Pros abandon percents of SIV

Professionals abandon %s of System Input Volume

Now 124 Supporters from 22 countries;

Register your support if you wish to join them

Einstein was quoted as saying that ‘everything should be as simple as possible, but no simpler.’ H.L Mencken’s aphorism stated that ’for every complex problem there is a solution that is simple, neat and wrong.’ Green considers that ‘Our world is complex, but we seek simple solutions that hide or ignore the complexity. The result is that our actions often have unintended side effects. These produce unwelcome trends, accidents and disaster’.

Over the past 25 years, an ever-increasing number of professional water conservationists have clearly recognized, from personal experience and analysis of managing diverse water supply systems, that expressing the volume of Non-Revenue Water and its components as a simple percentage of System Input Volume is usually fundamentally misleading, in terms of:

  • Setting targets and tracking changes in management performance in individual Utilities
  • Comparisons of technical efficiency of management between Utilities
  • General conclusions regarding management of NRW and its components

Water balance with several SIV options showing Water Exported and Water Supplied (WS)

Source:- Paper 2014G, based on Fig.7, Alegre et al Performance Indicators for Water Supply Systems, 2000

A reminder of some flaws in using % of System Input Volume, and % of Water Supplied,

  • Countries and Utilities define SIV differently; which System Input Volume was used ?
  • If water is exported, NRW or Leakage volume as % of SIV will be smaller than NRW or Leakage volume as % of WS; how do you know which figure is being quoted?
  • As sum of all components of % SIV or % WS must equal 100%, this is a Zero-sum calculation; if both consumption volume and NRW volume are reduced, Consumption % and NRW% cannot both reduce in the same year, so true progress cannot be assessed.
  • Utilities with a high consumption base can more easily achieve a lower % of SIV or % of WS than Utilities with a low consumption base, for the same volume of leakage
  • Hourly and seasonal variations in consumption cause large hourly and seasonal variations in leakage as % of SIV or as a % of WS, even when daily leakage volume is constant

Alternative KPIs which do not suffer from the deficiencies of % of SIV and % of WS have increasingly become available since 2000 – for example, NRW as % of SIV by Value; Apparent Losses Index (ALI) and Infrastructure Leakage Index (ILI).

Accordingly, since 1995, an increasing number of national organisations and countries have abandoned or downgraded the use of % by volume of SIV or WS for these purposes, and began to use more  appropriate and meaningful performance indicators, with more widespread beneficial results in sustainable water loss management. The world of water loss management is gradually changing for the better, but as always occurs with progress, some are leading from the front and others are lagging behind.

Even in countries where national organisations and Utilities have ceased to use percentages by volume for more than 10 years, it regularly occurs that some person or organization without the motivation, background or experience to understand the problem, or who considers they can ‘communicate’ better, or who simply wishes to generate headlines to emphasise perceived good or poor performance, will start quoting percentages and frequently draw misleading conclusions. In the internet age, this is then rapidly picked up, requoted by others and repeated. It is then necessary to spend time trying to inform the users of the unintended consequences and misunderstandings they have caused. Very few people welcome being advised after the event that they did not properly understand the issue they had commented on or the problems they had caused.

As a positive step to try to help people and organisations interested in NRW, who do not understand why percentages by volume can be so misleading, LEAKSSuite has decided;

a) to create this webpage ‘Professionals abandon Percentages of System Input Volume” (PaP)* in the KPIs Info-Hub to provide multiple readily accessible international references, papers, presentations and blogs (listed below) from around the world which confirm how misleading %s of System Input Volume can be, and

b) to encourage individuals, Companies and Organisations from many different countries to register your support of this initiative so that they can be included in a list of Water Professionals who have already abandoned the use of %s by System Input volume and % of Water Supplied as meaningful KPIs.

*The use of the acronym PaP is not accidental. The dictionary defines ‘pap’ as ‘material lacking real value or substance’.  

Dec 29th 2016

International Sources which highlight problems with using % of SIV or % of WS

In a hurry? Check-out a few PapNotes:

From the ‘Concepts’ Webpages of LEAKSSuite:

Free LEAKSSuite Software (for PCs only, not mobile phones or tablets)

InternationalMay 2017Presentation NRW as a % of System Input Volume just doesn’t work! PI 2017 Conference of the IWA-benchmarking and performance assessment group in Vienna, Austria, May 15-17 2017Presentation 2017C
Europe and InternationalOct 2016Presentation at Water Ideas 2016, BolognaRegulation in water loss management and
benchmarking to achieve efficiency: an
International perspective.
Presentation 2016P
United KingdomMay 2016BlogEngland & Wales Water Company Leakage Rises – or Falls? Consumer Council for Water Report confuses interpretation of leakage statisticsInterpreting Leakage in England & Wales
InternationalFeb 2016Presentation at IWA Water Loss BangaloreLeakage Performance Indicators ‘Fit for Purpose’: updates from Austria, Bulgaria, Croatia, Germany, South Africa (KwaZulu-Natal), Canada (Quebec), RomaniaPresentation 2016D
U.S.A.Feb 2016AWWA Manual of Water Supply Practices M36.
Water Audits and Loss Control Programs 4th Ed.
Table 1.1:
Non-preferred term: % of System Input Volume to measure Water Loss Performance.
Preferred term: Suite of Key Indicators for Water Losses as outlined in AWWA Audit method in Chapter 3.
Reason: A %-based expression obscures the underlying components of water loss and impedes realistic solutions based on system specifics.

U.S.A.Dec 2016Presentation, N. American Water Loss 2015 Atlanta Benchmarking Water Loss Performance: The Death of Unaccounted Water. Presentation 2015Q
InternationalJune 2015Article, Global Water IntelligenceHow can we optimise the way we measure leakage?Article 2015K
United KingdomJune 2015C.I.W.E.M.
Policy Position
Statement Update
CIWEM continues to recommend that leakage should definitely not be quoted in terms of percentages of system input volume; it is misleading for comparisons because of differences and changes in consumption, and it is a zero-sum calculation which cannot identify true reductions in leakage and consumption in the same year.CIWEM PPS‏
U.S.A.Feb 2015AWWA Flyer ‘Best Practice in Water Loss Control: Improved Concepts for 21st Century Water Management Extract from Flyer: Using percentage indicators to assess water loss in distribution systems gives a misleading and unreliable measure of utility performance because a percentage indicator
• is greatly affected by changing levels of customer consumption
• cannot distinguish among the specific components of non-revenue water occurring in a distribution system
• reveals nothing about water volumes and associated costs (the two most important factors in assessing water waste within a distribution system)
AWWA Improved Concepts for 21st Century Water Management
EuropeJan 2015EU Reference Document Good Practices on Leakage Management WDF CIS WG PoM:
Main Report
Leakage expressed as a % of System Input Volume (SIV) is simple and easy to calculate. However, it has several limitations in interpretation which have led some Member States to stop or to reduce the use of % of SIV as a leakage performance indicator. For example, it can result in substantial under- or over-estimates of true achievements in reduction of leakage volume (see e.g. the Belgian and Bulgarian case study accounts). This is because % of SIV is a ‘Zero-sum’ calculation, which is unable to identify
actual decreases in both consumption and leakage volume in the same period. See also Appendix B.2 ‘Unintended consequences of using % SIV as a leakage PI’
Good Practices on Leakage Management – Main Report_Final.pdf
InternationalOct 2014Blog and free softwareExplains what is meant by a Zero-Sum Leakage Indicator with examples of ‘Play the % Game’ free softwareZero-Sum Leakage PI
InternationalMay 2014BlogThe Water Conservation World Cup – using a cartoon approach to explaining the problemThe Water Conservation World Cup
United KingdomApril 2014Presentation, Smart Water Systems Conference, LondonBest practice international approaches to monitoring improvements in NRW and leakage management Presentation 2014 K

Presentation 2014K
Austria and EuropeApril 2014Paper, IWA Water Loss 2014 Vienna14 Years’ Experience of using IWA Best Practice Water Balance and Water Loss Performance Indicators in EuropePaper 2014 G
Presentation 2014F
Europe2014European Environment Agency Technical Report No 5, 2014
Performance of Water Utilities beyond compliance
Extract from Appendix 4: Targets for water loss (non-revenue water) are often expressed as % of system input volume. Seen as an indicator, this is, however, influenced by the water consumption, itself, in the sense that a lower (sustainable) water consumption for the same water loss in the system will result in a higher indicator value. Publications Office of the European Union, 2014
ISBN 978-92-9213-428-0
ISSN -1725 - 2337
Germany and AustraliaNov 2013BlogPercentages don’t work for tracking NRW and Leakage Reduction: two examples from Germany and an example from AustraliaLeakage Tracking – Percentages
InternationalJun 3013Article in Sustainable Cities, building for the future’‘Accounting for Water Leakage and Measuring Performance’. Climate Action, United Nations Environment Programme (UNEP), June 2013. ISBN: 978-0-9570432-8-2

Article 2013H
United KingdomMay 2012Article in BHS Circulation MagazineSome things Hydrologists need to know about Water Utility Leakage in the UK’. Article in British Hydrological Society ‘Circulation’ No 113, May 2012Article 2012Q
AustraliaJun 2010Article in ‘Water 21’ magazineManaging Australia’s real loss levels. Water 21, June 2010Paper 2010G
AustriaDec 2009Article in ‘Water 21’ MagazineAustria’s new guidelines for water losses for water lossesPaper 2009T
MalaysiaJul 2008Ranhill Utilities Berhad and US Aid PublicationThe Manager’s Non-Revenue Water Handbook: a Guide to understanding Water Losses. NRW has traditionally been expressed as a % of input volume. Although this is preferable to setting no targets at all, it is misleading as a PI because it favours Utilities with high consumption, low pressure and intermittent supply. ….. as a measure of the Utility’s year on year financial performance ….it should be expressed as the value, not the volume, of water lost. ’
InternationalSep 2007Paper at IWA Water Loss BucharestWater Loss Performance Indicators. IWA International Specialised Conference ‘Water Loss 2007’, Bucharest, September 2007. Conference Proceedings (3 vols) ISBN 978-973-7681-24-9; Session B2 of Volume 1Paper 2007T
U.S.A.2007AWWA Research Foundation: Evaluating Water Loss and Planning Loss Reduction StrategiesmarcoAWWA Research Foundation
American Water Works Association
IWA Publishing
Australia2004WSAA facts 2004: Water Services Association of Australia Inc., ISSN 1328-2751Section 5, Leakage. ‘Until 2003 WSAA published water loss as a percent of water supplied. Whilst easily calculated this does not measure the efficient management of leakage in a distribution system because it
• Does not distinguish between apparent and real losses: and
• Is strongly influenced by consumption (and changes in consumption)’
GermanyMay 2003DVGW W392 Guidelines on Water Loss Reduction German Association of the Gas and Water Sector ‘Expressing real losses as a % of system input volume is unsuitable as a technical performance indicator as it does not reflect any of the influencing factors. Systems with higher system input volumes (e.g. urban systems) will automatically have an (apparently) lower level of water losses if expressed in percentages. Systems with low water consumption (e.g. rural systems) will show high percentage figures of real losses. Therefore comparisons using percentages will always favour systems with high system input
Australia2003Paper in Water Science and Technology: Water Supply Vol 3 No 1-2 pp153-161. Applying the IWA Approach to water loss performance indicators in Australia.
Water Services Association of Australia stopped using % of System Input Volume as a KPI in 2001 and began to use ILI and litres/service connection/day instead.
InternationalAug 2002Paper and PresentationInternational Report on Water Losses Management and Techniques: Report to IWA Berlin Congress, October 2001. Water Science and Technology: Water Supply Vol 2 No 4, August 2002Paper 2002N
Presentation 2002M
InternationalDec 1999Paper in IWA ‘AQUA’ JournalA Review of Performance Indicators for Real Losses from Water Supply Systems. AQUA, Dec 1999. ISSN 0003-7214Paper 1999M
United Kingdom1998Report, Financial Times EnergyManaging Water Leakage: Economic and Technical Issues. ‘Expressing volumes or rates of leakage as a percentage of system input is the worst, and potentially most misleading, of the traditional performance indicators’ISBN 1 84083 011 5
United Kingdom1997Report on Leakage and Water Efficiency
ISBN 1 874234 32 9
Office of Water Services (the UK Economic Regulator): OFWAT does not favour % of water into supply as a measure of leakage. This can be misleading. For instance, a reduction in the volume of water into supply may occur as a result of a water company’s successful promotion of water efficiency by its customers, but would make its leakage problem appear to worsen.