided by Asterisk which till date have problems.

Recently in a discussion with a learner American gentleman he also expressed his views about voice logging capabilities of asterisk. Its best NOt to use Asterisk for mixing of your voice log streams, as it clogs the CPU and causes voice breakage.

Solutions that use Asterisk but not its queues (as far as my knowledge goes are: AHEEVA (canada), ALTITUDE (Europe), ASPECT (Yes … Aspect has a asterisk compatible version and DRISHTI (India). One is free to add more to this list.

Criterion is that

- Solution should not use Asterisk queues

- Soltuion should be able to work with multiple Asterisk servers and same application server

- Should have managed Load balancing in load conditions (say 400 channels per server with full Voice logging)

Basically a premise based solution is one where the technology is residing in your premises. There by you can own the technology but also have to maintain it. So you will need a IT Dept to manage the solution for you.

A Hosted technology is where the application is located remotly in some data center and is distributed through Internet. This is possible for International Centers only as in India VoIP at this level is still not allowed.

A few hosted providers that I came across are:

1. Five9    (Recently pulled operations from India)
2. Touchstar
3. DACX from Drishti
4. Aspect (A Reseller provides hosted service)
5. Vicidial based solutions

You have to sign up on a monthly leased basis and voIP may or may not be billed inclusive. Consider legal issues also. I am not sure if five9 has a voip license also in india.
In general if you are a small timer wanting to experiment in this buisness without getting into lot of tech issues go for hosted as IT is not a major hassle there but as always there is a cost involved to it.