Our leakage performance

We work with Ofwat, our regulator, to set stretching targets for reducing our leakage over a 5-year period. We report annually to Ofwat on how we’re doing against our regulatory target, which is our 3-year average leakage level. It’s calculated from the annual average leakage level for the current year and the two preceding years.

The starting position is the baseline leakage level. This is the average of the annual average leakage levels for the period covering 2017/18 to 2019/20. As part of our Leakage Reporting and Insight Improvement Programme (LRIIP), and to move further towards full compliance with the Ofwat common guidance on leakage, we have incorporated improvements into our leakage reporting methodology during 2021/22. You can read more about this below in the section on ‘Progress on our Leakage Reporting and Insight Improvement Programme’. So that our current performance is reported consistently we have reset the 2019/20 baseline for the regulatory period from 2020/21 to 2024/25, we’ve committed to reducing our baseline leakage level as follows:

We want to be able to report our leakage performance really accurately to our customers but we recognise that comparability of data over time is also important. We are planning to bring forward further data quality and methodology improvements during 2022/23 and to ensure our reported outturn is comparable with the outturn delivered in previous years. These improvements incorporate data from smart meters to inform us how households use water. We’ll restate both our baseline leakage targets and previous years’ performance so that our reporting is both accurate and comparable. Planned changes in reporting methodology and improvements in data quality will be reviewed by an independent third party auditor.

For 2022/23, we have set ourselves an internal leakage target of 531 Ml/d and will track our performance throughout the year. We will provide an update on how we are performing in our next quarterly report.

Highlights from the last year: April 2021 – March 2022

• We met our three-year average leakage performance target of 10.2% reduction against the baseline. Leakage from April 2021 to March 2022 was 605.6 Ml/d.
• To help achieve this, we fixed 61,671 leaks with an average 1,186 leaks a week.
• We have made improvements to our leakage reporting methodology which means that we can report leakage more accurately and with consistency. We have plans to continue to make further improvements so that we can continue to improve our data quality further.

What’s our leakage level?

The graphs below show the estimated amount of water lost from our network this year compared with the forecasts in our leakage reduction plan.

Graph 1: average leakage we’ve achieved each month, compared to the monthly average leakage level we would expect if we deliver our leakage reduction plan

Graph 2: our year-to-date average. This is the average of our monthly leakage levels from April to date compared to our forecast.

The solid blue line – our reported leakage – shows our actual leakage level, based on our daily operational data. The dotted lines show our forecast leakage level if we deliver our plan. Together, these three lines represent three bands:

• The central band (the dotted blue line) shows where we expect leakage to be if everything works out as we initially predicted
• The upper and lower bands (the orange and green dotted lines) reflect the possible impact of risks to our plan. These risks might include adverse weather, or weather that’s milder than we expected. Our plan might also be affected by finding more or fewer large leaks than we predicted.

Leakage tends to increase in winter when colder weather causes pipes to contract, increasing the chance of bursts. As we would expect then, our leakage increased in October and then more sharply in November, as temperatures cooled.

Each quarter, we review the assumptions we use to calculate leakage. These regular updates have been particularly important since the Covid-19 pandemic has changed our customers’ demand patterns for water. This quarter, we revised our night usage calculations, which meant that our calculated leakage level increased by around 16 Ml/d from the start of the year. This is shown in Graph 1 as the solid blue line (reported leakage) crossing above the dotted orange line (upper forecast).

At the end of each year, our final reported annual average leakage number is calculated using the water balance methodology that is set out in Ofwat’s reporting guidelines. We have been making improvements in our methodology to fully align with this guidance. Through our Leakage Reporting and Insight Improvement Programme (LRIIP), we’ve identified and delivered a number of opportunities that have led to improvements in the accuracy of several of the water balance components and as a result our data quality. We will continue to make further improvements as we progress through 2022/23.

As we make our data more robust we need to ensure we report consistently, so we’ve applied the improvements to our previously reported numbers and have reset our 2019/20 baseline (from 671.8Ml/d at AR21 to 674.4Ml/d at AR22).

How much water is that?

A leakage level of 589 Ml/d (our average monthly leakage in March 2022) is about 23% of the total water we supply. It’s equal to 236 Olympic size swimming pools full of water.

How many leaks have we fixed?

• We fixed 61,671 leaks between April 2021 and March 2022 – below the 73,538 that we’d planned
• That’s 1,186 leaks fixed on average each week
• All of our leakage reduction activities (including our pressure management and water efficiency initiatives) saved 438 Ml/d of water lost

We’ve fixed fewer leaks than we’d planned to due to the challenges of Covid-19 and difficulty in recruiting, which have both continued to impact our resources. By working smarter, using innovative technology and prioritising larger leak repairs, we have improved – we fixed more leaks than reported for the previous year(60,286).

Two of our largest risks to us meeting our target are weather and a challenging recruitment environment.

Having accurate and clear leakage reporting is an essential part of managing and reducing leakage efficiently, to ensure that we make best use of available data and resources. Our Leakage Reporting and Insight Improvement Programme (LRIIP) is designed to improve confidence in our data and processes, improve resilience, provide greater accuracy and consistency of reporting through assurance and demonstrate how we will use insight to effectively deliver the improved leakage performance expected by our customers and stakeholders.

The LRIIP comprises 21 key deliverables, across five themes related to:

• Data quality improvements
• Simplify/digitalise the water balance
• Create a self-assurance culture
• Implement assurance
• Enhance the capability and capacity of the team.

18 of the 21 deliverables are complete, one requires ongoing monitoring and the other two are on track to be delivered by their due date. We reported on the delivery of LRIIP to Ofwat on 29 April.

Highlights

• Embedding our compliance, assurance and risk management framework, which will help to improve confidence in our quarterly leakage reporting
• Scoping out the work required to update our models to make better use of smart meter data to understand our customers’ water use
• Tested automated digital products for monitoring and validating how water flows in our network
• Taking the learnings from the annual leakage reporting process to scope and prioritise our plans for the year ahead, which will build on the same five themes established for LRIIP and put more emphasis into sharing insights with our field teams who maintain the network and repair leaks and our customer teams who manage accounts and track billed consumption.

• Jargon buster
• Glossary of Terms used in this report