HK40046513B - Method, system and article for benchmarking performance in a contact center system - Google Patents

Description

Methods, systems, and artifacts for benchmarking performance in contact center systems.

This application is a divisional application of the invention patent application filed on May 23, 2017, by the applicant "European Afiniti Technologies GmbH", with application number 201780016552.3 and title "Technology for Benchmarking Performance in a Contact Center System".

Cross-references to related applications

This international patent application claims priority to U.S. Patent Application No. 15/176,899, filed on June 8, 2016, the entire contents of which are incorporated herein by reference.

Technical Field

This disclosure generally relates to contact centers, and more specifically, to techniques for benchmarking performance in contact center systems.

Background Technology

A typical contact center algorithmically assigns incoming contacts to agents that are available to handle them. Sometimes, a contact center may have available agents and be waiting to assign inbound or outbound contacts (e.g., phone calls, internet chat sessions, emails). At other times, a contact center may have contacts in one or more queues waiting for agents to become available for assignment.

In some typical contact centers, contacts are assigned to agents based on the time when those agents become available, and agents are assigned to contacts based on their arrival time. This strategy can be called a "first-in, first-out," "FIFO," or "round-robin" strategy.

Some contact centers may use performance-based routing (PBR) or similar methods to sort queues of available agents, or sometimes to sort contacts. PBR sorting strategies attempt to maximize the expected outcome of each contact-agent interaction, but typically do not consider the even utilization of agents within the contact center.

When a contact center changes from using one matching strategy (e.g., FIFO) to another (e.g., PBR), some agents may be available to receive contacts while others may be on another call. If average agent performance is uneven over time, the overall effectiveness of one matching strategy can be unfairly impacted by the other.

In light of the foregoing, it will be understood that a system may be required that can benchmark the performance of a contact center system, including the management of alternative routing strategies to detect and resolve imbalances in average agent performance within alternative matching strategies.

Summary of the Invention

Techniques for benchmarking performance in a contact center system are disclosed. In one particular embodiment, the technique can be implemented as a method for benchmarking the performance of a contact center system, comprising: circling between a first contact-agent pairing strategy and a second contact-agent pairing strategy for pairing contacts with agents in the contact center system, by means of at least one computer processor configured to perform contact center operations; determining, by means of at least one computer processor, an agent utilization deviation in the first contact-agent pairing strategy, the agent utilization deviation in the first contact-agent pairing strategy including the difference between a first agent utilization rate and a balanced agent utilization rate of the first contact-agent pairing strategy; and determining, by means of at least one computer processor, the relative performance of the second contact-agent pairing strategy compared to the first contact-agent pairing strategy, based on the agent utilization deviation in the first contact-agent pairing strategy.

According to other aspects of this particular embodiment, the method may further include adjusting the target agent utilization rate of the second contact-agent pairing strategy via at least one computer processor to reduce the agent utilization rate deviation in the first contact-agent pairing strategy.

According to other aspects of this particular embodiment, the method may further include, by means of at least one computer processor, determining the average available agent performance of a plurality of agents during at least one transition from a first contact-agent pairing strategy to a second contact-agent pairing strategy.

According to other aspects of this particular embodiment, the method may further include, by means of at least one computer processor, determining the average availability of at least one of a plurality of agents during at least one transition from a first contact-agent pairing strategy to a second contact-agent pairing strategy.

According to other aspects of this particular embodiment, the method may further include, via at least one computer processor, outputting a conversion management report of agent utilization deviations including a first contact-agent pairing strategy.

According to other aspects of this particular embodiment, the first contact-agent pairing policy may be a performance-based routing policy.

According to other aspects of this particular embodiment, the second contact-agent pairing strategy may be a behavior pairing strategy.

According to other aspects of this particular embodiment, the second contact-agent pairing strategy may be a hybrid behavior pairing strategy, and the hybrid behavior pairing strategy may be biased towards a performance-based routing strategy.

According to other aspects of this particular embodiment, the method may further include adjusting at least one parameter of the second contact-agent pairing strategy via at least one computer processor.

According to other aspects of this particular embodiment, at least one parameter includes a Kappa parameter for a hybrid behavior pairing strategy.

According to other aspects of this particular embodiment, the first contact-seat pairing strategy may target unbalanced seat utilization, and the second contact-seat pairing strategy may target balanced seat utilization.

According to other aspects of this particular embodiment, at one or more time points between the transition from the first to the second contact-agent pairing strategy and a subsequent transition from the second to the first contact-agent pairing strategy, the target utilization rate of the second contact-agent pairing strategy may be adjusted at least once.

In another specific embodiment, the technology can be implemented as a system for benchmarking performance in a contact center system including at least one processor, wherein at least one processor is configured to perform the methods described above.

In another specific embodiment, the technology can be implemented as an article of manufacture for benchmarking performance in a contact center system, comprising: a non-transitory processor-readable medium, and instructions stored on the medium; wherein the instructions are configured to be read from the medium by at least one processor, thereby causing at least one processor to operate in order to perform the methods described above.

The present disclosure will now be described in more detail with reference to specific embodiments as illustrated in the accompanying drawings. Although the present disclosure has been described below with reference to specific embodiments, it should be understood that the present disclosure is not limited thereto. Those skilled in the art who have access to the teachings herein will recognize additional implementations, modifications and embodiments, and other areas of use within the scope of the present disclosure set forth herein, and that the present disclosure may have significant utility in relation to these additional implementations, modifications and embodiments, and other areas of use.

Attached Figure Description

For a more complete understanding of this disclosure, reference is now made to the accompanying drawings, in which like elements are indicated by like numbers. These drawings should not be construed as limiting the disclosure, but are intended for illustrative purposes only.

Figure 1 shows a block diagram of a contact center system according to an embodiment of the present disclosure.

Figure 2 shows a schematic representation of a seating arrangement table according to an embodiment of the present disclosure.

Figure 3 shows a schematic representation of a seating arrangement table according to an embodiment of the present disclosure.

Figure 4 illustrates a schematic representation of a seating arrangement diagram according to an embodiment of the present disclosure.

Figure 5 illustrates a schematic representation of a seating arrangement diagram according to an embodiment of the present disclosure.

Figure 6 shows a schematic representation of a seating arrangement diagram according to an embodiment of the present disclosure.

Figure 7 shows a schematic representation of a seating arrangement diagram according to an embodiment of the present disclosure.

Figure 8 shows a flowchart of a benchmark transformation management method according to an embodiment of the present disclosure.

Detailed Implementation

A typical contact center algorithmically assigns incoming contacts to agents that are available to handle them. Sometimes, a contact center may have available agents and be waiting to assign inbound or outbound contacts (e.g., phone calls, internet chat sessions, emails). At other times, a contact center may have contacts in one or more queues waiting for agents to become available for assignment.

In some typical contact centers, contacts are assigned to agents based on the time when those agents became available, and agents are assigned to contacts based on their arrival time. This strategy can be called a "first-in, first-out," "FIFO," or "round-robin" strategy. For example, the longest available agent pairing strategy preferably selects the available agent that has been available for the longest time.

Some contact centers may use performance-based routing (PBR) or similar methods to rank queues of available agents, or sometimes to rank contacts. PBR ranking strategies attempt to maximize the expected outcome of each contact-agent interaction, but typically do not consider the even utilization of agents across the contact center. Some variations of PBR may include a top-performing agent pairing strategy, preferably selecting the available agent with the highest performance, or a top-performing agent pairing strategy for the contact type, preferably selecting the available agent with the highest performance for the paired contact type.

As another example, some contact centers may use a “behavioral pairing” or “BP” strategy, under which contacts and agents are intentionally (preferably) paired in a manner that allows for the allocation of subsequent contact-agent pairs, such that the combined benefits of all allocations under the BP strategy may exceed those of FIFO and PBR strategies. BP is designed to encourage balanced utilization of agents within the skilled queue while still improving overall contact center performance beyond what is allowed by FIFO or PBR methods. This is a remarkable achievement because BP operates on the same calls and the same agents as FIFO or PBR methods, utilizing the agents provided by FIFO roughly evenly, while still improving overall contact center performance. BP is described, for example, in U.S. Patent No. 9,300,802, the entire contents of which are incorporated herein by reference. Additional information regarding these and other features of the pairing or matching module (sometimes also referred to as “SATMAP,” “routing system,” “routing engine,” etc.) is described, for example, in U.S. Patent No. 8,879,715, the entire contents of which are incorporated herein by reference.

In some embodiments, the contact center may periodically switch (or “cycle”) between at least two different pairing strategies (e.g., between FIFO and PBR; between PBR and BP; between FIFO, PBR, and BP). Additionally, the outcome of each contact-agent interaction may be recorded along with the identifiers of the pairing strategy (e.g., FIFO, PBR, or BP) that have been used to assign the identifier to that particular contact-agent pair. By tracking which interaction produces which outcome, the contact center can measure performance attributable to a first strategy (e.g., FIFO) and performance attributable to a second strategy (e.g., PBR). In this way, the relative performance of one strategy can be benchmarked against another. At numerous switching periods between different pairing strategies, the contact center can more reliably attribute performance gains to one strategy or another. Additional information regarding these and other features of benchmarking pairing strategies is described, for example, in U.S. Patent Application No. 15/131,915, filed April 20, 2016.

When a contact center changes from using one matching strategy (e.g., PBR) to another (e.g., BP), some agents may be available to receive contacts, while others may be interacting with contacts (e.g., currently on a call). If average agent performance is uneven over time, the overall performance of one matching strategy may be unfairly affected by the other. For example, when a contact center uses PBR to match contacts and agents, high-performing agents are more likely to be busy interacting with contacts, while low-performing agents are more likely to be idle. Therefore, when transitioning from PBR to another matching strategy such as BP, the average performance of available agents over time may be lower than the average performance of all agents, including both available and busy agents.

Figure 1 illustrates a block diagram of a contact center system according to an embodiment of the present disclosure. As shown in Figure 1, the contact center system 100 may include a central switch 110. The central switch 110 may receive incoming contacts (e.g., callers) or support outgoing connections to contacts via a telecommunications network (not shown). The central switch 110 may include contact routing hardware and software to facilitate routing contacts between one or more contact centers, or to facilitate queuing or switching components within one or more PBX/ACDs or contact centers.

If there is only one contact center, or only one PBX/ACD routing component in contact center system 100, then central switch 110 may be unnecessary. If more than one contact center is part of contact center system 100, each contact center may include at least one contact center switch (e.g., contact center switches 120A and 120B). Contact center switches 120A and 120B may be communicatively coupled to central switch 110.

Each contact center switch in each contact center can be communicatively coupled to multiple agents (or "pools"). Each contact center switch can support a certain number of agent (or "seat") logins at the same time. At any given time, a logged-in agent may be available and waiting to connect to a contact, or a logged-in agent may be unavailable for various reasons—such as connecting to another contact, performing post-call functions such as logging information about the call, or taking a break.

In the example of Figure 1, central switch 110 sends a connection to one of two contact centers via contact center switches 120A and 120B, respectively. Each contact center switch 120A and 120B is shown to have two agents. Agents 130A and 130B can log in to contact center switch 120A, while agents 130C and 130D can log in to contact center switch 120B.

Contact center system 100 can also be communicatively coupled to integrated services from, for example, third-party vendors. In the example of FIG1, conversion management module 140 can be communicatively coupled to one or more switches in the switching system of contact center system 100, such as central switch 110, contact center switch 120A, or contact center switch 120B. In some embodiments, the switches of contact center system 100 can be communicatively coupled to multiple benchmarking modules. In some embodiments, conversion management module 140 can be embedded within components of the contact center system (e.g., embedded in a switch or otherwise integrated with a switch). Conversion management module 140 can receive information from a switch (e.g., contact center switch 120A), or, in some embodiments, from a network (e.g., the Internet or a telecommunications network) (not shown) regarding agents (e.g., agents 130A and 130B) logged into the switch and information regarding incoming calls via another switch (e.g., central switch 110).

The contact center may include multiple pairing modules (e.g., a BP module and a FIFO module) (not shown), and one or more pairing modules may be provided by one or more different vendors. In some embodiments, one or more pairing modules may be components of the transition management module 140 or one or more switches, such as the central switch 110 or contact center switches 120A and 120B. In some embodiments, the transition management module 140 may determine which pairing module can handle pairing for a particular contact. For example, the transition management module 140 may alternate between enabling pairing via the BP module and enabling pairing via the FIFO module. In other embodiments, a pairing module (e.g., the BP module) may be configured to emulate other pairing strategies. For example, the transition management module 140 or a transition management component integrated with the BP component in the BP module may determine whether the BP module can pair with BP or emulate FIFO pairing for a particular contact. In this context, "BP on" may refer to the time when the BP module is applying the BP pairing strategy, while "BP off" may refer to other times when the BP module is applying a different pairing strategy (e.g., FIFO).

In some embodiments, regardless of whether the pairing strategy is handled by a separate module or if some pairing strategies are simulated within a single pairing module, a single pairing module can be configured to monitor and store information about pairings performed under any or all pairing strategies. For example, a BP module can observe and record data of FIFO pairings performed by a FIFO module, or a BP module can observe and record data of simulated FIFO pairings performed by a BP module operating in FIFO simulation mode.

The embodiments of this disclosure are not limited to benchmark change management for only two paired strategies. Instead, benchmark change management can be performed for two or more paired strategies (e.g., benchmark change management for FIFO, PBR, and BP).

Figure 2 illustrates a schematic representation of an agent transition table 200 according to an embodiment of the present disclosure. In the example of Figure 2, four agents named "Alice," "Bob," "Charlie," and "Donna" may be assigned to a specific queue to interact with contacts. These agent names are for illustrative purposes only; in some embodiments, anonymous identifiers or other identifiers may be used to represent agents in the contact center. Furthermore, this highly simplified example shows only four agents. In some embodiments, hundreds, thousands, or more agents may be assigned to queues and may be illustrated in an agent transition table.

Seat transition table 200 shows five transitions labeled "201", "202", "203", "204", and "205". In some embodiments, each transition may represent the point in time when the contact center switches from one pairing strategy (e.g., FIFO) to another pairing strategy (e.g., BP). Transitions may occur multiple times per hour (e.g., every 10 minutes, every 15 minutes, every 30 minutes), or more frequently or less frequently over a day, week, month, year, etc. In some embodiments, transitions may be identified by the time they occur. For example, transition 201 may occur at 9:15 AM, transition 202 may occur at 9:45 AM, and so on.

At transition 201, agents Alice and Bob are unavailable, as shown in the shaded unit. For example, Alice and Bob may be interacting with the contact, or they may otherwise be occupied by post-interaction tasks such as recording sales or archiving customer service reports. Meanwhile, agents Charlie and Donna are idle or otherwise available to connect to the contact, as shown in the unshaded unit.

Similarly, at transition 202, agents Charlie and Donna are busy, while agents Alice and Bob are available. At transition 203, agents Alice and Charlie are busy, while agents Bob and Donna are available. At transition 204, agents Bob and Donna are busy, while agents Alice and Charlie are available. At transition 205, agents Bob and Charlie are busy, while agents Alice and Donna are available.

In any single transition, even pairing strategies aimed at balancing agent utilization (e.g., FIFO and BP, but not PBR) may appear to have skewed utilization at the time of the transition. For example, if Alice's normalized performance rating is 80, Bob's is 60, Charlie's is 40, and Donna's is 20, then the average performance of all agents is 50. However, the average performance of available agents at transition 201 (i.e., Charlie and Donna) is 30, below average. The average performance of available agents at transition 202 is 70, above average. The average performance of available agents at transition 203 is 40, below average. The average performance of available agents at transition 204 is 60, above average.

In some transitions, even pairing strategies that target unbalanced seat utilization (e.g., PBR) may appear to have balanced utilization at the time of transition. For example, at transition 205, the average performance of available seats (i.e., Alice and Donna) was 50.

While the average performance of available seats varies across any single transition, the average performance of available seats over time (e.g., throughout the day) across multiple transitions can reflect the statistically expected utilization of a given pairing strategy. Seat transition table 200 illustrates five transitions 201-205, which in some embodiments may not be statistically significant multiple transitions. However, for illustrative purposes, the average available seat performance across the five transitions 201-205 is (30+70+40+60+50)/5 = 50. In this example, the average available seat performance at the transitions across the five transitions 201-205 is balanced.

In some embodiments, in addition to determining or replacing the average performance of available seats on one or more transitions, the average availability of individual seats may also be determined and output. For example, in seat transition table 200, the average availability per seat for each transition 201-205 is 60% for Alice (3 out of 5 transitions), 40% for Bob (2 out of 5 transitions), 40% for Charlie (2 out of 5 transitions), and 60% for Donna (4 out of 5 transitions). For pairing strategies aimed at balancing seat utilization (e.g., FIFO or BP), statistically, each seat may have approximately the same number or proportion of transitions available. In this simplified example, which only describes five transitions 201-205, the average availability per seat varies between 40% and 60%. However, over time, the average availability per seat may statistically converge to the same percentage. For example, after 100 transitions, the average availability per seat may be approximately the same, such as 50%, 55%, 60%, etc.

Figure 3 illustrates a schematic representation of an agent transition table 300 according to an embodiment of the present disclosure. In contrast to the example of agent transition table 200 (Figure 2), agent transition table 300 illustrates an unbalanced pairing strategy, such as PBR, typically expected in contact centers. In some embodiments of PBR, the highest-performing agent (i.e., Alice) may preferably be selected to interact with the contact. Therefore, Alice is never available at any transition 301-305. Meanwhile, the lowest-performing agent (i.e., Donna) is always available at each transition 301-305.

The average performance of available seats is 30 at transition 301, 40 at transition 302, 30 at transition 303, 20 at transition 304, and 40 at transition 305. The average performance of available seats across the five transitions 301-305 is (30+40+30+20+40)/5 = 32, which is unbalanced. Over time, the magnitude of this statistically significant bias in the average performance of available seats, indicating seat utilization, could “contaminate,” bias, or otherwise affect the effectiveness of the alternative pairing strategy after each transition, leading to potentially unfair benchmark measurements.

In some embodiments, in addition to or instead of determining the average performance of available agents across one or more transitions, the average availability of each agent may also be determined and output. For example, in agent transition table 300, the average availability of each agent for each transition 301-305 is 0% for Alice (0 out of 5 transitions), 40% for Bob (2 out of 5 transitions), 60% for Charlie (3 out of 5 transitions), and 100% for Donna (5 out of 5 transitions). For pairing strategies (e.g., PBR) targeting unbalanced agent utilization, some agents (e.g., lower-performing agents) may be statistically significantly more available than others (e.g., higher-performing agents). Even in this simplified example illustrating only five transitions 501-505, the average availability of each agent varies significantly between 0% and 100%. The statistical significance of the changes in the average availability of each agent can be further confirmed over time. In this paper, unbalanced pairing strategies such as PBR always, or almost always, transfer lower-performing agents to the next pairing strategy (e.g., BP or FIFO), while higher-performing agents are never, or almost never, switched. As mentioned above regarding average agent quality during one or more transitions, the average agent availability over time, indicating a statistically significant skew in agent utilization, can “contaminate,” bias, or otherwise affect the effectiveness of alternative pairing strategies after each transition, leading to potentially unfair benchmark measurements.

Figure 4 illustrates a schematic representation of a seat transition chart 400 according to an embodiment of the present disclosure. In the seat transition chart 400, the x-axis represents a time period. For example, in a week, x = 0 could represent day 1, x = 1 could represent day 2, and so on. The y-axis represents the average performance of available seats during all transitions from a first pairing strategy to a second pairing strategy over a given time period. For example, at x = 0 (e.g., day 1), the average performance of available seats at the transition of the day is 50. At x = 1 (e.g., day 2), the average performance is slightly above average, and at x = 3 (e.g., day 4), the average performance is slightly below average. However, the seat transition chart 400 shows a relatively stable average performance over a relatively long time period (e.g., a week). In some embodiments, small daily variability may be statistically insignificant, and the overall seat utilization of the first pairing strategy is balanced.

Figure 5 illustrates a schematic representation of a seat shift chart 500 according to an embodiment of the present disclosure. In the seat shift chart 500, the overall seat utilization rate remains stable at approximately 25% per day, significantly lower than the average. Therefore, the overall seat utilization rate of this pairing strategy (e.g., PBR) is unbalanced.

When benchmarking across multiple pairing strategies, the first pairing strategy (e.g., PBR) may "pollute" or otherwise bias the performance of a second pairing strategy (e.g., FIFO or BP). At each transition from PBR to BP, the average performance of available agents may be significantly lower than the overall average performance of all agents assigned to the queue (i.e., imbalance). This "suppressed" agent pool at the start of a BP or FIFO cycle may weaken the overall performance of BP or FIFO for that cycle.

Conversely, in each transition from BP or FIFO to PBR, the average performance of available agents can be similar to or equal to the overall average performance of all agents assigned to the queue (i.e., balanced). This balanced agent pool at the start of each PBR cycle can enhance the overall performance of PBR for that cycle, since even a balanced agent pool can be better than the typical agent pool resulting from PBR.

Because each PBR cycle makes the agent pool more "contaminated" (unbalanced) than when it was received, and each BP or FIFO cycle makes the agent pool "cleaner" (balanced) than when it was received, some techniques used to benchmark PBR relative to BP or FIFO can make BP or FIFO appear to perform worse than PBR cycles when they do not contaminate their available agent pool at the beginning of each cycle. Therefore, it is helpful to compare the average performance of available agents at the beginning of the cycle ("first half") with the average performance of available agents at the end of the cycle ("second half").

Figure 6 illustrates a schematic representation of a seat transition chart 600 according to an embodiment of the present disclosure. The seat transition chart 600 illustrates an exemplary first/second half comparison. At x = 0 (e.g., day 1), the difference between the average performance of available seats transitioning to the first pairing strategy and those transitioning from the first pairing strategy during the day is 0. At x = 1, the difference is slightly above 0, and at x = 3, the difference is slightly below 0, but the overall difference remains close to 0 over the week. Conceptually, the pairing strategy results in seat pools with substantially the same average performance (e.g., quality).

A mean difference of 0 does not necessarily mean that both pairing strategies are balanced (e.g., the average performance of available agents is approximately 50). For example, if the first pairing strategy is PBR_A with an average available agent performance of 25, and the second pairing strategy is PBR_B with an average available agent performance of 25, the difference is still 0. From a benchmarking perspective, imbalance between two pairing strategies is acceptable if each pairing strategy is approximately as unbalanced as possible. In this way, each pairing strategy makes the agent pool as severe as it was when it was found, and the pairing strategy does not pollute the other.

Figure 7 illustrates a schematic representation of a seat transition chart 700 according to an embodiment of the present disclosure. The seat transition chart 700 shows another example of a first/second half comparison. At x0 (e.g., day 1), the difference between the average performance of available seats transitioning to the first pairing strategy and those transitioning from the first pairing strategy during the day is 25. At x=1, the difference is slightly higher than 25, and at x=3, the difference is slightly lower than 25, but the total difference over the week remains close to 25. Conceptually, one pairing strategy consistently and significantly pollutes the seat pool of the other pairing strategy during transition. For example, if the first half of the PBR strategy consistently receives a seat pool with an average performance of 50, and the second half of the PBR strategy consistently provides a seat pool with an average performance of only 25, the average difference is 25.

A mean difference significantly higher than 25 does not necessarily mean that either of the pairing strategies is balanced (e.g., the average performance of available agents is approximately 50). For example, if the first pairing strategy is PBR_A with an average available agent performance of 25, and the second pairing strategy is PBR_B with an average available agent performance of 0, the difference is still 25. The PBR_B pairing strategy still contaminates the benchmark detection, causing PBR_A to perform worse than when not cycled with PBR_B, and causing PBR_B to perform better than when not cycled with PBR_A.

Figure 8 illustrates a flowchart of a benchmarking transition management method 800 according to an embodiment of the present disclosure. At block 810, the benchmarking transition management method 800 can be initiated. The contact center system can cycle through at least two pairing strategies. For example, the contact center system can switch between BP and PBR pairing strategies. In each transition from BP to PBR or vice versa, the available agents for each transition can be determined.

In box 810, a first average performance of available agents over time during the transition from a first pairing strategy (e.g., BP) to a second pairing strategy (e.g., PBR) can be determined based on the determination of available agents for each transition and their relative or otherwise normalized performance. The first average performance can also be considered as a “first half” measure of the second pairing strategy for that time period.

In block 820, in some embodiments, a second average performance of available agents over time during a transition from a second pairing strategy (e.g., PBR) to a first pairing strategy (e.g., BP) can be determined based on the determination of available agents for each transition and their relative or otherwise normalized performance. The second average performance can also be considered a “second half” measure of the second pairing strategy for that time period.

In block 830, in some embodiments, the average performance difference between the first and second average performances can be determined. If the difference is equal to or close to zero, it can be determined that there is no significant difference between the average performance of available seats received from or provided to the first pairing strategy during the measurement period. If the difference is greater than zero, it can be determined that the average performance of available seats provided by the first pairing strategy (e.g., BP) is higher than the average performance of available seats provided by the second pairing strategy (e.g., PBR), indicating that the second pairing strategy may be contaminating the available seat pool and the benchmark.

In box 840, in some embodiments, a conversion management report may be generated. In some embodiments, the conversion management report may include a first average performance difference determined in box 810, a second average performance difference determined in box 820, an average performance difference determined in box 840, or any combination thereof. Data may be presented in various formats, including but not limited to agent conversion tables (e.g., agent conversion tables 200 and 300 (Figures 2 and 3)) or agent conversion charts (e.g., agent conversion charts 400, 500, 600, and 700 (Figures 4-7)). Reports may be generated dynamically and updated continuously or periodically. The report may include user interface elements for displaying, sorting, filtering, or otherwise selecting which data to display and how to display it. The report may be fully auditable, allowing viewers to examine the source data for each element. For example, the report interface may include user interface elements that display a list of agent identifiers available at a given conversion and their corresponding relative or normalized performance metrics.

In block 850, in some embodiments, at least one parameter of the first or second pairing strategy can be adjusted to, for example, reduce the average performance difference determined in block 830. Reducing or eliminating a non-zero average performance difference can reduce or eliminate the extent to which a pairing strategy inhibits performance or contaminates the benchmark detection of the second pairing strategy.

For example, in a contact center system that loops between PBR and BP, PBR may suppress the configuration of BPs aimed at evenly utilizing seats. Various techniques allow BPs to aim for uneven seat utilization. For example, adjusting the "Kappa" parameter can bias BPs towards PBRs in terms of seat utilization. Kappa is described, for example, in U.S. Patent Application No. 14/956,086, the entire contents of which are incorporated herein by reference.

If the Kappa is high enough, benchmark detection suppression or contamination can be eliminated (e.g., the average performance difference is zero). However, in some environments, a high Kappa value can degrade overall BP performance. In these cases, it may be necessary to compensate for PBR benchmark contamination by having a high initial Kappa value after the transition from PBR to BP, and to reduce or eliminate Kappa adjustments in the first 3 minutes, 10 minutes, etc. (e.g., reducing Kappa from 1.5 to 1.0). The rate of this "Kappa decay" can be adjusted to balance benchmark suppression from PBR with overall performance in the first half of the BP cycle.

Similarly, it might be desirable to have a high Kappa value before transitioning from BP to PBR, with Kappa adjustments made or increased in the last 3 minutes, 10 minutes, etc. (e.g., Kappa increasing from 1.0 to 1.5). The rate of this "reverse Kappa decay" can be adjusted to balance the baseline suppression from PBR with the overall performance in the latter half of the BP cycle.

In contact center systems that cycle between FIFO and BP, the average performance difference is typically zero because both FIFO and BP aim to achieve balanced agent utilization. However, in some environments, it is desirable or preferred for BP to aim for unbalanced agent utilization (e.g., a Kappa value greater than 1.0). If BP aims for unbalanced agent utilization, the average performance difference compared to FIFO may not be zero, indicating a suppressed or contaminated baseline detection. In these cases, it may be desirable to reduce or eliminate Kappa adjustments in the last 3 minutes, 10 minutes, etc. (e.g., reducing Kappa from 1.5 to 1.0). The rate of this "Kappa decay" can be adjusted to reduce the average performance difference between BP and FIFO to zero while simultaneously optimizing overall BP performance.

Following block 850, the benchmark change management method 800 may terminate. In some embodiments, the benchmark change management method 800 may return to block 810. In some embodiments, the steps may be optional, performed in a different order, or performed in parallel with other steps. For example, adjusting at least one parameter in block 850 may be optional, or may be performed before or simultaneously with generating the change management report in block 840.

At this point, it should be noted that the benchmark performance in the contact center system according to this disclosure, as described above, may involve, to some extent, processing input data and generating output data. This input data processing and output data generation can be implemented in hardware or software. For example, specific electronic components may be used in a switch management module or similar or related circuitry to implement the functions associated with the benchmark performance in the contact center system according to this disclosure, as described above. Alternatively, one or more processors operating according to instructions may implement the functions associated with the benchmark performance in the contact center system, as described above, according to this disclosure. If this is the case, these instructions may be stored on one or more non-transitory processor-readable storage media (e.g., a disk or other storage medium) or transmitted to one or more processors via one or more signals embedded in one or more carrier waves, both within the scope of this disclosure.

This disclosure is not limited to the specific embodiments described herein. In fact, various other embodiments and modifications thereof, besides those described herein, will be apparent to those skilled in the art from the foregoing description and drawings. Therefore, these other embodiments and modifications are intended to fall within the scope of this disclosure. Furthermore, although this disclosure has been described herein in the context of at least one specific implementation in at least one specific environment for at least one particular purpose, those skilled in the art will recognize that its usefulness is not limited thereto, and that this disclosure can be advantageously practiced in a variety of environments for various purposes. Therefore, the claims should be interpreted in light of the full scope and spirit of this disclosure as set forth herein.

Claims (12)

1. A method comprising: At least one computer processor communicatively coupled to the contact center system and configured to perform benchmarking operations in the contact center system cycles between a first contact-agent pairing strategy and a second contact-agent pairing strategy for pairing contacts and agents in the contact center system, wherein the cycle includes: alternately applying the first contact-agent pairing strategy and the second contact-agent pairing strategy in the contact center system to pair contacts and agents according to the corresponding pairing strategy applied for connection in the contact center system; The at least one computer processor determines a first set of transitions from the first contact-agent pairing strategy to the second contact-agent pairing strategy. The first average performance of at least one of the available agents in the first conversion set is determined by the at least one computer processor. The at least one computer processor determines a second set of transitions from the second contact-agent pairing strategy to the first contact-agent pairing strategy; The at least one computer processor determines the second average performance of at least one of the available seats in the second conversion set; The average performance difference between the first average performance and the second average performance is determined using the at least one computer processor; and The at least one computer processor adjusts at least one parameter of at least one contact-agent matching strategy based at least in part on the average performance difference. 2. The method of claim 1, wherein adjusting at least one parameter of at least one contact-agent matching strategy reduces the first average performance difference. 3. The method of claim 1, wherein at least one parameter of at least one contact-agent pairing strategy is adjusted such that the at least one contact-agent pairing strategy reduces the uniform utilization rate of agents in the contact center system. 4. The method of claim 1, further comprising: For each of the multiple cycles during the said cycle, determine the first average available agent performance at the beginning of that cycle; For each of the plurality of cycles, determine the second average available agent performance at the end of that cycle; and Based on the first average available agent performance and the second average available agent performance, determine whether at least one of the first contact-agent pairing strategy and the second contact-agent pairing strategy is polluting the available agent pool for the other contact-agent pairing strategy. 5. The method of claim 4, wherein adjusting at least one parameter of at least one contact-agent matching strategy is further based on: It is determined that at least one of the first contact-agent pairing policy and the second contact-agent pairing policy is polluting the available agent pool for the other contact-agent pairing policy. 6. The method of claim 5, wherein determining that at least one of the first contact-agent pairing policy and the second contact-agent pairing policy is polluting the available agent pool for the other contact-agent pairing policy comprises: Determine whether the first average available agent performance is greater than or less than the second average available agent performance; Based on the determination that the first average available agent performance is greater than the second average available agent performance, it is determined that the first contact-agent matching strategy is polluting the second contact-agent matching strategy; and Based on the determination that the first average available agent performance is less than the second average available agent performance, it is determined that the second contact-agent matching strategy is polluting the first contact-agent matching strategy. 7. The method of claim 6, further comprising: Based on the determination that the first average available agent performance is not greater than or less than the second average available agent performance, it is determined that at least one of the first contact-agent pairing strategy and the second contact-agent pairing strategy is not polluting the available agent pool for the other contact-agent pairing strategy. 8. The method as described in claim 4, Each of the plurality of cycles occurs over a period of time. The plurality of periods further includes a first set of periods and a second set of periods. The contact center system uses the first contact-agent matching strategy during the first period set. The contact center system uses the second contact-agent matching strategy during the second periodic set, and The time period corresponding to the period in the first period set is separate from the time period corresponding to the period in the second period set. 9. The method of claim 8, wherein, The time periods corresponding to the first period set are interleaved with the time periods corresponding to the second period set. 10. The method of claim 1, wherein, At least one of the pairing strategies includes a behavior pairing strategy. 11. A system comprising: At least one computer processor communicatively coupled to a contact center system and configured to perform benchmarking operations in the contact center system, wherein the at least one computer processor is further configured to: The system cycles between a first contact-agent pairing strategy and a second contact-agent pairing strategy for pairing contacts and agents in the contact center system, wherein the cycle includes: alternately applying the first contact-agent pairing strategy and the second contact-agent pairing strategy in the contact center system to pair contacts and agents according to the corresponding pairing strategy applied for connecting in the contact center system. Determine a first set of transitions from the first contact-agent pairing strategy to the second contact-agent pairing strategy; Determine the first average performance of at least one of the available seats in the first conversion set; Determine a second set of transitions from the second contact-agent pairing strategy to the first contact-agent pairing strategy; Determine the second average performance of at least one of the available seats in the second conversion set; Determine the average performance difference between the first average performance and the second average performance; and At least one parameter of at least one contact-agent matching strategy is adjusted, at least in part, based on the average performance difference. 12. An article comprising: Non-transient processor-readable medium; and Instructions stored on the medium; The instructions are configured to be read from the medium by at least one computer processor, which is communicatively coupled to a contact center system and configured to perform benchmarking operations in the contact center system, thereby causing the at least one computer processor to operate such that: The system cycles between a first contact-agent pairing strategy and a second contact-agent pairing strategy for pairing contacts and agents in the contact center system, wherein the cycle includes: alternately applying the first contact-agent pairing strategy and the second contact-agent pairing strategy in the contact center system to pair contacts and agents according to the corresponding pairing strategy applied for connecting in the contact center system. Determine a first set of transitions from the first contact-agent pairing strategy to the second contact-agent pairing strategy; Determine the first average performance of at least one of the available seats in the first conversion set; Determine a second set of transitions from the second contact-agent pairing strategy to the first contact-agent pairing strategy; Determine the second average performance of at least one of the available seats in the second conversion set; Determine the average performance difference between the first average performance and the second average performance; and At least one parameter of at least one contact-agent matching strategy is adjusted, at least in part, based on the average performance difference.