Mobile menu icon
Skip to navigation | Skip to main content | Skip to footer
Mobile menu icon Search iconSearch
Search type

Department of Computer Science

Intelligent software gets the right price

Researchers have developed techniques for price optimisation that enabled the creation of the first Intelligent Pricing Decision Support Systems (IPDSS) for the retail and petroleum sectors.

Graphic showing figure of £13.2 million.

KSS Fuels

KSS Fuels' revenue in 2012 was £13.2 million.

Graphic showing figure 60 under image of man and woman

KSS Retail

KSS Retail has more than 60 employees in the UK and the USA.

Spin-out companies are now the leading global suppliers of pricing software, with combined revenues of £18.2 million.

Correctly pricing goods and services – finding the balance between keeping customers happy and generating profit – is integral to commercial success. Before the widespread use of computers in the 1980s, picking the right price was a bit of a dark art.

However, early computerised price optimisation failed to deliver on its promise for some sectors, in particular the highly dynamic and competitive retail and petroleum sectors.

Early attempts to develop robust methods for optimising prices in these sectors proved unsuccessful because academics struggled to model their highly complex sales environments.

Researchers at Manchester made a breakthrough that allowed them to pioneer the first systematic pricing method, leading to the creation of Intelligent Pricing Decision Support Systems (IPDSS) for both retailers and petrol companies.

Software products incorporating the intelligent techniques were launched by KSS Ltd – a spin-off company established from the research group's intellectual property.

Graphic showing 10,000 and a petrol pump


10,000 petrol stations are using PriceNet software.

In 2007 KSS demerged into KSS Fuels (now Kalibrate) and KSS Retail. These two companies market PriceNet and PriceStrat software respectively. Today, PriceNet software increases oil company annual profits by up to £100 million per annum; PriceStrat raises the combined profits of its convenience and grocery store customers by up to $150 million.

Today over 400 companies in 80 countries use PriceNet and PriceStrat, including household names in the UK and the USA like Kroger, Tesco, 7-Eleven, O'Reilly, and Rite Aid.

KSS Fuels and PriceNet helped us revise our pricing process to be more accurate and timely in response to competition.

President of Miller Oil, USA

Impressive earnings

Since additional refinements to its software, KSS Fuels has become the leading global retailer of fuel pricing software with an annual revenue of around £15 million. The company employs 100 staff.

Graphic showing a range from 1,000 to 2,000


1,000 to 2,000 convenience stores worldwide use PriceStrat.

Since additional refinements to its software, KSS Fuels has become the leading global retailer of fuel pricing software with an annual revenue of around £15 million. The company employs 100 staff.

In 2012, KSS Retail, now owned by Dunnhumby Ltd, reported revenues of £6 million and a profit of £3 million. It is the premier global provider of price intelligence and optimisation software designed for grocery and convenience stores. The company employs 60 members of staff.

In 2013, Kalibrate floated on London's AIM market, with 33,227,848 Ordinary Shares issued at Admission, giving the Group a market capitalisation of £26.2 million at the Placing Price.

By using KSS PriceStrat, we will have advanced insight into our pricing and promotional decisions and will be able to ensure that we are priced right, on the right items, for our customers.

Joe Hanson, Vice President of Operations at Yoke's Fresh Food Stores, USA


The initial applications of pricing software were limited, especially for highly competitive businesses with complex dynamics and frequent fluctuations in price.

The team of researchers explored methods to create an intelligent system which would work across a variety of sectors.

Research into optimisation methods developed non-linear adaptive prediction models and learning algorithms which could model the demand for products and identify price-sale relationships from data. The researchers also used non-linear optimisation techniques to identify prices that would best achieve multiple pricing targets as well as meet strategic long-term goals.

With a generic methodology in place, the researchers created then sector-specific variants based on an industry's particular pricing dynamics.

Commercial partner organisations in the retail, petroleum, banking, and telecommunication sectors tested and evaluated pilot software.

Refinements based on recursive least square algorithms helped to overcome the problems of demand modelling and forecasting when historical price-sale data was limited.

The team

  • Madan Singh
  • Jean-Christophe Bennavail
  • Xiao-Jun Zeng
  • Nathalie Cassaigne