|
I've been aware of the outsourcing movement since my undergraduate days at Georgia Tech. I was actually a computer science major
there for a couple years. In addition to finding physics to be more fundamental than computer science (and thus more interesting to
epistemological side of me), I was also fairly convinced that outsourcing to places like India would eventually wipe out software
programming in the US much the same as outsourcing had wiped out manufacturing. Consequently I switched majors and got BS degrees in Math
and Physics rather than Math and Computer Science. Unfortunately, I was wrong. While physics was as philosophically rewarding as I
had hoped and I did get a chance to work at Stanford on one of the last big High Energy Physics (HEP) experiments conducted in the
US (BaBar),
there aren't any large HEP experiments going on outside of CERN anymore. In the end I found myself back programming.
So when I picked up Steve Hamm's book on Wipro,
Bangalore Tiger,
I was expecting to find a history of how the Indian company had managed to gather so much outsourcing work but also how, unlike in
manufacturing, they had failed to get it all. As a software developer today I don't feel like I'm in competition with
outsourceability. I've seen firsthand how one company's Costa Rican outsourcing attempt was bungled and worked for another where we
had to rewrite an entire application that was poorly built by an outsourced Ukrainian firm. Unfortunately Steve Hamm provides the
history of Wipro only in anecdotal bursts. For the most part it's a description of the "Wipro Way", i.e. how they were running the
business in 2006 and the specific challenges that they've encountered which lead them to run the business this way. It was more of
an extended case study, no doubt written for folks with MBAs rather than historians.
Wipro started out as a family firm, Western India Vegetable Products, that mostly made cooking oil. In a story worthy of a Bollywood
film (although there surprisingly hasn't ever been one), young Azim Premji is called home from his studies at Stanford in 1966 to
take over the family business when his father dies. Nearly 50 years later, Premji is still at the helm. Wipro spun the
cooking oil portion of its business off in 2013, but at the time of Hamm's writing it still represented some 10-15% of the company's
business. As with many top companies built with the strong leadership of a single individual, it will be interesting to see how
Wipro fairs after Premji is gone. I often wonder the same thing about a Berkshire-Hathaway without Buffett or an Oracle without Ellison.
We've all seen how Microsoft floundered after Gates stepped down and how AIG spectacularly imploded when Greenberg was rudely
ejected from the helm, but how IBM continued to thrive without a Watson in charge. The jury's still out on Apple post-Jobs, but
they didn't fair very well the first time they were without him.
Where Hamm's book really shines is in describing how Wipro motivates its employees. A number of these, like providing in house
training and certification, the innovation council, six sigma projects, and CMM certified processes, are really only feasible in
larger organizations. On the other hand, their lean mentality, celebration of employee success, collection of metrics, and
openeness are something any size organization can do. Running a business to optimize the right Key Performance Indicators (KPIs)
is what leads businesses like Wipro to succeed year after year. At the time of Hamm's books they were particularly focused on Net
Promoter Score, the difference between the percentage of your customers that would recommend you and those that wouldn't. Sadly
Hamm doesn't delve into too much depth about the details of how Wipro collects their metrics, how the KPIs have changed year over
year, and how their analytics process forms a feedback loop. It seems from Hamm's book like numbers always pass through people who
make decisions on how to improve them, but I can't imagine there isn't at least some level of automation to certain decisions. It
would have been interesting to know more, but Hamm was explicitly trying to keep his book non-technical. C'est la vive.
So what does the Wipro Way say about software development? Their Veloci-Q process is fairly well-aligned with industry best
practices:
1) Project Initiation
2) Requirement Study & Project Plan
3) High-Level Design
4) Low-Level Design & Test Design
5) Coding & Test Case Development
6) System Testing
7) Release
|
It would be interesting, though, to see if this has changed at all since the widespread adoption of
git has created the world of social coding.
|
