Assessment of FairPoint’s Stabilization Plan Status Report的費(fèi)爾波因特穩(wěn)定計(jì)劃狀況評估報(bào)告

1、.The Liberty Consulting Group July 13, 2009FairPoint Stabilization Plan Status Report AssessmentAssessment of FairPoints Stabilization Plan Status Report Liberty Consulting GroupJuly 13, 2009IntroductionOn March 31, 2009, FairPoint Communications Inc. (FairPoint) issued a Stabilization Plan, which F

2、airPoint stated was “designed to ensure FairPoint returns to Business as Usual by the end of the second quarter 2009.” FairPoints plan also identified the customer affecting issues it has experienced since it cutover from the Verizon Communications Inc. (Verizon) systems to the new systems and proce

3、sses developed for FairPoint by Capgemini, and FairPoints strategy for mitigating the impact of these issues on its customers. FairPoint issued an update to this plan on April 17 to address specific issues identified with the original plan, provide more detail on FairPoints recovery and mitigation p

4、lans, and define specific milestones for critical areas of the business. Since the middle of April, FairPoint has provided daily status reports to members of the regulatory staffs (Staffs) in each of the three states and to the Liberty Consulting Group (Liberty) to show its progress toward meeting i

5、ts end of the second quarter Business as Usual (BAU) commitment. FairPoint has also issued weekly “Milestone Update” reports, which are publicly available documents, and held weekly conference calls with Staff members and Liberty to discuss progress.FairPoint provided a “Stabilization Plan Status Re

6、port” (Status Report) on July 8, 2009, which was intended to provide the current status of FairPoints effort to return to BAU. FairPoints report focused on five primary areas of its business: · Call centers · Order flow · Service order backlog · Billing· Customer escalations

7、 to the regulatory staffs. For each of these five areas, FairPoints report reflects the progress it has made since cutover and highlights where additional improvements are needed to reach normalized levels, which FairPoint defines as “service from a customers perspective that equates to what they we

8、re receiving from FairPoint prior to cutover.” In its report, FairPoint contends that “there has been significant progress made during the second quarter,” but that “in most of the areas, additional systems improvements are needed to truly be at normalized levels.” Although FairPoint uses the term “

9、normalized levels” rather than “Business As Usual” in the Status Report, the implication is that FairPoint has admitted not achieving its original commitment to achieve BAU operations by the end of the second quarter 2009. After release of FairPoints Status Report, the Staff of the Vermont Departmen

10、t of Public Service requested Liberty to provide this summary assessment of the report. Although Liberty notes in this assessment some areas of inconsistency between the numbers reported in the Status Report and those provided in the daily reports to the Staffs, Liberty cannot verify the accuracy of

11、 the reported data in most cases. As a result, Libertys assessment relies on an assumption that the data are accurate, except in a few cases where Liberty has contrary evidence. .AssessmentFor convenience, Libertys assessment follows the same structure as FairPoints Status Report, considering each o

12、f the five primary operational areas FairPoint identified in turn. Call Centers FairPoint has three main call centers: the Consumer Call Center, the Business Call Center, and the Repair Call Center. For the Consumer Call Center, which has the largest call volumes of the three, FairPoint reported tha

13、t “it has experienced a dramatic improvement since March and for the month of June operated at a normalized level.” FairPoint cited many reasons for the improvement in this center including call volumes returning to pre-cutover levels, a strong management team, the productivity of the service repres

14、entatives working in the center, and time saving improvements identified with the aid of an outside consulting firm (Aricent). FairPoint noted that it is critical for the systems used to remain stable, because system failures and defects significantly affect the centers ability to perform effectivel

15、y.FairPoint reported that its Repair Call Center has also experienced a “marked improvement” since March and delivered normalized levels of service for the month of June. FairPoint noted that the results in this center depend significantly on outside factors, such as poor weather conditions, which a

16、ffect the volume of calls into the repair center. FairPoint indicated that such events affected the repair centers results prior to cutover and will continue to do so in the future.FairPoint stated that its Business Call Center has experienced improvement since March but did not perform at normalize

17、d levels. FairPoint attributes the poor performance of this center to increased workload. According to FairPoint, this workload increase resulted from FairPoints adding approximately 100 account managers to work directly with FairPoints large business customers. This increase in the business sales f

18、orce resulted in additional calls to the Business Call Center. FairPoint indicated that it is adding customer service representatives to the Business Call Center and expects this center to “consistently meet its objectives by the end of August.”Libertys AssessmentBased on the data FairPoint quotes,

19、Liberty generally agrees with FairPoints assessment of its Consumer and Repair Call Centers performance. FairPoints data shows a marked improvement in the performance of both the Consumer and the Repair Centers. However, these improvements were marred in late June by the increase in the volume of ca

20、lls into the Repair Center resulting from the service issues caused by the large amount of rain experienced in the northeast throughout the month and by a systems problem that affected the system response time for the service representatives in the Consumer Center in late June and early July. Libert

21、y agrees with FairPoints statement that it is critical that the systems remain stable for the call centers to perform efficiently. However, the FairPoints call centers have been subject to systems problems since cutover, involving both systems performance and effectiveness. The problems were suffici

22、ently severe that FairPoint engaged Aricent to help address these issues. The most recent performance problem calls into question whether the systems problems have been fully resolved and whether stability issues might cause continued problems with call center performance in the future.FairPoint cor

23、rectly indicates continued problems with the Business Call Center. However, Liberty believes that FairPoints Status Report has ignored some key issues. Performance in this center has shown little improvement since the middle of April, when FairPoint began to report its results to Liberty and the Sta

24、ffs. Furthermore, if FairPoints explanation is correct that the poor results in this center result from the addition of account managers without a complementary addition to the Business Call Center staff to handle the additional work load created, it indicates a dual failure of the FairPoint managem

25、ent team in planning for the additional work in the center and a failure to react quickly to the problem. Other call center problems have been reported to FairPoint by the Staffs that were not addressed by FairPoint in its report. For example, the Vermont Consumer Affairs and Public Information orga

26、nization has reported that customers continue to complain about (1) being put “on hold” for an inordinate amount of time by FairPoints service representatives and (2) having their call dropped by FairPoint while speaking with a service representative. FairPoint needs to investigate these problems to

27、 identify and report the root causes and the corrective action taken to resolve them.Liberty concludes that FairPoints reported data indicates its Consumer and Repair Call Center performance has greatly improved. However, as FairPoint notes in its Status Report, maintaining this improvement depends

28、on the systems stability, which is still in doubt. FairPoint admits that it needs to continue improving its Business Call Center performance, and it remains to be seen whether that can be accomplished by the end of August, as promised. FairPoint also should examine other complaints about the call ce

29、nters raised by the state Staffs.Order FlowWithin the order flow category of issues, FairPoint includes wholesale per-ordering, retail order entry, and order flow-through.The only pre-ordering issue FairPoint considered in the Status Report was Customer Service Record (CSR) inquiries. Systems issues

30、 experienced immediately after cutover generally prevented the competitive local exchange carriers (CLECs) from executing automated CSR inquiries. To compensate for this, FairPoint established a manual process for the CLECs to obtain the CSR data that they cannot obtain automatically. In its Status

31、Report, FairPoint indicates that it was able to process automated CSR requests starting in April. FairPoint indicates that it has “continued to improve the accuracy of the information requested by the CLECs and have been at normalized levels for over a month,” and quoted the success in reducing the

32、number of manual CSR requests as evidence of this.FairPoint states that “a primary reason we initially had issues right after cutover in processing our customers orders was a result of a higher percentage of orders falling out or not flowing through the systems than anticipated.” Orders that do not

33、flow-through require manual intervention to complete the transaction, resulting in service delivery delays. In the Status Report, FairPoint quoted flow-through rates for the seven different product types accounting for 99 percent of all orders. FairPoints data shows the overall flow-through rate for

34、 these seven product types increased to 87 percent from 54 percent in March. FairPoint also states that “we moved back to our standard interval schedule on May 22nd and are delivering most of our new requests within this due date.” FairPoint acknowledges that more improvement is needed, especially f

35、or retail DSL orders and for wholesale number port (“CB”) orders, for which FairPoint quotes current flow-through rates of 84 percent and 65 percent, respectively.FairPoint explains that an “unsubmitted” order is one that has not yet moved into the provisioning system and that there are many legitim

36、ate reasons for this condition; however, immediately after cutover, many orders were incorrectly routed to the unsubmitted queue. After a significant decrease in unsubmitted orders by the end of March, the number of such orders has held relatively constant, with 2,374 at the end of June. FairPoint b

37、elieves that the orders in the queue will continue to vary between 1,500 and 2,500, “but the primary measure of success is the ability of the customer service team to move orders through this queue as planned.”Libertys Assessment While it is true that the CLECs are having more success obtaining an a

38、utomated CSR response, the “improved accuracy” of the information contained in the CSR pre-order response remains questionable. As recently as July 9, during the weekly call FairPoint holds with its CLEC customers, many CLECs complained about inaccurate and missing information in the CSRs responses.

39、 It appears that when the CLECs receive an inaccurate CSR they work directly with FairPoint personnel to obtain the correct data rather than issue a manual CSR request, accounting for some portion in the decline in manual CSR requests that FairPoint quotes as evidence of the accuracy of the CSR resp

40、onses. Additionally, CLECs have had numerous problems with other pre-order transaction types, such as address validation and loop qualification, not discussed in FairPoints report and it remains unclear whether all of these other pre-order issues have been resolved. Thus, it may be premature to clai

41、m that FairPoints pre-order process is at normalized levels. FairPoint needs to devote more effort to investigate and resolve the pre-order problems the CLECs are still experiencing. Liberty agrees with FairPoint that a major reason for the service delivery problems since cutover is that the actual

42、flow-through rate did not meet expectations based on the pre-cutover test results reported by Capgemini. Liberty also agrees that FairPoint appears to have improved its flow-through rates since cutover. However, the flow-through rates quoted by FairPoint in its Status Report appear to be overstated

43、for a number of reasons. First, the current flow-through results reported by FairPoint in its Status Report are generally much higher than the results reported to the Staffs and Liberty in the daily progress reports. The only exception to this is wholesale standalone directory listing (JB) orders. I

44、n the Status Report, FairPoint did not provide any explanation of how it obtained the quoted flow-through results, how they were calculated, or why they are different from the results reported in the daily reports. Second, an order cannot be counted as one that flowed through to provisioning until t

45、he order completes, because it may fall out for manual handling anywhere in the provisioning process. As such, orders that were received in June and flowed through to provisioning completion will make the initial flow-through percentages appear high. Orders that fall out for manual handling will tak

46、e longer to provision and will not be counted in the flow-through percentages until provisioning is complete. As provisioning completes on these orders and they are included in the flow-through results, flow-through percentages will be reduced. Because FairPoint has not explained the flow-through ca

47、lculations quoted in the Status Report, it is possible that they are based on recent orders that were received and do not include the orders that did not flow through and remain to be provisioned. Third, there has been an increase in the number of late retail and wholesale orders over the past few w

48、eeks, which is inconsistent with the vastly improved flow-through percentages reported by FairPoint. Liberty agrees with FairPoint that there will always be some orders in the unsubmitted queue as a normal course of business, and as orders move out of this queue, they will be replaced by new orders

49、coming into the queue. However, FairPoint failed to provide a justification to support that the current number of orders in the queue represents a BAU state. FairPoint should explain the different reasons why orders fall into this queue, how many orders FairPoint has in each category at the end of J

50、une, and the length of time the orders have been in the queue. Without this information, it is not possible to determine whether the number of unsubmitted orders might not be masking other problems. For example, if orders are in the queue due to a lack of the facilities needed to provision the reque

51、sted service, there may be a problem with FairPoints network inventory data, which FairPoint must address internally or with Verizon. Another order flow issue not addressed in FairPoints reports is the ability of the CLECs to have their directory listing orders included in the published directories.

52、 CLECs have complained that they have had trouble making and verifying successful completion of directly listing updates before the directories are published. This is a critical issue for the CLECs, and possibly for FairPoints retail customers as well, which FairPoint must address before it can clai

53、m its order flow process is functioning at normalized levels.Liberty concludes that FairPoint has made progress in order flow. However, even FairPoint admits it has not achieved BAU in all aspects of this area, and some of the evidence it provides for achieving partial BAU, such as for CSRs and flow

54、-through, are suspect. In addition, there are additional areas needing attention besides those noted by FairPoint, such as, overall pre-ordering functionality and directory listing updates.Retail & Wholesale BillingFairPoint reports that it has been issuing timely retail bills since the end of M

55、arch and that as of the end of June it has lowered its rate of bills with known errors to less than one percent, suggesting that its retail billing has returned to normalized levels. FairPoint states that there currently are no wholesale bills with known billing errors but that some wholesale billin

56、g issues show up as billing disputes with specific CLECs and are resolved with each CLEC customer.Libertys Assessment Liberty agrees that, based on the data provided by FairPoint in its daily reports, FairPoint has met its objective of one percent or less of bills with known errors. However, this re

57、sult may be a somewhat misleading based on how FairPoint determines what it considers a “known error.” For example, some FairPoint customers recently received toll bills with the correct dialed numbers and rate for the call but the city name on the bill was shown as a city in the wrong state (e.g. a

58、 call to an 802 number with Austin, Texas as the city name on the bill). FairPoint has told the Staffs that it does not consider these incidences to be billing errors, because the customer was charged the correct rate on the bill. As another example, FairPoint has told the Staffs that it does not co

59、unt cases where customers have ported their service to a FairPoint competitor and but continue to receive bills from FairPoint, if that problem did not result from a known systems error. FairPoint has informed the Staffs that it believes there are relatively few remaining cases of such incorrect continued billing of former customers, but all three state Staffs report