Two Thirds Done

Thoughts and experiences of an eighty one year old Wild Duck.

Avram Miller by the sea
EN | IT

587 posts since 2006

Why Did Intel Capital Fail to Influence Intel’s Strategy

Written by

avram miller

in

,

I just read a very well-written article about this question, which you can read here. https://globalventuring.com/corporate/information-technology/how-intel-lost-touch-with-its-investment-arm/

It begins with the question: How could Intel Capital, a world-leading investment unit, have created more than $170 billion in market value from investing in more than 1,800 companies, and still not be able to steer its parent through market disruptions?

Reading the article was a bit of a punch in my stomach. Intel Capital was formally created in 1991. Prior to that, I had been doing a combination of M&A and early-stage investing. It had become clear to me that acquiring companies was not going to work out, as the Intel immune system would attack any foreign object, and I was concentrating on early stage. I had built a small group. Les Vadasz, who was my boss the entire time I was at Intel—and my mentor, and importantly, my friend—was asked by Andy Grove to focus on early-stage investments.

Les and I worked together to create Intel Capital. We established three objectives that would guide our efforts:

• Gain strategic insights that would influence Intel’s strategic direction
• Grow Intel’s current business
• Make enough money to keep the bean counters away

We never imagined we would make a lot of money as early-stage investors, especially since we had a strategic mission. But we made a great deal of money even before I left in 1999. We had a big impact on Intel’s core PC business, especially with respect to the home computer market, and we made so much money that at one time we were the second-largest contributor to Intel’s earnings.

One of the key reasons I left Intel was because I felt we had no influence on the company’s strategy. I tried my best to get Andy Grove to understand that the internet was not just a way to connect personal computers together, but I failed. I was not alone in this. I wrote my book, The Flight of the Wild Duck, partly to deal with my inability to influence Grove. It was Andy who suggested the title (indirectly). When I left Intel, he told Renee James (who later became the President of Intel) that I was “Intel’s Wild Duck, and that no one would replace me.” That was sweet, but it did not change the fact that he did not listen to what I had to say, and I hold myself partially responsible for that.  After Grove became CEO, Intel was all about execution. I once accused Grove of being the Mick Jagger of technology, saying that his strategy for tomorrow was to make today last. Sadly, it was true.

After I left, and then Les left (about five years later), Intel Capital became much less strategic. It was taken over by Arvind Sodhani, who had been brilliant as the Treasurer of Intel but would not have been my choice as leader of Intel Capital. By that time I had lost touch with Intel Capital, so I found the article interesting, as it was influenced by people who came after me.

After I left Intel, I tried to provide strategic advice to various Intel CEOs who came along (without compensation or acknowledgment), but they were really not interested. I felt I owed it to the company so I kept trying.

Enjoyed this? Get new posts by email.

Comments

6 responses to “Why Did Intel Capital Fail to Influence Intel’s Strategy”

  1. Lawer Avatar
    Lawer

    Very interesting blog post and topical based on recent events. Thanks for sharing Avram!

    Reply
  2. Patrick Avatar
    Patrick

    This article misses the mark on numerous points – failing to understand the breadth of the ecosystem Intel’s product served and Intel Capital enabled; and failing to understand that Intel’s core business units repeatedly and actively ignored guidance from Intel Capital and Intel Capital portfolio companies that should have shifted the Intel business unit strategy to service the new opportunity – however this would have required effort from the Intel Business Unit to understand and respond to the market need. The business units simply took a “profit maximization” route that involved not allowing change to the core product – iterating, cost reducing, incrementally advancing at a slow pace to milk maximum profits per cycle, and actively ignoring market signals across numerous areas, notably mobile devices and cloud GPU requirements, but prior to that display and storage, faster memory and networking – all areas that would have shifted profits from the 1-horse CPU company to other parts of the computer architecture. A much more interesting analysis can be had by understanding how much Intel sold each silicon gate for, versus say Micron in memory (or any commodity DRAM), and versus NVIDIA and ATI. In repeated internal Intel analysis driven by Finance and Mfg teams, it was decided not to produce a performant 64-bit architecture and it was decided that GPU’s were “giving away” Si at a price Intel FAB’s (and Finance) would not tolerate. This resulted in blind business groups with no market touch who actively avoided responding to any external signal (other than “meet competitive pricing” – meet-comp in the lingo). In this environment, Intel Capital did operate independently – there were far greater investment opportunities that it was hoped would finally wake-up the lethargic and catatonic business units – and very few to 0 profitable vc-class investments that would be aligned with the veritable strategic dessert that the business units created by sucking all profits from the surrounding space. This is not isolated to Intel – most large companies actually do this – it is their nature to embrace and extend and cut of the air supply of any entity threatening a change to the status-quo. So I would caution that far more of the “blame” for not changing direction or taking advantage of industry shifts lies with the Intel business units driven by finance priorities and mfg profit maximization. Intel Capital continually illuminated the way with great clarity, presenting numerous strategic opportunities to Intel that the internal teams simply lacked the urgency or vision to take advantage of.

    Reply
    1. avram miller Avatar
      avram miller

      You are correct, Intel Capital was not at fault. I would also not say the business units were not at fault. They were doing what was natural for them. When we first set up Intel Capital (CBD), Les felt that we should get a business unit to support every investment. I did not agree and eventually I could make investments without their approval. I still needed legal and finance to sign off but that was seldom a problem.

      It was the board including a series of ever worse CEOs that were responsible for what happened.

      Reply
  3. CaptPat Avatar
    CaptPat

    This article misses the mark on numerous points – failing to understand the breadth of the ecosystem Intel’s product served and Intel Capital enabled; and failing to understand that Intel’s core business units repeatedly and actively ignored guidance from Intel Capital and Intel Capital portfolio companies that should have shifted the Intel business unit strategy to service the new opportunity – however this would have required effort from the Intel Business Unit to understand and respond to the market need. The business units simply took a “profit maximization” route that involved not allowing change to the core product – iterating, cost reducing, incrementally advancing at a slow pace to milk maximum profits per cycle, and actively ignoring market signals across numerous areas, notably mobile devices and cloud GPU requirements, but prior to that display and storage, faster memory and networking – all areas that would have shifted profits from the 1-horse CPU company to other parts of the computer architecture. A much more interesting analysis can be had by understanding how much Intel sold each silicon gate for, versus say Micron in memory (or any commodity DRAM), and versus NVIDIA and ATI. In repeated internal Intel analysis driven by Finance and Mfg teams, it was decided not to produce a performant 64-bit architecture and it was decided that GPU’s were “giving away” Si at a price Intel FAB’s (and Finance) would not tolerate. This resulted in blind business groups with no market touch who actively avoided responding to any external signal (other than “meet competitive pricing” – meet-comp in the lingo). In this environment, Intel Capital did operate independently – there were far greater investment opportunities that it was hoped would finally wake-up the lethargic and catatonic business units – and very few to 0 profitable vc-class investments that would be aligned with the veritable strategic dessert that the business units created by sucking all profits from the surrounding space. This is not isolated to Intel – most large companies actually do this – it is their nature to embrace and extend and cut of the air supply of any entity threatening a change to the status-quo. So I would caution that far more of the “blame” for not changing direction or taking advantage of industry shifts lies with the Intel business units driven by finance priorities and mfg profit maximization. Intel Capital continually illuminated the way with great clarity, presenting numerous strategic opportunities to Intel that the internal teams simply lacked the urgency or vision to take advantage of.

    Reply
  4. patrick Avatar
    patrick

    Avram – I saw this related post which speaks a bit to intel’s dominance of the ever-shrinking PC pie while missing adjacencies and new opportunities –

    https://www.fastcompany.com/91393051/intel-history-trump-deal

    We saw a lot of this over the years. I think there is much to be learned from Intel’s arc and Intel Capital’s relationship to it – that perhaps informs other CVC efforts that must navigate that knife edge of strategic alignment and financial return. I think much of Intel Capital’s investment model and portco/industry engagement model are still the best way a VC can operate – and are copied by other CVC’s as well as some financial VC’s as templates for driving eecosystems forward. The alignment issue is a key one, often not discussed – the business units themselves must have incentives and motivation to innovate and the ability to take risks and perhaps fail – in ways that large companies have trouble with internally. I am forming a CVC unit now, so this topic was timely enough to make me reply;) I too will be thinking about the Intel arc and trying to apply the best while avoiding the worst with a new fund in a different industry – but similar challenges and a model similar to Intel Capital. Much to learn from our history to avoid repeating it;) Thanks for sharing this blog – you remain a beacon in our industry.

    Reply
    1. avram miller Avatar
      avram miller

      Thank you for your comments. Please feel free to email me at avram@avrammiller.com if you want to discuss this further. I wish you success.

      Reply

Leave a Reply

Recent Posts