UK Utilities Residential Cost-to-Serve Benchmarks and Metrics

SUMMARY

Ths report details the leading benchmarks and performance measures for the 6 leading UK suppliers in evaluating their cost-to-serve. Using Datamonitors UK Residential cost-to-serve model, the report performs further analysis to assess the key levers in controlling costs. The report works through a number of cost saving strategies and provides additional performance case studies.

Scope of this report

• Benchmark data on key performance metrics in assessing cost-to-serve
• Figures on savings calculated, using cost-to-serve model i.e. shifting customers to lower payment channels, to reducing average call handling times
• Sensitivity analysis, determining the prime levers across all 600+ metrics to achieve a 10% reduction in cost-to-serve for a leading UK supplier
• Datamonitors appraisal of operational performance across the 6 leading UK suppliers across their service related supply costs

Research and analysis highlights

Billing and the contact centre show the widest spread of costs for UK suppliers in their cost-to-serve per customer

There is a £6 variance between the highest and lowest cost-to-serve per customers among the UKs 6 leading suppliers

Staffing levels and billing queries are the levers that are more sensitive in delivering a 10% saving in cost-to-serve for one of the leading UK suppliers

Key reasons to read this report

• Discover which UK suppliers are outperforming or underperforming in their attempt to mamange their cost-to-serve
• Understand which cost drivers impact on cost-to-serve and by what degree
• Be informed on who and how leading UK suppliers are managing to deliver best practice under the four categories of Datamonitors cost-to-serve model

TABLE OF CONTENTS

CHAPTER 1 EXECUTIVE SUMMARY

• Determining cost-to-serve requires a clear structure and logic to set precise definitions that can then be measured
• Maintenance and Labour costs across the four sub components account for the majority of UK cost-to-serve
• The model can be used to identify cost savings and the levers needed to implement them
• Reducing the moving read failure rate reduces the cost-to-serve by £152,463, a 0.14% saving overall
• Key cost drivers can be identified through inductive analysis
• Key cost drivers can be identified through inductive analysis

CHAPTER 2 MODEL STRUCTURE AND LOGIC

• Cost to serve has increased in importance as fuel costs have risen in the UK
• Specific costs underlie each of these operational functions, e.g. wages and system maintenance
• Each of the cost sub-categories is fed by a set of performance measures, KPIs and cost drivers

CHAPTER 3 METHODOLOGY

• The model was constructing in three phases spanning the whole of 2004
• Datamonitor undertook three phases of investigation, research and analysis to ensure a robust methodology for the model
• Datamonitor undertook three phases of investigation, research and analysis to ensure a robust methodology for the model
• Phase 1 initiated the project with in-house conceptual model construction and data mapping
• Phase 2 involved further secondary and primary research to ascertain key metrics and operational drivers
• Phase 3 involved final data analysis in order to validate key metrics and drivers; as well as beta testing of model outputs

CHAPTER 4 BENCHMARKING AND COMPANY PERFORMANCE

• The model allows all of the 600-plus metrics to be benchmarked against the 6 leading UK suppliers
• SSE has the lowest cost-to-serve and npower the highest
• There is a £6 variance between the highest and lowest cost-to-serve among the UKs 6 leading suppliers
• Maintenance costs form the majority of UK suppliers cost-to-serve
• Service centre staffing is the largest individual component of cost-to-serve
• Service centre staffing is the largest individual component of cost-to-serve
• Datamonitors appraisal of operational performance is based on 4 performance categories
• EDF Energy achieves the lowest metering costs per customer, SSE the highest
• SSE achieves the lowest billing costs per customer, npower the highest
• SSE also achieves the lowest payment costs per customer, and npower is again the highest
• SSE also achieves the lowest payment costs per customer, and npower is again the highest

CHAPTER 5 DEDUCTIVE ANALYSIS

• The model can be used to identify cost savings and the levers needed to implement them
• Metering (npower) - reducing the moving read failure rate would deliver a 0.1% saving across npowers cost-to-serve
• Lowering the moving read failure rate reduces call volumes, so 6 fewer customer service agents are needed
• The biggest saving is made in the call centre
• Reducing the moving read failure rate reduces the cost-to-serve by £152,463, a 0.14% saving overall
• Billing (EDF Energy): Cost-to-serve can be reduced by 1.9% by increasing by 11 percentage points the number of customers who pay on receipt of their first bill
• This reduces by the total number of bills sent out in a year by 1.74 million
• Fewer bills and more prompter payment reduces staffing levels by 75 FTE
• The largest savings are gained through reduced staff numbers, postage and packaging costs and reduced high risk debt
• Increasing payment without reminders reduces the cost-to-serve by £1.69m, a 1.94% saving overall
• Payment (Powergen) - Reducing the proportion of customers who pay by cheque by 10 percentage points would deliver a £1.8 million saving
• Redistributing the method of payment metrics increases the Total number of direct debit transactions by 5.3m
• 101 customer service agents could be removed
• Total service contacts would fall by 0.1m, with an impact on staffing levels and wages
• Call volumes would also be affected
• The three largest savings come under Total cost of wages for call centre agents, Total cost of credit staff wages and Total cost of inbound calls paid to telecoms provider
• The overall cost saving amounts to £1.8m, £0.29 per customer - a saving of 1.40%
• Contact centre (Scottish Power) - increasing call automation and reducing handling times would reduce costs by 6.8%
• The average length of an inbound call in the call centre falls by 2.09 minutes
• The volume of calls into the contact centre across all queries falls by 4%
• 76% of the total savings are attributed to a reduction in contact centre FTEs
• A 4 percentage point increase in call automation and an average 2 minute reduction in call times saves £1.42 per customer

CHAPTER 6 INDUCTIVE ANALYSIS

• Key cost drivers can be identified through inductive analysis
• The issue of scale is redundant when performing sensitivity analysis across the leading UK suppliers, who all have over 4 million accounts
• Sensitivity analysis on Centricas KPIs shows that there are five main ways in which to reduce cost-to-serve by 10%
• The KPI metric most sensitive to a 10% change is the number of estimated first bills sent to regular customers each year
• AMR, staffing and number of bills sent are the key sensitive KPI levers that can deliver a 10% saving in Centricas cost-to-serve
• CSA average annual wages are the most sensitive to a 10% cost reduction
• Staffing levels and billing queries are the levers that are more sensitive in delivering a 10% saving in Centricas cost-to-serve
• BEST PRACTICE KPIS AND RECOMMENDATIONS
• The following section investigates the best practice supplier in each of the four service functions
• EDF Energy can thank its internal metering company for its superior performance
• EDF Energys performance in the key metering performance measures is better than the UK average
• EDF Energys success is based on benchmarking and staff training
• In-house metering has allowed EDF Energy to invest, producing improved performance
• Powergen has reduced labour costs by investing in its billing system
• Investing in IT reduced costs and improved customer service
• Powergen exceeds the UK average in all of the key billing performance measures
• SSEs relies for its success on channelling customers onto preferred payment channels
• SSEs performance exceeds the UK average in the key payment performance measures
• SSE cuts costs by encouraging customers onto low cost payment channels and by improving bad debt collection
• SSE has managed its customer base onto lower payment channel options and supports this with technological innovation
• Centrica plans to improve customer satisfaction by raising performance in its contact centres
• Centricas performance on the key contact centre performance measures exceeds the UK average
• Datamonitors research provides recommendations on improving contact centre performance
• European utilities can be divided into three categories: premium providers, price players and the herd
• British Gas has invested in order to maximise its CIS and CRM capabilities and reduce its cost-to-serve

CHAPTER 7 APPENDIX

• Future readings
• SPP writing team
• How to contact experts in your industry

List of Figures

Figure 1: The cost-to-serve model - top level

Figure 2: Share of overall cost, broken down by sub-component

Figure 3: The service centre and billing costs show the widest spread of costs for UK suppliers in their cost-to-serve per customer

Figure 4: Actual and percentage change of overall, metering and contact centre costs

Figure 5: Total cost differential between original and new final values for npower

Figure 6: Sample from the model

Figure 7: Percentage change to deliver a 10% saving for overall cost-to-serve

Figure 8: Comparing Centrica against the UK average across five service criteria

Figure 9: The cost-to-serve model - top level

Figure 10: Detail of metering costs, KPIs and cost drivers

Figure 11: Actual view of three of the four service areas

Figure 12: The cost-to-serve model timeline

Figure 13: Cost-to-serve per customer for the leading 6 UK suppliers and their average

Figure 14: The relative cost-to-serve per customer for the leading 6 UK suppliers against the average of the 6 leading UK suppliers

Figure 15: Share of overall costs broken down by labour, non-labour and maintenance costs

Figure 16: Share of overall costs broken down by sub component

Figure 17: Cost range of sub-components of cost-to-serve of UK suppliers

Figure 18: Performance ratings for each company across four service functions

Figure 19: Metering performance ratings

Figure 20: Billing performance ratings

Figure 21: Payment performance ratings

Figure 22: Customer service performance ratings

Figure 23: Metrics impacted by reducing the moving read failure rate from 12% to 5%

Figure 24: Breakdown of cost savings on a increased performance on moving read failure rate for npower from 12% to 5%

Figure 25: Actual and percentage change of overall, metering and contact centre costs

Figure 26: Total cost differential between original and new final values for npower

Figure 27: Sample from the model

Figure 28: Metrics altered by improving first time bill payers by 11 percentage points

Figure 29: Sample from the model - fewer bills and prompter payment

Figure 30: Breakdown of cost savings on a increased performance on % of customer paying on first bill for EDF Energy, from 59% to 70%

Figure 31: Actual and percentage change of overall, billing, payment and contact centre costs

Figure 32: Total cost differential between original and new final values for EDF Energy

Figure 33: Sample from the model - EDF Energy

Figure 34: Metrics affected by shifting other payments to direct debit and online payment methods

Figure 35: 101 customer service agents could be removed

Figure 36: Total service contacts would fall by 0.1m, with an impact on staffing levels and wages

Figure 37: Call volumes would also be affected

Figure 38: Breakdown of cost savings from a 10 percentage point reduction in cheque payment, with a 7 percentage point increase in DD and 3 percentage point in online payment

Figure 39: Actual and percentage change of overall, billing, payment and contact centre costs

Figure 40: Total cost differential between original and new final values for Powergen

Figure 41: Sample from the model - Powergen

Figure 42: Metrics impacted by increasing the proportion of automated calls by 4 percentage points and reducing query handling times by an average of 2 minutes

Figure 43: The volume of calls into the contact centre across all queries falls by 4%

Figure 44: Breakdown of cost savings from increasing the proportion of automated calls by 4 percentage points and reducing query handling times by an average of 2 minutes

Figure 45: Actual and percentage change of overall and contact centre costs

Figure 46: Total cost differential between original and new final values for Scottish Power

Figure 47: Sample from the model - Scottish Power

Figure 48: Fixed costs become insignificant at high customer levels

Figure 49: Sensitivity Analysis - KPIs

Figure 50: Sensitivity analysis of leading KPIs by delivering a 10% decrease in Centricas cost-to-serve

Figure 51: Sensitivity Analysis - Cost Drivers

Figure 52: Sensitivity analysis of leading cost drivers by delivering a 10% decrease in Centricas cost-to-serve

Figure 53: Comparison of key metering performance measures for EDF Energy against the UK average

Figure 54: Comparison of key billing performance measures for Powergen against the UK average

Figure 55: Comparison of key payment performance measures for SSE against the UK average

Figure 56: Centricas performance on key contact centre performance measures exceeds the UK average

Figure 57: Comparison of key contact centre performance measures for Centrica against the UK average

Figure 58: European service metrics for the three types of utility to 2008

Figure 59: British Gas drives customer planning through data-based segmentation