This is the third post in our series on the ROI of adding call-backs to your call center. Each post looks at one of the paths to cost reduction or revenue lift.
We’re purposefully putting aside the “softer” benefits of eliminating hold-time, i.e. the increase in customer satisfaction, agent happiness, NPS scores, etc. Those benefits are well understood and, for many companies, sufficient motivation to add call-backs. If you’re one of those companies, the ROI covered in these posts is icing on the cake. For companies where “hard” ROI numbers are the better path to budget approval, these posts will provide the business case you need.
In Part 1, we looked at reducing abandon rates. In Part 2, we looked at decreasing handle time. Today, we’re going to look at reducing telco costs.
Understanding Telco Cost
Telco costs can be divided into fixed costs such as hardware (e.g. switches, gateways), and variable costs that grow with call volume (e.g. toll-free and long distance fees). Not all costs fit cleanly into these categories (e.g. port licenses and trunk fees grow with concurrency, which is related to call volume but not always proportional), but they are a useful simplification for us.
In today’s discussion, we are just going to focus on costs that are purely driven by call volume. We will further simplify by combining all these costs into a single per-minute cost and then converting to cost-per-call as you see in Figure 1.
Before we get into the numbers, it’s important to appreciate that telco costs overall comprise a small fraction of total operational costs in the contact center. As Figure 2 shows, “Telecom and Networking” contributes only 5-10% of the total cost-per contact. This figure is taken from a research paper by Strategic Contact in which 5 different kinds of call center situations were modeled. For more detail, please see the original report.
Since we’re looking at a smaller component of cost overall, one should expect the cost savings to be small as well. Telco cost reduction is not often the primary driver of a call-back deployment. That role usually falls to reduction of abandon rate (covered in Part 1), reduction in handle time (covered in Part 2) or smoothing out volume spikes (will be covered in Part 4).
Even though telco cost reduction is a “supplemental” motivation, this exercise is still worthwhile because:
A) You want to make sure that costs are not going to increase and,
B) It never hurts to get a bit more ROI.
Calculating Call Duration
The dominant factor that drives variable telco costs in a call center is the inbound toll-free rate charged by their carrier. This number has a huge range; we’ve seen prices from $.002 to $0.06 per minute in the past year. For today’s exercise, let’s use $0.03 which is around the median. If the average call duration is 10 minutes, then the costs of the toll-free line to the call center is $0.30 per call.
Note that we are using “call duration”, which is different from “handle time” and different from “talk time”. This can be confusing, especially because there isn’t a universal agreement on how these terms should be used. Figure 3 illustrates the way we use these terms at Fonolo, which is consistent with most writing we’ve seen on the matter.
Note that call duration includes the time in the IVR, as well the queue time. From the carrier’s perspective, this is the only number that matters: it tells them how many minutes of raw telephony the call center consumes.
The Impact of Call-Backs on Telco Cost
To understand how adding call-backs impacts telco cost, we need to divide the call timeline into 3 stages. Figure 4 shows the timeline of a call that includes virtual queuing, lined-up with our familiar illustration of the caller experience. The stages appear in numbered circles:
- First, the caller dials in on the company’s toll free line and works through the IVR prompts to get to a queue.
- At some point in the queue, the caller hears an offer for a call-back and accepts it. At this point, the caller hangs-up and the toll-free line no longer needs to be used.
- When the caller’s turn arrives, an outbound call is made to him, and he is connected with the agent for the conversation.
Now let’s walk through how the telco costs change when call-backs are added. Figure 5 shows the timeline again, now with two “cost charts” lined-up below. These charts show the cost-per-minute at each stage (not to scale).
- During stage 1, the per-minute cost is the inbound toll-free rate, which we’ll label “C” in the cost charts. This is the same with or without call-backs.
- During stage 2, the caller has accepted the call-back and hung-up. This is the stage where the savings really adds up. For the duration of the queue time, the relatively high cost (rate “C”) of inbound toll-free is replaced with a very low cost connection (rate “A”). This rate is so low we will assume it is negligible for our calculations here. (See the section below “The Connection During Queuing” for an explanation why.)
- Stage 3 begins when the caller accepts an outbound call. The cost of this connection is a standard outbound telco rate (or possibly a long-distance rate) and is labeled rate “B”. This has a range from free (if the call center has outbound calls included in its flat-rate telco package) to $.02 per minute. We’ll use $.005 here. (To be precise, the connection to the call center from stage 2, rate “A”, is still active during this stage, but since it is negligible, we can just focus on rate “B”.)
Adding up the savings is fairly straightforward. Figure 6 works through a case where stage 2 is 5 minutes and stage 3 is 10 minutes. (Roughly corresponding to ASA of 5 minutes and Handle Time of 10 minutes.) With these numbers, the cost-per-call gets reduced from $.48 to $.08.
To complete the process of finding the total cost savings, let’s bring back “ExampleCo” from parts 1 and 2 of the series. This fictitious company receives 500,000 calls per year and has a 20% call-back rate. Therefor they would see a $.40 per call reduction on 100,000 calls, for a total savings $40,000.
The Connection During Queuing
In the previous section, we talked about the telco cost of stage 2 being negligible. Let’s understand why. During this stage, the caller’s place is being held in the queue. The way in which that a place is being held will determine the cost.
There are some call center platforms, such as Unified IP from Aspect or CIC 5 from Interactive Intelligence, that have a native virtual queuing built in. Other platforms, have add-on modules for virtual queuing, such as Callback Assist for Avaya or Courtesy Callback for Cisco. In all these cases, there is also no telephony required and the cost is truly zero.
In the case where the call-backs are provided by a 3rd party, such as Fonolo, there needs to be some external communication.
- The communication could be entirely on-premise, in which case the telco cost would still be zero. (With Fonolo this is achieved using our Call-Back Appliance).
- Or there could be VoIP-based telephony connection between the call center and the 3rd party, in which case the cost would just be the IP bandwidth, which is truly negligible.
- Finally, there could be PSTN connection, in which case the standard termination/LD rate would apply. Even so, the cost is still low and we’ll call it negligible for today’s exercise. (But, to be thorough, there may be cases where it is not.)
One final note: The description above applies to the scenario where a caller dials into the call center first and then accepts an offer to get a call-back via the IVR or ACD, i.e. responding to a message like, “Press 1 to get a call from the next available agent, rather than wait on hold.” (At Fonolo, this experience is provided by our In-Call Rescue product.)
Call-backs can also be triggered through interaction with a web or mobile interface using Web Call-Backs (formerly Visual IVR) or Click-to-Call-Back. (At Fonolo, our Web Rescue and Mobile Rescue products handle this.) All the logic and workflow described above could be adapted by simply skipping stage 1.