The Art of Partnering [from]

The Art of Partnering

Img 3385

When I went through the security line at San Francisco
International Airport this morning, I noticed this laptop with an Apple
sticker pasted over its Dell logo (click to enlarge the photo if you
don’t believe me). I thought this was very funny, so I asked its owner
why he did this. He explained that he was tired of explaining why he
had a Dell. I told him that I’d never heard of an Apple owner pasting a
Dell sticker over the Apple logo, and he agreed that this was unlikely
to happen. (At that point, he noticed my PowerBook’s Tony Hawk
autograph, but I digress…)

This got me to thinking about how companies form
partnerships–pasting each other’s logos on products and services and
ending up with crap. The fallacy of partnerships–and how “partner”
became a verb–is rooted in the days of 1998-2000. During these
years, most startups didn’t have a business model, so they blew smoke
about having “partnered” with big firms. Surely if a company partnered
with Microsoft or IBM, it would be successful.

To this day, whenever an entrepreneur uses “partner” used as a
verb, it bothers me because I hear, “Bull-shitake relationship that
isn’t going to increase revenue.” However, I am not an angry little
man, so in the spirit of improving what has become a flawed process, I
offer The Art of Partnering.

  1. Partner for “spreadsheet” reasons. Most companies
    form partnerships for the wrong reason: To make the press and analysts
    happy. This is stupid. The right reason to form a partnership is to
    increase sales or decrease costs. Here’s a quick test: Will you
    recalculate the spreadsheet model of your financial projections if the
    partnership happens? If not, then the partnership is doomed. You can
    wave your hands all you like about “visibility,” “credibility,” and
    “acceptability,” but if you can’t quantify the partnership, then you
    don’t have one.
  2. Define deliverables and objectives. If the primary
    goal of a partnership is to deliver “spreadsheet reasons,” then
    execution is dependent on setting deliverables and objectives such as
    additional revenues, lower costs, penetration of new markets, and new
    products and services. The only way to determine whether a partnership
    is working is to answer quantifiable questions such as, “How many more
    more downloads of software occurred because our two web sites are now
  3. Ensure that the middles and bottoms like the deal.
    Most partnerships form when two CEOs meet at an industry boondoggle.
    The next thing you know they’ve concocted a partnership that “the press
    and analysts will love,” and the next step is to get the PR people to
    draft an announcement. Is it any wonder partnerships seldom work? Some
    people believe that the key to successful partnerships is that
    top-management thought of it. They’re wrong. The key is that the
    middles and bottoms of both organizations like the partnership–after
    all, they will be implementing it. Indeed, the best partnerships occur
    when the middles and bottoms work together and wake up one day with a
    de-facto partnership that didn’t involve top management until it was
  4. Designate internal champions. Long after the press
    conference and announcement, one person inside each organization must
    remain the champion of the partnership. “A bunch of people contributing
    to the partnership when they can” doesn’t cut it. For example, during
    the desktop publishing days of Apple, John Scull (not Sculley) was “Mr.
    Desktop Publishing” at Apple. His counterpart at Aldus was Paul
    Brainerd. So the responsibility for the success of desktop publishing
    came down to John and Paul–not John, Paul, George, Ringo, and a host
    of other part-time contributors.
  5. Accentuate strengths, don’t hide weaknesses.
    Companies form most partnerships to hide their respective weaknesses.
    For example, Apple and DEC formed such a partnership in the 1980s.
    Apple’s weakness was a lack of data communications strategy. DEC’s
    weakness was a lack of a personal computer strategy. So the two
    companies tried to put two and two together. In the end two and two
    didn’t even add up to four because DEC’s data communications strategy
    couldn’t help Apple, and Apple’s personal computer strategy couldn’t
    help DEC. The deal between Apple and Intel has better prospects because
    it is based on the companies respective strengths: Apple’s ability to
    design great consumer devices, and Intel’s ability to build fast chips
    with low power requirements.*(see footnote) And this partnership
    certainly has “spreadsheet” reasons for both parties.
  6. Cut win-win deals. A partnership seldom takes
    place between equals. As a result, the more powerful side is tempted to
    squeeze the other party. The weaker side, for its part, will
    begrudgingly accept such deals and try to get what it can. Bad idea.
    Bad karma. Bad practicality. If the partnership is a win-lose deal, it
    will blow up because concrete walls and barbed wire cannot hold a
    partnership together. Only mutually beneficial results can. In the
    long, the bitter seed of resentment planted at the start of a
    partnership will grow into a giant, destructive weed.
  7. Put in an “out” clause. No matter how great the
    deal looks, put in an “out” clause so that both parties may terminate
    the partnership relatively easily. This may seem counter intuitive, but
    if companies know that they can get out of something, they’ll work
    harder to make it successful. This is because easy out clauses can
    increase motivation: “We’d better keep up our end of the bargain
    because we need these guys, and they can walk.” Frankly, if all that’s
    holding the partnership together is a legal document, then it’s
    probably not going to work anyway. It’s hard to imagine that indentured
    servitude is a motivating model of employment.
  8. Ask women. Men have a fundamental genetic flaw.
    Actually, they have many fundamental genetic flaws, but I am only
    concerned with one here: The desire to partner (verb!) with anything
    that moves. They don’t care about practicalities and long-term
    implications. If something is moving, men want to partner with it.
    Women, by contrast, do not have this genetic flaw. When you come up
    with an idea for a partnership, don’t bother asking men what they think
    about it because they will almost always think it’s a good idea.
    Instead, ask women and gain some real insight as to whether the
    partnership makes sense.
  9. Wait to legislate. Remember in the Art of
    Recruiting entry when I said that an offer letter is the last step in
    the process? An offer letter is not properly used as a “strawman” to
    get negotiation going. The same thing applies to a partnership. After
    you’ve reached closure on the deal terms–the result of many meetings,
    phone calls, and emails–then you draft an agreement. This happens at
    the end of the process because you want the people to have
    psychologically committed themselves to the partnership. If you start
    the drafting process too early, you’re asking for nit-picking delays
    and blowups. Incidentally, if you ask for legal advice too early,
    you’ll kill the process. The best way to deal with lawyers is to simply
    say to them: “This is what I want to do. Now keep us out of jail as we
    do it.”

Written at: Marriott Hotel, Park Ridge, New Jersey

* Please God, take these two strengths and give us a laptop that has
the Macintosh interface and a six-hour battery life. But then, God, why
didn’t Steve talk about battery life in his keynote address?

Thanks, Tom Kang, for your outstanding contributions to this entry.

The Effective Emailer [from]

The Effective Emailer

Istock 000000120894Medium

Because of my recent post about schmoozing,
you might think I’m a warm, fuzzy, and kumbaya kind of Guy. Most of the
time I am, but I have strong feelings about email etiquette and what it
takes to get your email read–and answered. As someone who gets dozens
of emails every day and sends a handful of emails every day to get
strangers to do things (“digital evangelism”), I offer these insights
to help you become a more effective emailer.

  1. Craft your subject line. Your subject line is a
    window into your soul, so make it a good one. First, it has to get your
    message past the spam filters, so take out anything about sex and
    money-saving special offers. Then, it must communicate that your
    message is highly personalized. For example, “Love your blog,” “Love
    your book,” and “You skate well for an old man,” always work on me. 🙂
    While you’re at it, craft your “From:” line too because when people see
    the From is from a company, they usually assume the message is spam.
  2. Limit your recipients. As a rule of thumb, the
    more people you send an email to, the less likely any single person
    will respond to it, much less perform any action that you requested.
    (Thanks, Parker, for mentioning this.) This is similar to the Genovese Syndrome (or the “bystander effect”):
    In 1964, the press reported that thirty eight people “stood by” while
    Kitty Genovese was murdered. If you are going to ask a large group of
    people to do something, then at least use blind carbon copies; not only
    will the few recipients think they are important, you won’t burden the
    whole list with everyone’s email address. Nor will you reveal
    everyone’s email address inadvertently.
  3. Don’t write in ALL CAPS. Everyone probably knows
    this by now, but just in case. Text in all caps is interpreted as
    YELLING in email. Even if you’re not yelling, it’s more difficult to
    read text that’s in all caps, so do your recipients a favor and use
    standard capitalization practices.
  4. Keep it short. The ideal length for an email is
    five sentences. If you’re asking something reasonable of a reasonable
    recipient, simply explain who you are in one or two sentences and get
    to the ask. If it’s not reasonable, don’t ask at all. My theory is that
    people who tell their life story suspect that their request is on shaky
    ground so they try build up a case to soften up the recipient. Another
    very good reason to keep it short is that you never know where your
    email will end up–all the way from your minister to the attorney
    general of New York. (courtesy of Jonathan) There is one exception to
    this brevity rule: When you really don’t want anything from the
    recipient, and you simply want to heap praise and kindness upon her.
    Then you can go on as long as you like!
  5. Quote back. Even if emails are flying back and
    forth within hours, be sure to quote back the text that you’re
    answering. Assume that the person you’re corresponding with has fifty
    email conversations going at once. If you answer with a simple, “Yes, I
    agree,” most of the time you will force the recipient to dig through
    his deleted mail folder to figure out what you’re agreeing to. However,
    don’t “fisk” either (courtesy of Brad Hutchings). Fisking is when you
    quote back the entire message and respond line by line, often in an
    argumentative way. This is anal if not downright childish, so don’t
    feel like you have to respond to every issue.
  6. Use plain text. I hate HTML email. I tried it for
    a while, but it’s not worth the trouble of sending or receiving it. All
    those pretty colors and fancy type faces and styles make me want to
    puke. Cut to the chase: say what you have to say in as brief and plain
    manner as possible. If you can’t say it in plain text, you don’t have
    anything worth saying.
  7. Control your URLs. I don’t know what’s gotten into
    some companies, but the URLs that they generate have dozens of letters
    and numbers. It seems to me that these thirty-two character URLs have
    almost as many possible combinations than the number of atoms in the
    universe–I don’t know how many URLs a company intends to create, but
    it’s probably a smaller number than this. If you’re forwarding an URL,
    and it wraps to the next line, it’s very likely that clicking on it
    won’t work. If you really want someone to click through successfully,
    go through the trouble of using SnipURL to shorten it. SnipURL also provides the functionality of showing you how many people have clicked on the link.
  8. Don’t FUQ (Fabricate Unanswerable Questions), I.
    Many people send emails that are unanswerable. If your question is only
    appropriate for your psychiatrist, mother, or spouse, then ask them,
    not your recipient. When I get this type of message I go into a deep
    funk: (a) Should I just not answer? But then the person will think I’m
    an arrogant schmuck; (b) Should I just give a cursory answer and
    explain that it’s not answerable? (c) Should I carefully craft a
    heartfelt message probing for more information so that I can get into
    the deep recesses of the sender’s mind and begin a long tail of a
    message thread that lasts two weeks? Usually, I pick option (b).
  9. Don’t FUQ, II. There’s one more type of
    unanswerable message: the open-ended question that is so broad it
    should be used in a job interview at Google. For example, “What do you
    think of the RIAA lawsuits?” “What kind of person is Steve Jobs?” “Do
    you think it’s a good time to start a company?” My favorite ones begin
    like this: “I haven’t given this much thought, but what do you think
    about…?” In other words, the sender hasn’t done much thinking and
    wants to shift responsibility to the recipient. Dream on. The purpose
    of email is to save time, not kill time. You may have infinite time to
    ask essay questions but don’t assume your recipient does.
  10. Attach files infrequently. How often do you get an
    email that says, “Please read the attached letter.”? Then you open the
    attachment, and it’s a dumb-shitake Word document with a three
    paragraph message that could have easily been copied and pasted into
    the email. Or, even worse, someone believes that his curve-jumping,
    paradigm-shifting, patent-pending way to sell dog food online means
    you’ll want to receive his ten megabyte PowerPoint presentation? Now
    that lots of people are opening messages with smartphones–sending
    files when you don’t have to is a sure sign of bozosity.
  11. Ask permission. If you must ask unanswerable
    questions or attach a file, then first seek permission. The initial
    email should be something like, “May I tell you my background to
    explain why I’m contacting you?” Or, “May I send you my PowerPoint
    presentation to explain what our company is doing?”
  12. Chill out. This is a rule that I’ve broken many
    times, and each time that I did, I regretted it. When someone writes
    you a pissy email, the irresistible temptation is to retaliate. (And
    this is for an inconsequential email message–no wonder countries go to
    war.) You will almost always make the situation worse. A good practice
    is to wait twenty-four hours before you respond. An even better
    practice is that you never say in email what you wouldn’t say in
    person–this applies to both the sender and recipient, by the way. The
    best practice is to never answer and let the sender wonder if his email
    got caught in a spam filter or didn’t even matter enough to merit a
    response. Take my advice and do as I say, not as I have done–or will
    do. 🙂

Addendumbs (ie, stuff that should have been in here in the first place, but I was too dumb):

  • Per Russell Willis and Grace Lee, add a good signature.
    That is, one that includes your name, title, organization, email
    address, web site, and phone. This is especially true if you’re asking
    people to do something–why make it hard for them to verify your
    credibility or to pick up the phone and call you? Also, I often copy
    and paste people’s signatures to put them into the notes field of an
    appointment. The email client that I use, Entourage, won’t let you
    easily copy the sender’s info from the header, so I have to create a
    forward, copy everything, and then delete the forward.
  • Never forward something that you think is funny.
    The odds are that by the time you’ve received it, your recipient
    already has too, so what is intended as funny is now tedious. However,
    I do have the Neiman-Marcus recipe for cookies…

Written at: United Airlines flight #230; Denver-SFO, seat 2J.



HTML语法大全 [copied one]


  <! – – … – -> 批注  

  <A HREF TARGET> 指定超级链接的分割窗口  

  <A HREF=#锚的名称> 指定锚名称的超级链接  

  <A HREF> 指定超级链接  

  <A NAME=锚的名称> 被连结点的名称  

  <ADDRESS>….</ADDRESS> 用来显示电子邮箱地址  

  <B> 粗体字  

  <BASE TARGET> 指定超级链接的分割窗口  

  <BASEFONT SIZE> 更改预设字形大小  

  <BGSOUND SRC> 加入背景音乐  

  <BIG> 显示大字体  

  <BLINK> 闪烁的文字  


  <BODY> 显示本文  

  <BR> 换行  

  <CAPTION ALIGN> 设定表格标题位置  

  <CAPTION>…</CAPTION> 为表格加上标题  

  <CENTER> 向中对齐  

  <CITE>…<CITE> 用于引经据典的文字  

  <CODE>…</CODE> 用于列出一段程序代码  

  <COMMENT>…</COMMENT> 加上批注  

  <DD> 设定定义列表的项目解说  

  <DFN>…</DFN> 显示”定义”文字  

  <DIR>…</DIR> 列表文字卷标  

  <DL>…</DL> 设定定义列表的卷标  

  <DT> 设定定义列表的项目  

  <EM> 强调之用  

  <FONT FACE> 任意指定所用的字形  

  <FONT SIZE> 设定字体大小  

  <FORM ACTION> 设定户动式窗体的处理方式  

  <FORM METHOD> 设定户动式窗体之资料传送方式  

  <FRAME MARGINHEIGHT> 设定窗口的上下边界  

  <FRAME MARGINWIDTH> 设定窗口的左右边界  

  <FRAME NAME> 为分割窗口命名  

  <FRAME NORESIZE> 锁住分割窗口的大小  

  <FRAME SCROLLING> 设定分割窗口的滚动条  

  <FRAME SRC> 将HTML文件加入窗口  

  <FRAMESET COLS> 将窗口分割成左右的子窗口  

  <FRAMESET ROWS> 将窗口分割成上下的子窗口  

  <FRAMESET>…</FRAMESET> 划分分割窗口  

  <H1>~<H6> 设定文字大小  

  <HEAD> 标示文件信息  

  <HR> 加上分网格线  

  <HTML> 文件的开始与结束  

  <I> 斜体字  

  <IMG ALIGN> 调整图形影像的位置  

  <IMG ALT> 为你的图形影像加注  

  <IMG DYNSRC LOOP> 加入影片  

  <IMG HEIGHT WIDTH> 插入图片并预设图形大小  

  <IMG HSPACE> 插入图片并预设图形的左右边界  

  <IMG LOWSRC> 预载图片功能  

  <IMG SRC BORDER> 设定图片边界  

  <IMG SRC> 插入图片  

  <IMG VSPACE> 插入图片并预设图形的上下边界  

  <INPUT TYPE NAME VALUE> 在窗体中加入输入字段  

  <ISINDEX> 定义查询用窗体  


  <KBD>…</KBD> 表示使用者输入文字  


  <LI TYPE>…</LI> 列表的项目 ( 可指定符号 )  


  <MARQUEE> 跑马灯效果  

  <MENU>…</MENU> 条列文字卷标  


  <MULTIPLE> 可同时选择多项的列表栏  


  <NOFRAME> 定义不出现分割窗口的文字  


  <OL>…</OL> 有序号的列表  

  <OPTION> 定义窗体中列表栏的项目  


  <P ALIGN> 设定对齐方向  

  <P> 分段  

  <PERSON>…</PERSON> 显示人名  

  <PRE> 使用原有排列  


  <SAMP>…</SAMP> 用于引用字  

  <SELECT>…</SELECT> 在窗体中定义列表栏  

  <SMALL> 显示小字体  

  <STRIKE> 文字加横线  

  <STRONG> 用于加强语气  

  <SUB> 下标字  

  <SUP> 上标字  


  <TABLE BORDER=n> 调整表格的宽线高度  

  <TABLE CELLPADDING> 调整数据域位之边界  

  <TABLE CELLSPACING> 调整表格线的宽度  

  <TABLE HEIGHT> 调整表格的高度  

  <TABLE WIDTH> 调整表格的宽度  

  <TABLE>…</TABLE> 产生表格的卷标  

  <TD ALIGN> 调整表格字段之左右对齐  

  <TD BGCOLOR> 设定表格字段之背景颜色  

  <TD COLSPAN ROWSPAN> 表格字段的合并  

  <TD NOWRAP> 设定表格字段不换行  

  <TD VALIGN> 调整表格字段之上下对齐  

  <TD WIDTH> 调整表格字段宽度  

  <TD>…</TD> 定义表格的数据域位  

  <TEXTAREA NAME ROWS COLS> 窗体中加入多少列的文字输入栏  

  <TEXTAREA WRAP> 决定文字输入栏是自动否换行  

  <TH>…</TH> 定义表格的标头字段  

  <TITLE> 文件标题  

  <TR>…</TR> 定义表格美一行  

  <TT> 打字机字体  

  <U> 文字加底线  

  <UL TYPE>…</UL> 无序号的列表 ( 可指定符号 )  

  <VAR>…</VAR> 用于显示变量