A growing number of companies are realising that in business, comparing chalk with cheese is not only a fruitful endeavour, but an increasingly necessary one.
Dallas-based Southwest Airlines used to have a problem. Most of its aircraft were on the ground between flights for an average of 40 - expensive - minutes. Although it had benchmarked its refuelling processes against other airlines, it found it was already one of the leaders. Some sort of quantum leap for improvement was needed, so it looked outside the airline industry for the most efficient refuellers in the world - Formula One racing. Adopting the turnaround processes used during pit stops, Southwest Airlines can now refuel an aeroplane in 12 minutes.
The tool which Southwest Airlines and many other companies are using to achieve this type of leap forward is benchmarking. It's an easy enough concept when that means measuring yourself against your direct competitors, but increasingly, companies are looking further afield and realising that to be the best means measuring yourself against the best in whatever sector and in whichever country they may be found. It's an issue that is of vital interest to British businesses. A recent report from the Confederation of British Industry (CBI) put the increase in UK GDP were UK manufacturers to achieve US productivity levels at a staggering £60 billion. The report, Fit for the Future, also suggests that if the UK economy as a whole adopted US best practices, the increase in GDP could be around £300 billion.
And yet still all too few British companies have taken this lesson on board. In a report at the end of September, Partnership Sourcing - a joint venture between the CBI and the DTI - highlighted how few British companies have adopted benchmarking techniques. Of the 600 companies surveyed, 88% were aware of the concept, yet only 62% practised it. The take-up was particularly low among small and medium-sized enterprises.
As a management tool, benchmarking's canonisation began in the '70s with Xerox in the US attempting to beat off Japanese competition. The definition of David Kearns, former chief executive of Xerox, is still the best starting point: 'Benchmarking isn't just the hard measure of where someone else is, it's how they are doing it ... how the competition engineered the product to lower manufacturing cost.' But by benchmarking against best practice in other sectors and other countries, against the ABBs, the Motorolas, the Fedexes of the world, rather than just remaining within the industry, some companies are managing improvements in productivity and profits by leaps and bounds.
The mechanics of benchmarking are nothing new. The Greek philosopher Aristotle benchmarked various currencies to the Athenian one in the 4th century BC, and at its most basic level that is what a stock exchange is - a benchmark of what a company is worth. Its value to a chief executive is that benchmarking along with business process re-engineering, Total Quality Management and IT, is a tool, albeit a powerful one, that can be used for improvement. Michael Useem, professor of business management at Wharton, explains that there are two main reasons for benchmarking.
First to establish best practice, and second, to establish norms within the industry, for example pay levels. The most basic question for a company is how to get the benchmarkable data, which, as Mark Czarnecki, president of Houston-based consultants The Benchmarking Network, points out, 'falls into the grey area between the secret formula and public record'. Certainly Paul Ruggier, group quality director of Consolidated Finance Insurance at GE Capital, admits that it is 'the hardest task we have'. No one seriously considers industrial espionage, although one consultant dismisses what he calls 'competitive analysis' as 'low in quality'.
Essentially there are four methods that firms use to source this information.
First, and most obviously, as Edward Collier, partner at Mercer Management Consultants, advises, use the information that you already have. 'Don't discount the information that is available for free - from the press, analysts' reports, annual reports and so on.' Second, listen to your people.
'With the service industries perhaps more than the primary or secondary industries, there is a high degree of staff turnover. If people move they bring new practices,' says Tony Burgess-Webb, head of international for public relations agency Hill and Knowlton. He gives the analogy of the early days in Silicon Valley when rapid turnover led to a swift dissemination of information.
Third, use consultancies. 'In the narrowest scenario take executive compensation.
If you go to a consultancy it will sell you that information without naming names,' says Useem. More valuably, consultants can help you work round what seem like insurmountable problems. Werner Kreuz, European partner in charge of benchmarking for consultants AT Kearney, cites a two-part benchmarking study he was asked to do for a major British steel manufacturer.
The first part was market benchmarking across 17 European countries. 'It was easy to summarise the criteria on which decisions were based, even if the findings were different for each country.' The difficulty was part two - a cost-chain analysis for the company's nine major competitors.
Naturally enough if the consultants had asked for the cost chain of each of the firms, 'they would have kicked us out', but Kreuz identified that the yield was the critical factor in the cost chain, which the companies were, he says 'proud to tell us'.
Fourth, ask others how they went about it. As Czarnecki points out: 'To talk of a payroll process as secret is foolish.' The question is often raised why companies would share data. Malcolm Graveling, partner at AT Kearney, sardonically suggests a Baldridge award syndrome - referring to the Malcolm Baldridge National Quality Awards in the US - where the winners want to show off why they are the best. It's good PR for them. Rob Garner, head of Rob Garner Ltd, who has more than a decade's experience with continuous-improvement programmes at companies like ICL and Price Waterhouse, cites the construction company Mowlem, which is happy to share on the principle that by the time others have picked up on what it is doing it will have moved on. Others prefer to do it anonymously under the aegis of organisations like Price Waterhouse's Global Benchmarking Alliance, which for the last three years has been working with 60-70 global companies - like British Petroleum, BT and MCI.
The most plausible answer, however, is that companies are selfish. They do it for what they can get out of it. Kreuz relates how ABB, in wanting to install a purchasing information system, set up what it calls a 'tiger team' to visit companies like Kawasaki Heavy Industry and Boeing, to see how others had done this within a decentralised organisation. The pay-off for the participants was that on completion they got to examine what ABB had developed.
That companies have to benchmark cross sectorally to compete there is no doubt. We have moved a long way from Du Pont in the early '80s which only used to benchmark how far ahead of its competitors it was. Mark Rodriguez, vice president of consultants ASM, follows the financial sector and admits that even banks - often the more conservative sector of the economy - realise they have to catch up and that to compare themselves solely with other banks is folly. When trying to convince clients of the value, he turns the question round: 'If I am General Motors, what can I learn from a bank? Well, for many years now, both Ford and GM have earned huge revenues from fleet management of leasing and hire cars. If I am in that business, who am I leasing to? I have to learn how to segment the price on loans, evaluate and score customers. I have to look at the consumer banks.' While the more innovative companies have generated the impetus themselves, increasingly, pressure for change has been from the customer. With reference to call centres and customer helplines, John Gallagher, European director of market research at IBM, points out: 'The customer has a set of expectations and regardless of whether he's calling a bank, to order a computer, or calling about an auto offer, we don't have a different personal reaction.' This mirrors the findings of the Henley Centre's Teleculture Futures report published last year. 'Even if you are at the top of your class, if you don't believe that you can improve, then you are taking a risk,' he continues.
Cross-sectoral benchmarking however, brings its own problems, not least of which is that of language. As Paul Pederson, partner and global leader of knowledge management at Price Waterhouse, points out: 'If you compare billing in the petroleum industry it is not that different from the accounts-payable process in other sectors.' Where he, naturally enough, sees the role of the consultants is that they can take the various outputs and 'map them back to a generic structure', and he talks of getting teams into a room to decide what to call the systems. Associated with this is how the data is presented especially if a cross-border element is introduced. Much benchmark data - notably the Probe system commonly used across Europe - is given on a one-to-five scale. But Colin New, professor of manufacturing strategy at Cranfield Business School, points out that there is scope for error in some performance-related methodology. 'If say, for delivery performance you get above 95% you would automatically get a five (in some standard benchmarks). However ideal that might be in the auto industry, if you did that for (a retailer dependent on fresh produce like) Tesco you would be out of business in five minutes.'
'You have to make sure that you are comparing apples with apples,' says Kreuz, explaining how he did this for a US chemical giant. The company wanted to compare its costs with its competitors. The greatest challenge in the assignment was not getting the information, but working out what every company deemed a cost, and how they had reached that figure. But even with benchmarking data on common ground there is still scope for error, and success comes down to judgment and implementation. While it may sound blindingly obvious, identifying the correct information among the data smog is one of the biggest issues for firms. As AMS's Rodriguez says, the greatest challenge for chief executives is judging which benchmark matters. Citing David Packard, the co-founder of Hewlett Packard, he quotes: 'More entrepreneurs die from indigestion than starvation.' A further difficulty is that firms often benchmark the wrong data, simply because it is easier. So firms end up doing something like a spend in procurement versus revenue benchmark instead.
Then, says Czarnecki, although there is indeed very little that is non-benchmarkable, companies can take this to ridiculous extremes. They have to admit that certain processes within a firm are, if not industry specific, then industry-scale specific. It would be lunacy says IBM's Gallagher, for a construction company to benchmark its cheque-payment process on a credit card company. 'It just doesn't make sense,' he says. 'If you are a credit card company you would look at firms with a similar scope and scale like insurance companies, banks or telephone companies.'
The greatest pitfall of all can be at the end - what to do with suggestions brought up by the data. As Wharton's Useem says, it is less than helpful to try and replicate another company's processes exactly. 'Observe and adapt and assimilate,' he says. 'Don't copy.' The chief executive has to use the information from the benchmark study in the appropriate manner and impart the aims clearly. Employed correctly, benchmarking can be used to get out of the Not Invented Here Syndrome as part of a change process.
A common error, says Garner, is for the company not to make clear the difference between the standard and the target. He cites the difficulty encountered by BT when it advertised that its aim was for nine out of ten telephone boxes to work. 'Engineers would fix 18 out of 20, and then call it a day. The target had become the standard.'
In terms of taking into account cultural differences when benchmarking, the camp is clearly split whether you need to or not. Ruggier is convinced that it would be impossible to 'implement exactly what works in Toyota in Wolfsburg (VW's headquarters). You have to adapt to your culture and situation and business'. Useem, who would add corporate cultures to the list, points to the fact that no one has been able to get close to the customer focus that US retail store Nordstrom has, which he believes is 'embedded'. Poppycock, says AT Kearney's Graveling, disagreeing vehemently.
Given that every process has an input or output, how can that be impossible to measure?
While that debate will continue, there are few who would now dispute that benchmarking is essential. The only question is where it will stop.
Most believe that there is value to be had in benchmarking every part of the organisation. Certainly benchmarking studies are moving higher and higher into the more visible parts of the company such as communications and human resource departments. The trend seems to be, as Pederson has identified, for benchmarking to be done on a more long-range level and in directional trends. 'That is harder to map,' he admits, but as long as information can be presented which can help companies maintain a competitive edge, it will be used.