Average Speed of Answer - Best Practicw

Topic Views - 1262

Resource and Scheduling Manager

Royal and SunAlliance

Average Speed of Answer - Best Practice
I am the Resource and Scheduling Manager in Royal Sun Alliance. I am putting a paper together for one of our Corporate Clients and they have asked me to advise them 'What is the industry standard in Call Centres for the Average Speed of Answer'

Can anybody help me with this?

MI Capability Manager

Vertex DataScience Ltd

The industry standard response is....

"There isn't one"

ASA is determined by what is acceptable to the client, taking into account available staffing levels (i.e. what they will pay for) and agent occupancy.

What you do need to understand is what your direct competition accept and then decide on whether your company can afford to match or better that target. However, getting your direct competition to give you their ASA is another thing altogether.

In the absence of this you will need to calculate staffing levels you will need for various ASA, show this to those clients and see what they are prepared to pay for.

Call Centre Manager


most common
is probably 80% in 20 seconds




No standard
I agree with Julian there is no standard. I've usually come across 90% of answerables in ten secs.



I've been asked
To calculate staffing levels on 95% in 50s


Dechaine Consulting Inc

I've seen 60/40 and 80/20 most commonly used.

but yeah....judging by the variety of factual answers there is no real answer.

Contact Centre Consultant

DarrylBeckford Limited

Sorry to be an a$$hole...
but all of these are an indication of Grade of Service, and not ASA.

Whilst there's no average or normal for either - there may be in your particular sector.

For example, OFWAT demand that a water company's GOS should be 90% in 30 seconds. This should translate to an ASA of under 30 seconds - although this would depend on how long the 10% that were over 30 seconds queued for. AFAIK (or remember) they don't specify an ASA.

Looking below there's some weird and wonderful figures. There's a huge range: 60/40 (JD) is likely to leave some pretty dissapointed customers, 90/10 (MS) must have left a lot of agents twiddling thumbs between calls.

It reminds me of a post by the lovely Dylan that you may find interesting:Click here


Call Centre Manager


yeah but no but
all centres I have dealt with use GoS as the primary, with ASA as a secondary measure.

As with most statistics, in isolation they don't tell you much! It's very possible to achive a reasonable ASA with a shocking GoS



ABC Ltd.

I agree with Zoe and Justin
In our multiskiing enviroment we follow 80/20 for few operations and 90/10 for few operation.
I have also observe 85/20 in many call center.


Call Centre Manager


there is no such thing...
as aleady well desrcibed by all who have replied, there is no such thing. I work in an industry where we need to answer a customer as quickly if not better than our competitors. That equates to an average of less than 2seconds over a year period. To estimate busy time for agents though, this sits at about 50% for a 37.5hr contract. Not very producive I know but a eutopia for all those trainers out there that want to have an opportunity to develop staff!



Lean Process Consultant

Worth Solutions Limited

Understand variation
>>For example, OFWAT demand that a water company's [Grade of Service] should be 90% in 30 seconds. This should translate to an ASA of under 30 seconds - although this would depend on how long the 10% that were over 30 seconds queued for.

Exactly, what happens to the 10%? Or the 40% with a GoS of 60/40?

In fact a better measure is to understand the capability of your system (not IT system, the system of work). This means plotting a run chart of all the calls and how long it takes to answer them. Add on Upper and Lower Control Limits as specified by Statistical Process Control and then you can understand if your system is in control.

By using something less crude than an average and more instructive than a binary target (we had more than 90% of calls answered in 30 secs, or we didn't), you can understand the system better and know when to act and when not to. I.e. you can stop firefighting and direct efforts to improving the system.

If anyone wants more explanation you can email me at xxxx or you can read "Understanding Variation" by Don Wheeler





Average Speed of Answer - the statistics
I agree with the previous responses, but they're not very helpful in answering the question and I happen to be sitting on a load of stats for the last 12 months so...

Average speed of answer in Financial Services is 16s, (75% under 21s)

and average time before callers abandoned the call is 53s

Of course this is just one measure and most customers want the question/problem resolved quickly (and first time), but answering the phone is a good start.




This all depends on the type of call.
Differing service levels may be appropriate for sales cales over service calls!

Channel Strategist


ASA factors to consider
Big factors to consider are: caller tolerance levels for holding, competition, if your call is acquisition or retention based, and if getting them to your delay announcement/filter statement contributes to your strategy.

Another big question to ask is if you are referring to the average daily/weekly or if you are managing to the interval level. Often the ASA is helped during a day by the non-spike periods that are more manageable, but will not help the experience/perception of those callers that tried during the busy times.

In the acquisition world, I think you would be more likely to see the :10, :20, :30, :40 second ASA's. However, I would also doubt that many tech support/customer support groups manage to that aggressive of a support level, particularly if you're managing to the interval.

I also would be wary of the surface results you'd find. Many centers mask ASA, with a call discriminator that adjusts offered volume to a threshold that filters out early abandons. This is a fine operational tool to improve daily results, but for a marketing team that wants to define the cost for media, it can make cost per response detail very inaccurate and inflated.

Call Centre Manager


Just wondering how you got on, and whether you were in a position to share any of your findings with teh Community.



Want to add a comment?

Not found what you were looking for?

1. Try searching through our site.
2. Still not got an answer?

Why not ask the Call Centre Helper Community? Click here to ask your question