Secrets of Software Success: Management Insights from 100 Software Firms

Posted by Hemos on Thursday February 24, 2000 @11:40AM from the learn-from-masters dept.

Allan Carscaddon has contributed a review of Secrets of Software Success: Management Insights from 100 Software Firms Around the World. As the title would indicate, this is book for people who are trying to figure out the best way to get work done in a more corporate environment. Click below if you need to learn more to survive in the corporate world.

Secrets of Software Success: Management Insights from 100 Softw
author	Detlev J. Hoch ; Cyriac R. Roeding ; Gert Purkert ; Sandro K. Lindner
pages	256
publisher	HBS Press, 2000
rating	7/10
reviewer	Allan Carscaddon (allan@carscaddon.com)
ISBN	1578511054
summary	Study of the software industry and the differences between successful companies and unsuccessful ones. Some interesting insights on the nature of success in the software industry

The Scenario

Overall, an enormous number of software firms fail, and, worldwide, estimates are that 84% of software projects fail. Yet, software is clearly one of the top investment themes and is the driver of much of the growth in the U.S. and world economies. Almost all business innovation is tied to software today, making the software industry a very attractive one for investors, managers, and workers.

The authors, who all work at McKinsey & Company in Germany, attempted to describe the differences between successful software companies and unsuccessful ones by surveying and interviewing the management of 100 companies from around the world (U.S., Europe, India, Israel) and comparing the differences between the top 34 performers (based on financial return) and the bottom 34 performers. By analyzing the differences between these companies, the authors derived a set of characteristics needed for success.

Many of the results are unsurprising. For instance, one of the secrets to being a successful software firm is attracting and retaining top talent. Successful software firms tend to be run by charismatic and visionary CEOs. Successful companies have a large web of "Partners" who help to sell the software and derive profit from these sales. They write component-oriented code and use it to make future development faster.

Some of the results are surprising. One of the most surprising to me is how poor the authors found the development procedures to be at many of the failing firms. A number of reasons are advanced for this problem, and the authors have plenty of blame to spread around. One clear message is that in many firms, and all of the unsuccessful ones, the management of software projects is seriously flawed. The authors note that estimates for the time to complete software projects usually vary by a factor of 2.5. Yet, managers are often willing to commit to a timetable that is ambitious (if not impossible) before the project scope and requirements are even determined. Less successful companies commence projects with unrealistic time and budget expectations and then fall prey to pressure from marketing, management, and customers to add features after coding has begun.

Within the successful companies, project risk is minimized by spending more time on the design and requirements stage. At this point, the most successful companies are in almost constant contact with their customers, including customers on the teams that create the initial design and requirements documents. These companies spend much more time in the design stage than unsuccessful firms, but they are much less subject to feature creep after coding has begun. When pressure is put on the teams producing the software, successful firms make marketing and the project teams negotiate directly with one another, often leading to some features being removed if new features must be added.

The authors also focus on marketing as a major difference between successful and unsuccessful firms. Their conclusion is best summed up in the chapter title "Marketing Gods Make Software Kings." They describe many of the common marketing practices of successful firms (although they use kinder terms for FUD, it is one of the primary marketing tactics of successful firms). The other major lessons of this section are how strongly successful firms concentrate on the company brand name, rather than the individual products, and how much successful firms rely on PR, especially making a "celebrity" of the CEO.

What's Bad?

The main "theme" of the book is that managing in the software industry is "Like Riding a Bull," to quote from the title of the first chapter. Although the authors do create a picture of a very dynamic competitive environment in the software industry, I thought that this was a bit silly, and was not really supported by the language or structure of the book.

The book includes a section on professional services firms. Professional services firms do write a lot of code, and are very important to software companies as a part of the web of partners that help to drive sales. However, I thought that the discussion of service firms directly was a bit misplaced. Services firms do act as threats to some software firms and complement others, but there is a very significant difference in business model and approach between a software services firm and a product-oriented software company.

The book relies heavily on a very few sources for quotes and anecdotes to illustrate the points that it makes. It relies especially heavily on quotes from a few European company managers, especially from SAP, while I would have preferred a more balanced mix.

What's Good?

The book is relatively easy to read for a business management book. The textual descriptions are enhanced by excellent explanatory graphics that will be helpful in communicating the main points of the book to the reader and to others. While I wasn't blown away by this book, I was interested in many of the insights that it provides on the structure of the marketplace in which software is sold. The book is long on analyses of market forces, but perhaps a little light on really practical "Secrets," that the title would suggest are contained in it.

The book does have an excellent review of the history of the software industry, from its origins as a portion of the services that were provided along with hardware to an industry of its own. The condensed history of the industry in the appendix will be invaluable to people who are interested in the history of the software industry.

"Secrets" and Open Source software

As I read this book, I was thinking about what it would have learned had it included Open Source software in its search for the secrets of success in the software industry. Because the book project started in 1996, it omitted Red Hat and Cygnus, both of which would have (I believe) met the authors' criteria for successful firms had they done the analysis in 1999.

One of the primary claims of the Open Source, or "Bazaar" development approach is that it is more efficient than the closed, or "Cathedral" approach practiced by all of the companies in the book. So how do Open Source development teams avoid the pitfalls that well-funded companies seem to be unable to avoid?

One of the secrets of the Open Source movement has been its ability to attract and retain top talent. Open Source projects may have a unique and sustainable advantage over software "companies" in this area because the creation of a successful Open Source project necessarily involves significantly experienced developers. In addition, it is unlikely that a project will gain much momentum if the maintainer is seen as unskilled. As a project becomes mature and is widely seen as useful, it will attract additional skilled developers-- some volunteers, and some within for-profit companies built on the open source model.

Open Source projects may also have an advantage over Cathedral projects in that the feature creep issue is typically not an issue. Rather, in the Bazaar model of development, the "customer" who needs a "feature" bears the cost of adding that feature to the code, and then returns the code (hopefully) to the community.

Marketing and building partner webs will be challenges for Open Source projects in the future. Some Open Source projects have been very successful at building a web of partners Apache, for instance, has a number of "partners" promoting it as a product and selling add-on services. Linux, under the various distributions, is beginning to see real progress. Marketing may come from the migration of Open Source to big business, but this will always be a challenge for worthy, but emerging projects

So What's In It For Me?

"Secrets" will be a good read for the person considering a software venture or working in one at any management level. The description of software industry dynamics is a valuable. People who are interested in the software industry may be interested in the book, especially the historical sketches of the industry. However, those most interested in the historical sketch may want to borrow, rather than buy the book.

"It's Like Riding a Bull"
A New Business Called "Software"
Exceptional Software Leaders Are the Rule
Winning the War for Software Talent
Software Development: Completing a Mission Impossible
Marketing Gods Make Software Kings
Professional Software Services: Experts at Marketing Trust
Grow Your Partners to Grow Yourself
The Landscape of the Future
Staying on the Bull