GUEST POST by Thomas Howe, CTO at TEN DIGIT Communications

In the classic movie, Ben Hur, there’s a Roman galley scene. Below decks, a rather large man sits near the bow, banging rhythmically on the drum in front of him. The drum coordinates the rhythm of the slaves as they pull the oars, propelling the boat through the water. As the galley enters battle, the Roman captain screams “ramming speed” down the hatchway, causing the man to drum faster, and the slaves to pull harder. It doesn’t end well for anyone.

The rolling drums of a contact center

The modern-day contact center often feels similar, but substitute the slaves for agents, and the drum for the ringing phone. Agents spend their work days draining inbound call queues, tied to the rhythm of the inbound caller. Perhaps the greatest innovation of conversational communications – using asynchronous messaging channels that aren’t tethering a customer to an agent and thus enforcing waiting time (see also Chat vs. Messaging) – is the inversion of control from both the agent’s and customer’s perspective. Since conversational communications is asynchronous, it’s the humans that control the pace and flow of conversations. Ringing phones must be answered in real time; a text message can be answered when attention is available. The human impacts are nothing but positive: both customers and agents enjoy conversations that don’t have the stress of the ringing drumbeat. It’s a more efficient model as well, as many conversations can be had simultaneously. Conversations can span arbitrary time periods and are no longer constrained by 10-minute voice conversations and often are had after 60-minute on-hold times.

This transformation from synchronous to asynchronous conversations has many implications. Asynchronous teams:

  • are smaller, as they are more efficient
  • are less sensitive to surges in volume
  • have happier agents
  • have less conversation abandonment

Asynchronous messaging teams simply operate differently and must be managed and measured differently

Asynchronous messaging teams simply operate differently and must be managed and measured differently #cctr #custserv Click To Tweet


New channels require new metrics – A Proposal

The current approaches of measuring contact center performance are inappropriate for this asynchronous communications model. Service Level Agreements like on-hold time become meaningless when the conversations are text based. Contact center staffing is no longer as sensitive to call volume or patterns; spikes and valleys in demand are much more easily managed when conversations aren’t real-time.
After two years of designing and deploying smartphone contact centers, let me propose a more reasonable set of conversational contact center metrics:

  • Sessions: the most fundamental metric is “how many conversations did you have”? Sessions allows the administrator to gauge the number of customer problems and questions addressed.
  • Duration: for voice-centered communications, duration translates into the amount of time dedicated by an agent, solely for the call. For conversational contact centers, duration is not a measure of work, but is instead a measure of the “clock time” required to solve the problem.
  • Total Messages: the total number of messages sent or received by an agent measures both efficiency and engagement. Messaging, as opposed to voice, is more efficient not only for timing reasons, but for content reasons. Messaging conversations are shorter and more focused. Furthermore, they can include links to external content, and embed rich media. However, in situations where relationships are more important than efficiency, total messages represent the level of engagement between the agent and the customer. For most of our customers, messaging conversations typically involve between 10 and 12 messages.
  • Customer Delay: the amount of time that a customer requires to respond to the agent is a direct measure of the attention and interest this customer has. Customer delay is an important metric for measuring overall customer population interest, or to identify the eager among them.
  • Agent Delay: the amount of time that elapses before an agent responds to a customer directly measures agent responsiveness. In contrast to assignment delay (below), agent delay focuses in on how long it takes an agent to respond.
  • Agent Handling Time: Agent handling time is a measure of how much effort was required by an agent to support a given conversation. Unlike voice, agent handling time is almost always an estimate; it’s near impossible to determine how much time was actually needed by the agent to respond to an inbound message.
  • Assignment Delay: The best measure of appropriate levels of conversational contact center staffing is assignment delay – “How long did it take for me to find an agent?” Assignment delay is the conversational equivalent of on-hold time in voice.


Get blog articles delivered to your inbox