Introduction
In this discussion post, process benchmarking is defined, and it is compared with seemingly related concepts as well as other types of benchmarking. The application of benchmarking in the information technology (IT) industry is discussed, and then it is applied to a fictitious software company. Finally, it is noted that while all companies and organizations can benefit from benchmarking, contemporary industrial manufacturing (Industry 4.0) companies can especially benefit from it, due to Industry 4.0’s similarity to the IT industry.
What is Process Benchmarking?
Goetsch & Davis (2021) define benchmarking, specifically process benchmarking, as “the process of comparing and measuring an organization’s operations or its internal processes against those of a best-in-class performer from inside or outside its industry.” (p. 360)
This is different from competitive analysis, which is what customers do. Benchmarking is not about comparing the final product or service of two companies. Rather, benchmarking involves comparing the underlying processes used to produce a product or service, distribute it, and support it. The goal is to find a “secret sauce” another company uses and adopt that to one’s own company. That other company is called a benchmarking partner. If possible, the benchmarking partner should be “best-in-class,” at least in the particular process under comparison.
Benchmarking should also not be confused with so-called “best practices.” In software development, best practices are frequently arbitrary constraints not backed by any measurement. Best practices’ popularity is simply based on the insecurities that many IT employees and managers have.
Yarrow & Prabhu (1999) present three different modes of benchmarking: metric benchmarking, diagnostic benchmarking, and process benchmarking. The first concerns comparisons of performance data. They state:
So long as we are comparing `apples with apples’, metric benchmarking can serve a useful purpose as a `call to action’. However, its emphasis is on the `what’ rather than the `how’. This form of benchmarking can help an organization to pinpoint aspects of performance that need to improve, but on its own it cannot help them to learn how to improve. (Yarrow & Prabhu, 1999, p. 794)
Process benchmarking “involves two or more organizations comparing their practices in a specific area of activity, in depth, to learn how better results can be achieved.”
The third approach, diagnostic benchmarking, “seeks to explore both practices and performance, establishing not only which of the company’s results areas are relatively weak, but also which practices exhibit room for improvement.”
Benchmarking in IT Departments
In large companies with multiple independent software products, it is possible for different product teams to become benchmarking partners, if one of the teams has some metric that is significantly better than the other team’s. Even then, benchmarking with other companies could still be needed to achieve best-in-class quality.
Companies or organizations which have IT departments (which means all companies and organizations) all have certain problems in common. These problems include: what is the best database to use? What is the best way to test software? What is the best way to make websites publicly available? What is the best way to secure a computer system? and so on.
Because of this, it is possible to have benchmarking partners from widely varying industries.
The solutions to these problems are exchanged on an informal basis. IT workers frequently participate in user groups, either in-person or online. There are also online forums, product-specific support groups, etc., that can be used as sources of information.
Information gained through user groups or forums is purely anecdotal. Once this (unverified) solution to a problem is found, it must be evaluated and reviewed. This usually starts with a competitive analysis. Once it is determined that the proposed solution should indeed solve the specified problem, the solution will be implemented on a trial basis. If the trial is successful, the solution will be deployed on a wider basis. This process is called A/B testing.
Benchmarking in a Fictitious Organization
Software companies like Gaggle dot Com often find that there are two different ways to improve some aspect of a company’s IT department. To be concrete, suppose a certain database’s performance has been subpar. This can be improved by either changing some configuration property of the database or by replacing the database with one made by a different vendor.
By reading documentation and the forums of the database vendor, a configuration change may be found that improves the database’s performance. Or the problem can be described in a forum post along with a request for help. This is not benchmarking, however – it is more like self-improvement.
Real benchmarking would happen when Gaggle dot Com finds a benchmarking partner, say at a conference or user group meeting. This would be an informal and ad hoc relationship and would be the situation when a replacement database can be recommended. This is again anecdotal evidence, and it must be verified using A/B testing as described above.
In situations like this, a metric comparison with the benchmarking partner does not make sense simply because the entire processes (database plus everything else) are too dissimilar, like apples-to-oranges.
There are two quantitative comparisons that can be made, however. The first is to compare the performance of the database, acting alone, used at Gaggle.com and the benchmarking partner. In the classification that Yarrow & Prabhu (1999) present, this would be process benchmarking, where the “process” is strictly limited to the database.
The second comparison involves measuring the performance of Gaggle.com’s entire system both with the old database and with the new database in place. This is what Yarrow & Prabhu would call diagnostic benchmarking. As Yarrow & Prabhu (1999, p. 794) note, “diagnostic benchmarking is more akin to a `health check’ for the company, helping to identify which practices need to be changed and the nature and extent of performance improvements which should follow.”
In using either process or diagnostic benchmarking, Gaggle dot Com om has improved its database performance, perhaps even exceeding that of the benchmarking partner, and so have delivered a higher quality service to our customers.
Conclusion
The approach to benchmarking described here – combining process and diagnostic benchmarking – is not limited to the IT industry. As Wolniak & Grebski (2023) notes, any company can benefit from competitive benchmarking. In particular, Industry 4.0 is a transformation of traditional manufacturing processes using digital technologies, data exchange, and automation. As such, Industry 4.0 companies share the tools of IT companies, and the process and diagnostic benchmarking should be applied there, too.
References
Goetsch, D. L. & Davis, S. B. (2021). Quality management for organizational excellence: Introduction to total quality (9th ed.). Pearson.
Wolniak, R., & Grebski, W. (2023). The usage of benchmarking in Industry 4.0 conditions. Zeszyty Naukowe Politechniki Śląskiej. Organizacja i Zarządzanie, 188, 665-676. http://dx.doi.org/10.29119/1641-3466.2023.188.40
Yarrow, D. & Prabhu, V. (1999). Collaborating to compete: Benchmarking through regional partnerships. Total Quality Management, 10(4-5), 793-802. https://doi.org/10.1080/0954412997820
No comments:
Post a Comment