Perception is Reality: Is the "Accent Effect" Hurting Your Agent Performance?

Say it isn't so. Actually, I do not think one contact center in the U.S. would knowingly allow conditions in the workplace that are unfair or knowingly subject individuals to hostile work environments. I would even go as far to say that some over-compensate to avoid even the slightest appearance of discrimination. Many contact centers are also proactive at protecting their agents against abuse from customers.

Or are they? Could many contact centers be accepting prejudice from its customers and allowing discrimination of its agents to occur? Is your contact center one of them?

The Quest for Equality

Contact centers are driven by and driven to various key metrics that define performance. Among the many, agent performance metrics need to be measured consistently and applied evenly and fairly across the workforce.

Media, personal opinions, and dare we say prejudice, has created a negative perception of off-shoring in American society...

Performance weaknesses, as defined by the customer and by internal standards, are identified and addressed via various performance management techniques such as product training, listen skills training, system usage training. In-house versus out-sourced centers are held accountable to the same metrics. Best practices are shared among the centers and teams. Focus is on improving performance metrics.

Perceived Differences

A client of Customer Relationship Metrics expressed concern that service by its off-shore center may not be as effective as their on-shore location. Media, personal opinions, and dare we say prejudice, has created a negative perception of off-shoring in American society and this organization was experiencing this effect.

As the rumble within the organization grew louder, management worried that the impact to the organization would be far larger than the operational savings. They needed to prove or disprove the validity of this perception with scientifically valid customer intelligence before a multi-million dollar mistake was made. Customer Relationship Metrics set off to further analyze our client's feedback collected via the External Quality Monitoring (EQM) program for the contact center agents located domestically and overseas.

First, we analyzed overall company satisfaction to see if the off-shore center had a negative impact on the brand ratings. Was this the case? It was not.

Second, we looked to see if customers were less likely to recommend the company when they were served by the off-shore center. This has to be true. Again, no.

Third, we looked at a customer-defined First Contact Resolution (FCR). Certainly the off-shore centers performed worse at once-and-done and this was to be the defining issue. Struck out again. Surprisingly, we discovered the off-shore centers perform slightly better at contact resolution. Big blow for the off-shore naysayers in the organization. We pressed on.

Lastly, we turned to the agents. Do off-shore agents achieve comparable satisfaction evaluations? They did not. In fact their scores were significantly lower.

Better, but Not Better?

But wait, how could the off-shore agents be receiving lower "satisfaction with agent" scores when we always find "satisfaction with agent" scores to be highly correlated with resolution. The off-shore agents should be experiencing this same effect. How is it possible that service could be "better" (resolution as defined by the customer) while experiencing lower "satisfaction with agent" scores?

Ah, the technology had to be different? No. Their queue times longer? No. The IVR was different? No. Nothing of significance is different between these virtual centers except the physical location. The caller did not know which call center the call was routed to and did not know that a location was off-shore. What is different about the agents? Are customers reacting not to the proficiency or skill of the agents but rather something else?

The Accent Effect

Is it possible the presence of an accent can mentally frame the call for the customer? Is the stereotypical perception that off-shore service is not as good influencing the results? Is there a prejudice against non-American accents? It is something because, clearly, the data proved the service was better.

We weren't done yet. If it is prejudice that is impacting the results, then a measurable difference would exist between scores obtained by domestically-located agents who spoke with non-American accents compared to those with American accents. We needed to control for the effect of accent.

Customer Relationship Metrics analyzed nearly 5,000 customer service evaluations (surveys) collected over a six month period via an EQM program. Each survey was linked to the specific contact center agent who provided the service experience. For the analysis, each survey was classified as belonging to an agent with an American accent or an agent who spoke with a non-American accent.

First we analyzed if the "overall company satisfaction" evaluation was lower for the non-American accent? It was not.

Next we checked to see if the "likelihood to recommend the company" was lower for the non-American accent? Again, we found that it was not.

Is the service center perceived to be a less valuable resource to the customer after a call with a non-American accented agent? Again, no, this was not the case.

Okay, do agents with non-American accents resolve a smaller percentage of their calls? Nope, it was statistically the same.

The moment of truth. Do agents with non-American accents achieve comparable satisfaction evaluations? The answer: No, they do not. Scores are significantly lower.

The findings revealed that the diction of the agent had no significant impact on a customer's perception of the company, its products or the value of the contact center to that customer. Little impact was noted in comparing a customer's likelihood to repurchase or recommend the company's products. However, when asked to rate their satisfaction with the agents' performance during the call, customers distinctly favored those agents with American accents.

The proof (or prejudice) simply stated is that customers expressed significantly greater satisfaction with key aspects of their service experience with the American accented agents.

Prejudiced against non-American accents

Within the first few seconds of the call when a customer hears an accent, he or she immediately makes the assumption that non-American accent means a lower level of service is about to delivered. The call is framed differently in the customer's mind based on the stereotypical prejudice. The framing effect means that the caller assesses a low number of service experience points for this transaction and the agent must earn them back during the interaction.

Conversely, when an accent is American, the customer mentally awards all points for the experience and the agent works to keep the points during the transaction. It is more difficult to earn the points back than to keep them.

Unfair Labor Practices

Your organization is affected by a perpetuating stigma pertaining to off-shore contact centers whether you off-shore or not. There is a prejudice that an accent equals off-shore and that off-shore service will not be as good because of the communication challenges. By mentally framing an accent as an inferior service provider, agents are subjected to a service prejudice that is formed in less than 15 seconds.

If you are like our client, you have heard it on calls. Your customer feedback program includes derogatory comments about off-shoring (even if you are not off-shoring). You have read it on blogs. Agents with non-American accents are often times subjected to abuse by customers.

How large of an issue is this for your contact center? What does the management team do to insure a consistently high quality service experience? Avoid hiring agents with accents? Of course that cannot be the plan. So what can be done to leverage on-shore agents with non-American accents as well as the off-shore agents? Research as described above can easily identify the magnitude of the Accent Effect in your center(s). Without an empirical piece of research, the answer cannot be known.

When an Accent Effect is quantified, the management team has options to avoid discriminating against these agents by reducing the effect of the prejudice through accent neutralization activities, particularly with respect to the greeting and early portion of the call flow.

Acknowledge that this inequity exists and support the agents as they work to provide delightful customer experiences. And lastly, knowing that these agents must work harder than their American-accented counterparts, it may be necessary to set different performance goals for these two distinct groups of agents. Continuous improvement over time is less discriminatory than hitting the same marks on the metrics.

A Cost of Doing Nothing?

Is there a cost to being one of the contact centers that accepts prejudice from its customers and allows discrimination of its agents to occur? Subjecting individuals to a hostile work environment by not providing them the tools to effectively serve customers and to then hold them accountable to same standards is no different than not accommodating persons with disabilities. Certainly, there is a cost associated with doing nothing to correct something that is discriminatory.