Benchmarking methods varies from theory to theory. Corporations determining what to benchmark generally follow Spendolini!|s structure of thinking outside of the box (The Benchmarking Book, 1992). According to Spendolini, corporations should start from internal examination. Then, taking one step further, inspect and learn from the outstanding practices of their competitors, leading companies of the sector, and even from other businesses. The last step is to seek and integrate the best practices of the world into the corporation!|s own executing process.

Figure 1. From whom to learn in benchmarking

Bhutta and Huq (1999) designed an integrated matrix, with two dimensions; "to what is compared" and "what the comparison is being made against." Bhutta and Huq suggest every cell in the matrix be carefully evaluated in accordance with relativity.

1. To What Is Compared
There are three kinds of benchmarks implemented in comparative analysis:

(1) Performance Benchmarking -
According to Bhutta & Huq"s definition, is to compare process of competitors with their own, in order to evaluate their own performance.

(2) Process Benchmarking -
Relates to every methods and process capable of improving business process.

(3) Strategic Benchmarking -
When companies need to change their strategy, Strategic Benchmarking can be implemented to find out how their competitors" strategies are formed. Thus improving their own plans and design.

2. What the comparison is being made against.
The four ways of comparison are:

(1) Internal Benchmarking -
Implemented when comparing performance of different departments or sections within the company in order to find the best process. For instance, multinational enterprises can learn through implementing the same process in different subsidies. Such information not only is very easy to acquire, but also greatly facilitates information exchange within the organization.

(2) Competitive Benchmarking -
Implemented on competitors who provide similar goods and services. Companies, through comparing and analyzing, are able to find out the best practices of competitors in services and processes. Making it possible to close the gap between them. Due to the similarity with competitors, the best practices found are easy to apply on companies themselves.

(3) Functional Benchmarking -
Applied upon non-competitors within the sector or technologically related. Through comparison on process and function, companies are able to learn from these outstanding examples.

(4) Generic Benchmarking -
Applied to other industries. Companies learn different processes from different sectors, stimulating new ideas and methods and enabling companies to break through traditional thinking.

The two dimensions mentioned above can be integrated into the fallowing 3*4 matrix:

Source：Bhutta and Huq（1999,p.257,original adapted from Leibfried and Mcnair,1992）
﹡Relativity：★★★ High ★★ Medium ★ Low

Anderson and Petterson (1996) indicated internal benchmarking is relatively limited in terms of acquiring new information and achieving dramatic improvement. Implementing Competitive Benchmarking is more effective in regards to improving performance and strategy. However, due to the difficulties in exchanging information resulting from competition, Competitive Benchmarking is less beneficial in Process Benchmarking. Functional Benchmarking and Generic Benchmarking, when combined with Process Benchmarking, will be able bring out the most of them. The combination will enable companies to acquire more and deeper information about processes than Competitive Benchmarking. Nevertheless, when the former two combine with Performance Benchmarking and Strategic Benchmarking, the result is not satisfactory. Because, the reason is there is no competition, nor is the benchmarked target within the same sector.