This website uses cookies and other technologies to help us provide you with better content and customized services. If you want to continue to enjoy this website’s content, please agree to our use of cookies. For more information on cookies and their use, please see our latest Privacy Policy.

Accept

cwlogo

切換側邊選單 切換搜尋選單

Morris Chang’s master plan for 50 years: The art of sowing despair

Morris Chang’s master plan for 50 years: The art of sowing despair

Source:Monte Jode Science & Technology Association of Taiwan

Morris Chang, founder of TSMC, reflected upon the wisdom he’s gained in the past 60 years during a speech. He also disclosed—for the first time—the roadmap he personally devised for TSMC back in 1998.

Views

2976
Share

Morris Chang’s master plan for 50 years: The art of sowing despair

By Hannah Chang
web only

Semiconductor Century ConversationMorris Chang x Chris MillerBook your seat: https://bit.ly/3Yd6IUv

On October 26th, TSMC Founder Morris Chang spoke at the 20th anniversary forum and gala dinner hosted by the Monte Jade Science & Technology Association of Taiwan. For the first time, Chang divulged the “TSMC Strategy” that was designed in 1998, stressing: “The most vital lesson a leader can learn is that whether it’s your individual mindset, your company strategy, or your business model, everything must constantly be made anew, because this is the only way to make a difference and continually grow your company.”

Chang says TSMC’s business model is nothing more than continuous, ceaseless innovation. Depending on who the customer was, TSMC’s thinking would go beyond technology and product.

If TSMC had stuck with making semiconductors all those years, its clients would still be PC manufacturers, instead of other semiconductor giants. TSMC broke the mold by inventing the “semiconductor contract manufacturing and design service” business model and focusing on being an independent “pure-play” semiconductor foundry. “By turning fabless semiconductor companies into our clients, we made TSMC unique.”

He also says TSMC’s market position will become more and more unshakeable. Even though countries around the globe are scrambling to build their own semiconductor plants, at the moment, TSMC’s head start is insurmountable. 

Here are the key points from Chang’s speech:


At the age of 20, I was studying at the Massachusetts Institute of Technology. My father gave me 50 shares of IBM stocks, which taught me the ins and outs of price-to-earnings ratios, P/B ratios and stock prices. As a stockholder, I learned to read about the stock market in newspapers borrowed from the school library, and I studied the quarterly and annual reports that IBM mailed me. I dissected how big companies operated. 

At 27, I began working in Texas Instruments, and I made quite an impression. Texas Instruments sent me to get my PhD from Stanford University, which I managed to do in two and a half years. I returned to Texas Instruments and spent the next quarter century working there.

I was promoted to department head after returning to Texas Instruments. You must always have a clear grasp of the organization structure of your company: Are departments organized according to the business they oversee, or the functions they perform? Organizing departments according to the business they do is a good way to train general managers, but not every company is suited to this kind of structure.   

For example, since the technology at TSMC is constantly evolving, if a customer needs more than one kind of technology, they will need to negotiate with separate departments. And since even the most advanced technologies will no longer be new after a couple of years, if the same department is put in charge of all the new tech, and the department that oversees older production processes keeps shrinking—that’s no way to run a business.

At Texas Instruments, I was the head of many different sales departments before they put me in charge of semiconductors. I began with 3,000 workers under my watch; I ended up supervising around 40,000 employees. 

I was at Texas Instruments from when I was 33 to 52— the best years of my life. But I learned much from these years. I believe technology is of the utmost importance, and I believe that in a tech company, nothing is more important than having a CEO with a tech background.

Pat Gelsinger, the current CEO of Intel, has a background in tech. “None of the last few CEOs had a tech background, so how could the company have performed well?" Those are his own words.

But it’s more than just about technology. In 2015, he visited me in Taiwan. At the time, he was head of another company, called VMWare. I asked him what VMWare did.

Gelsinger was able to explain VMWare’s business very clearly in 15 minutes. I thought to myself: This man will go far!

This is something I cannot stress enough: A manager needs to learn how to prepare a convincing briefing. The secret is this: You must really believe what you are saying. 

I remember the first time I was asked to give a speech at Texas Instruments; I was in my thirties. I had no experience and was very anxious. I asked my boss: What should I talk about?

His answer: “Talk about something you know.” I’ve lived by that advice ever since.

(Source: Monte Jode Science & Technology Association of Taiwan)

Why do CEOs make about 50 times what engineers make? The difference is their ability to assign value to things

Other important lessons I learned included the difference between “products” and “tailor-made goods”. You must ask yourself: Do I provide products or tailor-made goods? Like Michael Porter famously said, if there is another company making what you make, you are selling products. If you want to stand out and make unique, tailor-made goods, you need to have the right strategy.

For example, pricing strategy. A CEO usually makes about 50 times what an engineer makes, or around 400 times what an operator on the assembly line makes. Why do you deserve a bigger paycheck? Because CEOs need to understand the company’s pricing strategy. If the CEO could increase the asking price of a product by 1%, that’s akin to a thousand engineers coming together to lower the production cost by 1%. Hence the difference.

Needless to say, it’s hard to change the price. Especially if all you sell is a product, you have little control over pricing; it’s all been decided for you by market demand and your competitors. Only tailor-made, customized solutions have a chance to make their own fate, but this also involves market understanding, marketing, and customer management. 

I never take notes when a person speaks. Notes are worse than useless. I spend my time watching the person and thinking, why is he using these words, what is he trying to say? What is the underlying motive?

I also look at their body language, which often tells you more than their words. Look at me now, my body language shows my passion for this topic!

Tolerance is also important. When I first came back to Taiwan, I did not have my own team. A colleague of Chinese descent from Texas Instruments came to visit me. He said, you don’t have your own team, that’s no way to do business.

My thoughts are to the contrary. A team is a cage you build for yourself. Don’t lock yourself in your own little cage. You need to tolerate different ways of thinking and expand your horizons. 

There are two ways to define what makes a leader. One is there are people who are willing to follow you. The other is you know the right way to go. A good leader, then, is someone people are willing to follow, and someone who actually knows where to go.

All I’ve said so far are the basics. “Strategy” is on a higher level.

In 1974, when I was still with Texas Instruments, I used the strategy of “continuing to cut prices on TTL”.

This was a very successful strategy. At the time, Texas Instruments controlled nearly half of the market, but we still faced many competitors. Like Jack Welch, former Chairman and CEO of General Electric, famously told his workers: “We need to become number one or number two in every business we are in within five years, or else we need to get out of that business.” 

In the semiconductor business, we always asked for profit margins in excess of 50%. My strategy at the time was to sow despair in the minds of my opponents. When even Texas Instruments could only achieve a profit margin of 40% by asking the lowest price, they knew they had to get out of the business.

After I turned 54, I founded TSMC. I only had 120 employees when I started; when I retired, 50,000 people were on my payroll. The keys are “business model” and “strategy”, but unless you are the founder, you have little chance of creating a new business model. 

TSMC’s business model was our most important innovation. At the time, TSMC’s business model depended on who the customer was rather than what the product was.

If TSMC had stuck with making semiconductors, we’d still be selling to PC companies. But TSMC chose the business model of “semiconductor contract manufacturing and design service”. We turned other semiconductor companies into our customers. This was how we were different. 

When I came back to Taiwan in the 90s, the business model pioneer I admired the most was Starbucks. They took coffee that cost 30 or 40 cents and sold it for two and half bucks. Because the customer wasn’t someone who wanted to grab a thirty-cent coffee and go; the customer wanted to stay and enjoy life. I have nothing but respect for what they’ve accomplished.

Afterwards, Google and Facebook came onto the scene. Everyone is Google’s customer; we are also Google’s tools. Google collects data from its users and sells it to someone else, and it sells ads. This is also a new, revolutionary business model.

Revealing TSMC’s strategy from 1998 for the first time: Every employee was a salesperson

TSMC followed a very clear strategy when it designed its business model. I can now reveal the only handwritten strategy I penned for TSMC in 1998: First and foremost, understand what the customer needs. For example, you can offer the customer technology that surpasses their competitors, or adaptively satisfy their technical needs, or lower the price, or shorten the time to mass production,; or provide reliable quality, or provide seamless turn-key services, or protect their privacy. 

(Source: Monte Jode Science & Technology Association of Taiwan)

TSMC’s strategy was to beat its competitors in every field. Because of this, we had to maintain a good relationship with our clients. We had to be flexible in our pricing, so we could establish a “first and last look” relationship with the clients.

The company also had to put marketing and services at its core. Every employee was a salesperson; even professional engineers, accountants, and managers were TSMC’s salespersons. 

At the time, the end goal of TSMC’s strategy was to push ROE above 20%, become the world’s largest semiconductor foundry, and reach US$10 billion in revenue by 2010. TSMC’s revenue was actually US$13 billion in 2010. As you can see, it was a successful business model and strategy. 


Have you read?

♦ TSMC’s bold net zero pledge: Cutting a Taipei’s worth of emissions
♦ How TSMC may change Taiwan’s ‘Rust Belt City’
♦ TSMC Recovered Firefly Population in Plant Zone

Translated by Jack Chou
Edited by TC Lin
Uploaded by Penny Chiang

Views

2976
Share

Keywords:

好友人數