But they have their limitations. They cannot, for example, tell what percentage of customers browse without purchasing.
An American computer installation company, Datatec, believes it has a product to fill the gap. Six high-street UK chains are now testing Shoppertrak, already used in the United States by retailers such as Saks Fifth Avenue and K-Mart.
Shoppertrak counts customers with an infra-red beam system as they enter and leave the store. Because this is mounted vertically it is, says its manufacturers, more accurate than some horizontal systems that can, for example, mistake three people as one.
In addition, it cuts off at a given height to exclude articles not defined as potential shoppers, notably children and trolleys. It can also distinguish direction.
Most importantly, Shoppertrak can be linked to an electronic point of sale (EPOS) system so the information can be merged with sales data and converted into management intelligence to form a more meaningful basis for planning.
For example, the percentage of customers who make a purchase - the conversion rate - will not be constant. By recognising this, sales staff can be employed more tactically to convert browsers into buyers.
The effect can be enormous. In 78 stores in the US, Datatec found the average conversion rate to be 30 per cent. The company estimates that increasing this by 3 per cent is equivalent in profit terms to a 10 per cent increase in sales because it can be managed at little extra cost.
The usefulness of conversion rate does not end there. It is also, Datatec says, a better means of comparing stores within a chain.
Customer traffic can provide an alternative measure of the effectiveness of advertising or window displays. Sales information might suggest that a promotion is failing, but it could be that although a greater number of people are being drawn into the shop, the problem is one of conversion. The product they are specifically looking for may be out of stock, for example, or queues may be putting off potential new customers.
This type of investigation can be extended by positioning Shoppertrak modules within the store to measure traffic into specific departments, aisles or even the entrance to a fitting room.
'This may all sound very obvious, but when you start measuring it can make a big difference,' said Joanna Schuller, general manager for Shoppertrak in the UK. She estimates that managers' approximations of traffic through larger stores are often completely inaccurate.
Shoppertrak is priced by the unit, at pounds 800 per 16 inches. This includes installation, training, two months' information analysis and a link into a PC, which will give raw statistics.
Linking up with an EPOS system for more sophisticated analysis is extra.
For this, Datatec has an arrangement with IBM, which has developed a software package that can be adapted for any EPOS system. IBM is also working on software to analyse the Shoppertrak data without an EPOS system, through a conventional PC. This will soon be tested by two retailers and a major bank.