Dylan (apparently)

A new way to measure KPI

Categories: Call Centre Talk
Written by Dylan on 23/2/2005 at 2:50 pm

Measuring agent productivity

“Calls per Hour, Talk Time, Idle, and Wrap (ACW) are nonsensical targets in the majority of contact centre environments, yet they remain the number 1 key performance indicators for contact centre agents around the world!”

In the perfectly managed contact centre, it may be possible to guarantee a consistent flow of calls to all agents, however this requires impressive resource and workflow management. It is feasible that with a limited product and service range a genuine optimum “talk time” and “ACW” target can be applied to all calls. But do any of us work in such a simplistic and idealistic environment? (If so I have a CV ready to forward to you!)

So if we take away these foundations of our performance management culture what do we use to measure the productivity and efficiency of our contact centre agents? And who is responsible for ensuring team productivity and efficiency remains high?

Let us start by examining the traditional targets…

Calls per Hour
This is dictated by numerous factors controlled not by the agent but by the operation’s managers; therefore it is the managers who must be targeted on their teams’ efficiency (occupancy and utilisation)
? The resource levels of the team determine the “available” time in the team. “Available” time can be directly influenced through utilisation management, but not by the agents.
? The call traffic volumes dictate the frequency of calls hitting the queue. Only the customers can decide to call in.
? The length of calls and associated “Wrap” (ACW) determines the availability of agents to take calls.

Wrap (ACW)
Wrap time will vary depending on the nature and complexity of the call, as well as being influenced by the skills of the agent. However, poor Wrap is more likely to be caused by quality issues than productivity issues. If Wrap is a productivity issue then this will come out through the productivity measure of Occupancy.

Average Call Length
This is the same as targeting Wrap – the nature & complexity of the call (and of the callers) decide the length of call required.

Idle
“Idle” is signed on time where the agent is not involved in calls. Other than personal (comfort) breaks “Idle” must be planned, or where reactive must be tracked and recorded. As “Idle” usage is planned and therefore beyond the control of the agent, it cannot be targeted for the agent performance; to do so results in agents penalised for training, etc. A typical “Idle” breakdown for an inbound contact centre profiles as:

? Breaks 7.25 % (30 minutes per day)
? Personal breaks 2% (Average 8.5 minutes per day)
? Training / coaching 5% (Average 1 hr 45 minutes per week)
? Query for use by lower benchmark agents only against individual targets
? Meetings 2% (Average 3 hr per month)
? Idle “Admin” only to be used for accountable time, such as being taken off the phones to support administration
? Idle “None” must not occur for any reason.
? Idle “Training” must be accountable by either the training team or the Team Leader. Any ad hoc coaching should be recorded against “Training”.
? Idle “Query” must only be used by agents in the lower benchmark as agents’ queries must be answered during the call by either an assist or handoff, or during the wrap part of the call (recorded as “Wrap” or “ACW”). If calls are being handed off due to a skills or knowledge gap then this must be picked up as coaching or training, to be delivered when call flow allows. Daily ‘buzz sessions’ and ad hoc meetings such as daily target setting must be recorded against “Meeting”
? Agents must sign off the phones for lunch, as this is unpaid time. Remaining signed on distorts the true occupancy and shrinkage figures.

So if we are asked to manage performance without these cornerstones, what targets are left that can give us a useful measure of productivity in the contact centre? ….

Occupancy & Efficiency
? “Talk” time - Targeted against individual performance
? “Wrap” time - Targeted against individual performance
? “Available” time
? “Occupancy” = “Talk” + “Wrap” + ”Available” / “Talk” + “Wrap”

? Talk time is only partially under the control of agents. The requirement in most businesses is dependant on the nature of the individual call. Excessive talk time, when caused by lack of call control or knowledge, is a measure of quality issues not productivity, and should be addressed as such.
? Wrap time (ACW) is the only state an effectively managed agent can place themselves in to avoid taking further calls. However the length of wrap required will be determined by the nature of each call. It is therefore the correlation between Talk + Wrap that needs to be managed.
? Talk time vs. Wrap time provides an accurate measure of agent efficiency. As an agent’s knowledge, experience, and technical proficiency increases the % of talk time must increase and the % wrap time decrease.
? Talk time & wrap together provide the “occupancy” measure. Efficient occupancy is usually considered to be around 80-85%; the maximum occupancy before agents start displaying defensive and negative behaviours is 88% plus. The occupancy measurement is a % of time signed into “Talk”, “Wrap”, or “Available”. E.g. 65% “Talk” + 12% “Wrap” + 23% “Available” = 77% “Occupancy”

Quality
As well as the use of specific QMS procedures, quality can be highlighted through some of the traditional productivity KPI used in contact centres

? Talk time – Excessive talk time may indicate poor call control or lack of knowledge, very low talk time may be indicative of not fully servicing the client, or rushed work.
? Wrap – if wrap is disproportionate to talk time, then there is likely to be a knowledge issue.

Managing the business imperatives
While this form of targeting is fair & effective on agents, it does not necessarily address the business targets for your contact centre. This is where traditional targeting works far better.

Team Calls Per Hour
The CPH performance of your team is dictated by the resource profile against call arrival. A high CPH with poor SLA means understaffing, a low CPH is indicative of over resource. Where CPH is low against a high SLA managers should look to utilise their resource more effectively, such as training sessions, or supporting other functions. If CPH is consistently low and SLA are achieved then the centre is over resourced and redundancy or redeployment should be considered

Team Idle
As, in this model, all Idle is planned, excessive Idle in a team means poor resource planning and management. Idle is managed through scheduling of non-phone activity, and is supported by conventions such as “no more than X people out per shift”. Training, meetings, etc must be planned in line with resource profile vs. call arrival.

Sickness
Sickness is either genuine, or a conduct issue. Genuine sickness must be supportively managed, but can legitimately lead to dismissal if handled correctly, and when it is determined that an individual is not fit to work in their role. Setting a target for sickness at agent level penalises genuine sickness and sanctions limited “sickies”

Ok… so if you have made it this far then maybe you see some value in what I am proposing? I hope so! In which case get out into your business and canvas agents views to see if this could be applied in your contact centre… you may find (as I did) that your overall performance improves, agents feel they own their performance, managers manage; and who knows what additional benefits a happy & productive well managed environment may bring.


Powered by WordPress