The Art And Science Of Call Answering
In this article today, I’d like to discuss a metric Evolve Tele-services has found to be very useful in the Direct Response (DR) world. Unique ANI Throughput, or UAT, is a term we coined twenty years ago while working on our first DR project, The Total Gym (prior to that our backgrounds were in CS and Tech support). UAT is the concept of measuring individual callers versus gross call volume, the entrenched industry standard. Callers responding to DR advertising have distinct behaviors from the standard call center customer service model, and we believe that UAT is the most effective measurement of said behaviors.
The Value Of A Call Vs. A Caller
The core difference between UAT and the customer service standard is how incoming calls are viewed. The customer service model is to assign a value to every attempt: a call is a call. UAT assigns a value to each unique phone number that attempts to call: a caller is a caller. The customer service model will count every attempt by a single phone number as a distinct event, while UAT will only count each phone number once. This is critical in the DR environment, where a single ANI may make multiple attempts to reach a live operator. When using UAT, what becomes important is not how many attempts were made by a single ANI, but what the ultimate outcome was. That is, if one ANI had ten separate calls on a DR spot, the ten calls are only relevant to the extent that one of them was eventually answered or not. We believe this process is critical in understanding DR call behavior and ultimately maximizing production, for several reasons.
I’m sure we’re all aware of the challenges that DR programs pose when compared to standard customer service dynamics. In a typical customer service environment, call volume is predictable from day to day. While there may be unexpected events, you generally know from historical patterns how many calls will come and when they will come. This allows for staffing to address the patterns at a level the client deems sufficient. For example, XYZ company knows they’ll receive approximately 3,000 customer service calls per week, peaking on Monday and falling off through Friday. They know that with operating hours from 8:00 AM to 5:00 PM, the call volume will fall from interval to interval in a set pattern. Given this data, they will apply the correct number of agents to deliver call handling that meets their client’s expectations. These are usually driven by metrics such as abandon percentages, average speed of answer, average abandon time, etc. The problem arises when you try to apply the same logic to DR.
Handling DR Call Volume
DR is quite the opposite of customer service when discussing call handling. Call volume on DR programs will fluctuate wildly from day to day, week to week, etc. The days of the week the volume falls on will change as well. The hours of the day where the volume comes will vary. And most importantly, how the volume arrives is significantly different. For example, in DR you can’t just say you’ll have 100 calls between 9:00 and 10:00. You need to anticipate 100 calls between 9:13 and 9:34. In addition to managing the needs of call handling in this fluid environment, you have the additional component of closing the sale. Handling the call is obviously the first critical step of the process, but the ultimate resolution is closing the sale. Thus, in addition to handling the call volume to a satisfactory level, the DR call center will take every step possible to drive the calls to the most successful agents. For example, in a customer service environment, virtually every staff member is “logged in” to the same group and has the same anticipated outcome. In DR, some staff members may be excluded from a group, as their proficiency in selling a given product is not sufficient to risk losing the opportunity. This is where UAT becomes absolutely necessary to maximizing production.
A New Metric
After many years of analysis with multiple companies, the staff at Evolve has learned to take DR caller behavior and direct it through the process to provide the highest degree of success on the sales side. Specifically, this involves limiting the queue size of each individual group and prompting excesses to a “forced busy” status. In the old model, a call center could and would accept every call presented to it. Calls are routed to all available agents, and the rest are put into a holding queue, where they will remain until someone becomes available, regardless of time. With the UAT model, once all available agents are on call, a portion of the remaining calls will be brought into the holding queue. Anything incoming call beyond that will be played a “forced busy”, where Evolve’s switch technology has instructed the carrier to return a busy signal to the caller. In the vast majority of cases (85 – 90% on average), the caller will make additional attempts to get through. While there are limits on the number of retries the average caller will attempt, data indicates that busy signals in moderation may actually spur the callers buying interest, especially when the alternative is staying on hold for excessive periods of time. Bottom line, hold time generally serves to weaken a callers buying instinct, while a modicum of forced busies has the opposite effect. Additionally, this method allows for more calls to be routed to what we would classify as “A” agents, or those that give the call the highest degree of likelihood for a sale outcome.
Evolve has learned the secret to using UAT as a measurement tool and can track UAT real time. If an ANI abandons and never calls back, we know. If they call back five minutes later and connect to an agent, we know that as well. Thus, UAT as provided by Evolve is an absolute figure. For example, if Evolve reports 88% UAT on a given day, it truly means that 88 out of 100 callers went to an agent and had some type of outcome. The remaining 12 never reached an agent. That 12 will include abandons that didn’t call back, hang ups before the call was answered, etc. Evolve’s goal for UAT on a long form advertisement is 88-92% on a weekly average. We’ve found through years of data that this range provides the right balance of outcome for both client and call center, and that the remaining callers were not comprised of many sellable opportunities. This is not to say that other metrics (such as average speed of answer or abandon time) are not used, but rather they are driven off UAT. If UAT is at or above acceptable levels, the other metrics are typically in line. If UAT falls short, it will be observed in these other metrics as well.
While some in our industry still use the older methodology for call handling, many more are using some version of UAT as their core, and the metric has become more prevalent. We believe that eventually all DR centers and clients will move to UAT as a measurement tool. It simply addresses the uniqueness of the DR caller and focuses on the preferred outcome, which for DR, is closing the sale.