“Changing Business Processes is Like Building Invisible Factories”

Erik Brynjolfsson
Erik Brynjolfsson

There is a lot of talk about agile organizations, and the ability of a company to manage its business processes has become more and more important. Erik, what do you think about the importance of operational excellence?

Brynjolfsson: A lot of the innovation that is happening today in companies is not particularly in the products but rather in the processes, the way that the businesses are run. Nokia is a good example of that, Zara is a good example of that, and Walmart, too. These are all companies that have been very innovative in their business processes. We found that when companies make investments in information systems, like large ERP systems such as an SAP installation, there are about five times as many dollars spent on investing in new business processes and training and education of the workers as there is spent on the software and the hardware. Often these investments in designing new business processes are not recorded as investments but are treated as expenses, and that can lead to a misunderstanding of how big the computerization effort really is. We call this investment organization capital. Even though it is not measured very well it is one of the most important things that is driving productivity today.

How much do you think is this organization capital worth in US dollars?

Brynjolfsson: When companies make these business process changes I think of it as if they are building invisible factories. According to our research, the total value of these is about $2 trillion of additional productive capacity which has not previously been measured in the GDP statistics on the balance sheets of these companies.

Let me give you a specific example. Michael Dell invited me to visit the factory of Dell Computers in Austin, Texas. I had a tour from the Vice President of Operations, and I was very impressed by the way the factory was run. I found they were very efficient in the way that materials were delivered and in the way they took orders from their customers over the Internet and arranged the deliveries from their suppliers.
One of the things that surprised me, however, was that when I looked out over the factory, which was the size of two football fields, I saw that almost half of the factory was empty and not being used. I asked the VP of Operations why this was so and he said that just three months earlier the whole factory had been full, but that they had done a big reorganization.
They had installed some new software and they had resourced their inventory policies and flow of materials. As a result they were now able to produce 20 per cent more computers while using 40 per cent less floor space. One of the things they changed, for instance, was that they now have the suppliers deliver the materials every four hours.
Dell has continued to ramp up their production. The company is now using all of the floor space in the factory including the floor space that was not necessary before. If Dell had not improved its business processes they would have had to build a second factory to keep up with demand. Instead of building a second factory, it was able to use the existing factory more efficiently, so essentially it built a factory within a factory. The company built a new factory made of software and business processes instead of one that was made of bricks and mortar. This is happening all over the United States, and all over the advanced countries, where literally thousands of these invisible factories are being built.

Can you give us any advice on how firms successfully change their business?

Brynjolfsson: We had an interesting result on this question. We found that the firms that had the very lowest performance were the ones who had made a few changes in technology and small changes in business organization but had not gone all the way. The firms that had made virtually no changes and stayed with the traditional organization at least had a working, coherent system, and the firms that had been aggressive and made changes in both the technology and the organization did the best of all.

The lesson that I take from this is that it is very important to have what we call a coherent system, a set of practices that fits together. Although the traditional organization is not as effective as the digital organization, at least the traditional organization is coherent; all the pieces fit together and work. The worst thing seems to be to take some pieces of the new system and some pieces of the old system and try to have a sort of mongrel system where the pieces do not fit together very well.

Who in a company should drive such an aggressive change in both technology and organization?

Brynjolfsson: Because of the importance of having all the pieces fit together, we found that it is essential to have the CEO, the very top of the organization, driving the change. If only one or two division leaders or Vice Presidents or functional leaders try to drive the change in some cases it may backfire and make the system less efficient. For instance, if the CIO aggressively puts in some technology but the supply chain is not changed, there may be no real benefits. Or if the Human Resource Manager does not work in a coordinated way to change the incentive systems and the training, that may not be effective.

What then is a CIO’s role? Is it up to them to focus on costs or on innovation?

Brynjolfsson: The CIO’s role has really changed, or it should be changing. Today the CIO needs to work much more closely with the business division leaders, and co-ordinate the technologies with the business process changes. It no longer makes sense to just focus on the IT function narrowly and simply try to cut costs or get efficiency optimized just within the IT function. This I think ignores the potential of information technology to really transform the business and bring cost savings throughout the entire business and not just the IT function, and more importantly increase customer satisfaction and ultimately hit revenues and profitability of the corporation as a whole.

What is the right mix of costs for a CIO?

Brynjolfsson: I had a good conversation with the Chief Financial Officer of Citigroup. They spend about $9 billion on technology and operations, and I asked him whether he was looking for ways to cut that budget because of hardware costs coming down and development costs coming down with outsourcing. I asked whether in the next five years he expected there to be a smaller share of their total cost structure, and he said something which surprised me. He said that he hoped not, that his goal was to make it a bigger share of the cost structure. I said ‘Why is that?’, and he said that the technology and operations budget was the place where there were the best gains and efficiency in technological advances. If he could shift costs from labor and other parts of the bank into the IT budget he would be on a curve of increasing cost reductions with Moore’s Law and the other improvements. His goal was to reduce costs elsewhere by automating and converting them into technology and operations budget areas. More generally there is a tremendous opportunity for companies to substitute labor, to save labor by automating business processes and allowing the information technologies to handle more and more functions.

What do you think about the importance of new technologies like RFID from an economic point of view? It is just an emotional issue or is there real value in it?

Brynjolfsson: There is a tremendous amount of value in the whole wave of new innovations that are hitting the market now, including RFID. Sometimes you read that people claim that IT has become mature and that there are no more important innovations, and it is simply a matter of exploiting the innovations from 20 years ago. I disagree in a profound way with that.

IT is not like the railroads or the steam engine where there was one big innovation and then it was simply a matter of applying that over a period of years.
Most of the benefits from information technology do not come from simply implementing these new technologies. Rather they come from inventing new business processes and new ways of interacting with customers and suppliers that have been enabled by the technology. For instance RFID makes possible a revolution in the supply chain, a revolution that companies like Walmart are pioneering. That is where the big benefits will come from, with the new way of organizing the supply chain and interacting with customers, not from simply installing new technology into old business processes.
Although I am very optimistic about the continuing changes in technology, another way of thinking of it is to suppose some catastrophe happened and technology froze for the next 10 years. Imagine that there were no new advances in technology. I believe that we would continue to have innovations in business processes for at least another decade. In fact, I think there is so much opportunity to improve the operations of companies simply using existing technology that we are far from the end of the productivity revolution.

Apart from that view – what do you think about innovations like for example a service oriented IT architecture in an enterprise?

Brynjolfsson: A service oriented architecture is an innovation which I expect to see spread more widely in the coming years. It is made possible in part by the lower communication costs. Some people see service oriented architectures as just ways of saving a lot of money on the hardware and software of the company. It is true that there can be some big savings, but I do not think that is the main benefit. I see the bigger benefit in the increased flexibility and potential for organizational innovation which a service oriented architecture makes possible. It is not simply a matter of doing what you have been doing before but more cheaply, it is a matter of increased flexibility and agility that a service oriented architecture creates. That is where the big benefits will come, I believe. This is what makes operational and organizational excellence.

The Interview was held by Dr. Paul Hofmann und Dr. Rainer Stoll