800x100 static WP 3
WP_Term Object
(
    [term_id] => 50
    [name] => Events
    [slug] => events
    [term_group] => 0
    [term_taxonomy_id] => 50
    [taxonomy] => category
    [description] => 
    [parent] => 0
    [count] => 1221
    [filter] => raw
    [cat_ID] => 50
    [category_count] => 1221
    [category_description] => 
    [cat_name] => Events
    [category_nicename] => events
    [category_parent] => 0
)

Kathryn Kranen at IEDM

Kathryn Kranen at IEDM
by Paul McLellan on 12-23-2014 at 7:00 am

 It is the 50th year of IEDM, the International Electron Devices Meeting. The fact that it has been going for so long reveals why it has such an odd name: back in 1964 most “electron devices” were tubes (valves in UK lingo). This year they gave all of us a USB stick with all the papers from all 50 years of the event, something that would have been unbelievable to even think about during the first conference when a few bits of memory would cost a king’s ransom.

This year at lunch on Wednesday there was an IEDM/Women in Engineering co-sponsored lunch. The setup was Kathryn Kranen, most recently CEO of Jasper Design Automation, being interviewed by Thuy Dao of Freescale.

Thuy started by asking Kathryn about the early part of her career. She studied in Texas and then worked for Rockwell, firstly in board design and then in ASIC design. She was then recruited by Daisy Systems (in Texas) and eventually moved to headquarters in Sunnyvale (in what are now Synopsys’s buildings I presume). Initially she was an application engineer. She then moved to Quickturn and learned to be a bag-carrying salesperson with a product that didn’t really work and wasn’t selling well. Technical products simply do not sell themselves. Quickturn went public (and would eventually be acquired by Cadence, the Palladium product line is the descendent of those original products).

Kathryn went to Verisity, at the time an Israeli startup. She ran the US subsidiary and got it profitable. She left when she got pregnant and Moshe Gavrielov (now CEO of Xilinx) took over. Verisity went public so presumably Kathryn made some money and she was retired for 4 years bringing up her children.

In 2002 Jasper recruited her to be CEO. At the time there were 6 people. She closed 2 rounds of funding but after a couple of years the market froze. All 3 big EDA companies had their own formal products (from acquisitions) and so customers stopped being interested since they had a formal product that was “good enough” and bundled into big deals. The founders left Jasper and recruited many of their friends so they ended up with only one of the original software developers. Luckily they had a couple of loyal customers, Sony and Cisco, who kept buying. Then they started to grow again. They opened their Brazil development site that eventually became the largest. And it turned out that their technology really was the best and JasperGold started to become the market leader.

So Jasper needed to decide what to do. They were more profitable and had more revenue than prior EDA IPOs such as Verisity. But post Sarbanes-Oxley that is not enough any more. It would be another 3-5 years to get big enough to go public. So they decided to sell. They had just gone through a failed acquisition when they discovered bad stuff during due diligence. So they decided to spend a lot on cleaning up the books, putting all the documentation online and generally making it so that once an acquisition process started it would go fast. They had regular meetings with two companies, one of which was Cadence (and I don’t think you win any prizes for guessing the second). Cadence made a lowball offer that they turned down, then they grew 35% in one year and Cadence came back with something acceptable. They didn’t use an investment banker.

So, Thuy asked, what are the 3 lessons for a startup from founding to liquidity?

[LIST=1]

  • The real enemy is time. Don’t wait for everything to be perfect or the market will pass you by.
  • It is all about the customers. They need to be willing to pay for your products. Don’t give it for free to early customers. Jasper’s first customers paid more than later ones but they got the entire company including all of engineering working for them. If the customers pay a lot of money they will invest effort to make the product work, whereas if they get it for next to nothing they will walk away.
  • Don’t clling to things that are not working, find people you need and be picky about team members.

    One thing Jasper did is not spread itself thinly. At the time Cadence acquired them they only had 25-30 customers but each on was a $1M customer. A couple of customers paying $1M is much better than 20 customers paying $50K. How hard will they work on a cheap product that is not yet mature and working cleanly?

    Kathryn was asked about whether she was treated differently as a woman. She said that at the time she didn’t think so but now looking back she can see some weirdness. She has recently got involved in STEM mini-course at the middle school that her 8th grade daughter goes to. But it is necessary to have an explicit conversation about the biases and not pretend. For example, there is a recent Barbie book about designing a video-game. Barbie designs it but then when it comes to writing the code she needs to find some men. Not exactly the ideal message!

    Kathryn is currently temporarily retired. And she has no wish to return to EDA (because it is all tiny companies or huge ones) but she would like to work for a 100-150 person company, probably some sort of B2B software since that is what she knows. Expect to see her around soon!


    More articles by Paul McLellan…

    Share this post via:

  • Comments

    There are no comments yet.

    You must register or log in to view/post comments.