The
    Deal
    on the Table
     After talking to virtually every venture capitalist in the
     business, in May of 1984 it appeared that we were finally going
     to close a deal.  Frank Chambers, who had been introduced to us
     by Jack Stuppin, indicated that he was willing to make an
     investment in our company.  Despite our cynicism, born of endless
     tiresome and fruitless meetings which consumed our time when it
     was desperately needed to develop the company, we were eager to
     obtain funding we could apply to increase our marketing efforts
     and seize the market before better-funded competitors entered the
     fray.
     Mike Ford described the situation as “being in a verdant field
     with gold bars lying all around.  The question is how many can
     we throw in our pickup truck before a big vacuum cleaner comes
     down from the sky and sucks them all up.”  In essence then,
     being greedy and aggressive suckers ourselves, we wanted the cash
     to nail down the largest possible market share while we still
     could.
     After numerous and lengthy discussions, Frank Chambers
     communicated the terms of the deal.  What follows is a
     transcription of Dan Drake's notes taken while hearing the terms
     over the phone, with annotations explaining what the terms mean.
Terms I:  Frank Chambers
-  $500–700K.  Size of the investment in dollars.  This would
represent about 1/10 of the company.
-  $2 preferred, convertible to 1 share common.  The investment
would be preferred stock, so that if the company failed the investor
would get his money from the remains before any of the common
shareholders.  The stock could be converted to common stock at any
time.  This is conventional in venture capital deals.
-  If not liquid in  4 1/2 years, we offer to repurchase at
2× price (plus accrued dividends).  If the company did not
go public or get acquired, thus providing an opportunity for the
investor to “cash out”, we would buy back the investment at twice
its original value.
-  8% dividend from 2/1/85; with majority of preferred holders or
the board able to substitute change in conversion ratio.  We
would pay this dividend from the company's earnings.  If we didn't
have the earnings to pay the dividend, the foregone income would be
used to increase the venture capitalist's ownership share of the company.
Most venture capital deals work this way, but it's unusual to have the
dividend start so soon after the deal.
-  One demand registration, unlimited piggyback.  The investor
could, at will, demand that we make a public offering in order to
enable him to cash out his stock.  The company would bear all of the
costs of the public offering.  In this, or any other offering the
company made, the investor would be able to sell his shares
“piggyback”, up to the total amount of the shares owned.
-  Options.
    
    -    110,000 at $.75.
-    Remainder at least $2 + 5% per 6 months.
-    Dilution protection, unspecified, above 500,000 options.
-    Vesting at least 4 years.
-    Forfeited options are canceled, not returned to pool.
 
-  Board of directors.  7 people: preferred elects one, one by
agreement of preferred and common shareholders.  Jack Stuppin to be on
the board.  Advisory committee to the board
up to 3 people chosen by the investor.
-  Representative of preferred investor approves: “All
compensation matters”, and all capital expenditures over $25,000.
-  John Walker: key man insurance $500K; employment contract, 2
years, non-competition.
-  Frank Chambers approves investors.
-  Preferred has first refusal on private equity offerings.
     We didn't like it.  While many of the terms were conventional and
     were what we expected, several totally unexpected constraints on
     our ability to develop the company in the way that had brought us
     to the present point were contained in the deal.  In particular,
     our ability to grant stock options to new employees was severely
     constrained by limits on the number available, by forcing the
     option exercise price to above the price paid by the investor
     (who received much better terms on his preferred than the
     employee would on his common stock), by retiring from the pool
     any options granted to an employee who subsequently left the
     company, and by imposing a four-year vesting period on all
     options, which the founders of the company felt transformed the
     options from their original purpose of allowing employees to
     share in the company's success to a kind of twentieth century
     indentured servitude which compelled employees to stay with the
     company or face forfeiture of their financial gains.
     We also thought that the general tone of the deal was far from
     consonant with the percentage of the company being
     purchased and the demonstrated performance of the company to date
     and the track record of its managers.  But we still wanted the
     cash.  So…we came back with the following suggested terms.
 
Terms II:  Us
-  $500–700K.
-  $2 preferred.
-  Repurchase at 1.5× if not liquid in 4 1/2 years;
call on preferred at $4 + 20% per year.  The call provision
would allow us to forcibly buy out the investor for twice the
investment plus an increment of 20 percent per year.  This was meant
to be symmetrical with the repurchase provision benefiting the
investor.
-  8% dividend from 2/1/85; reduction in conversion price on
missed dividend.  This is equivalent to the original requested
terms, just more lucidly put.
-  One demand registration if proceeds
 exceed $X, X >= $5 million?
-  Options: dilution protection by issuing proportional shares at
the same price as exercised options.
Vesting set by board when option issued.  No cancellation of
forfeits—return to pool.
-  Board of directors.  7 people: preferred elects one, common
elects 6.
-  Ceiling on executive salaries and bonuses until some numbers
(sales and profit) achieved; override by preferred representative.
Preferred representative approves capital expenditures over $100,000.
-  John Walker: key man insurance; contract, no non-competition.
Being an ornery S.O.B., I said that signing a non-competition
agreement with a company in which I was the largest shareholder was
beneath my dignity and that I wouldn't do it.  Moral: don't pick an
asshole to be president of your company.
-  Frank Chambers approves investors, but present stockholders can
buy in subject to $ ceiling and regulatory problems.  We didn't
want the venture capitalist to be able to veto further investments by
people who got in before he did.
-  In general, the constraints have drop-dead provisions.  This meant that every constraint on our freedom to run the
company would expire when we reached some well-defined performance
milestones.
He didn't like it.  So, we got together and attempted to come up
with another offer which would be acceptable.  Here it is.
Terms III:  5/10/84
-  $500–700K.
-  $2 preferred, convertible.
-  Repurchase terms OK (?)  We acceded to the original
terms.
-  Forced conversion if we get to $10 million (?) annual sales and
$1 million annual profit or make a public stock offering over $5
million.
-  One demand registration, piggyback, after 1/1/87 for
>$5 million.
-  Options: 100,000 shares at $.75, 200,000 shares at $1.60,
        200,000 shares at $2.00, 200,000 shares at $3.00, 200,000
        shares at $4.00.  Majority of each class of stock to approve
        any new plan.  Vesting at best 50–25–25 after the first
        100,000 shares.  This meant that after the first 100,000
        shares of options (which were committed to fulfilling options
        we'd already granted to people), those who received options
        would receive 50% of their shares the first year, 25% the
        second year, and 25% the third year.  Forfeited options
        return to pool.
-  Board: ask Chambers*, Ellison, Stuppin.  (? Unsettled).
      What are the terms on paper?
 * If not on board, an advisor.
-  $75,000 ceiling on officer and director salaries.  Override by 6
or 7 directors.
-  $25,000 on capital expenditures.  Override how?
-  John Walker: key man insurance; employment contract to “devote
substantially all his time”.
-  Frank Chambers approves investors.  Common holders can buy in
(n.b. might have to keep it intrastate to avoid sophisticated investor
problems).  Since people who joined the company in the beginning
and worked themselves to exhaustion to build it to the point that
venture capitalists were interested in investing were not ipso
facto “qualified” to risk their savings by investing
in their own company since they were not already wealthy.  This is an
example of what was referred to in the 1980's as the
“opportunity society”.
-  First refusal on private equity offerings.
     After these terms were presented, it was clear that we would
     never come to an agreement on the issue of awarding stock options
     to employees to give them a real stake in the company's success.
     In addition, the overall flavour of the deal seemed to us totally
     inappropriate for a company which was, at the time of these
     negotiations, generating sales equal to the size of the deal
     every month and generating after-tax profits close to the size
     of the deal every quarter.
     We couldn't believe that this was the best deal obtainable for
     venture funding of the company, and we were inclined to ask
     around to see if this was reasonable.  But, our Distinguished
     Financial Advisor informed us that this would constitute
     “shopping the deal” when “a deal was on the
     table” which was right out by the genteel standards of the
     venture community, and that he could not countenance such
     unrefined behaviour (notwithstanding the fact that in the real
     world this kind of collusion is called “conspiracy in
     restraint of trade” and people go to jail for it).
     So, after a brief weekend meeting in which we discovered we all
     agreed on the obvious conclusion, we decided to graciously
     decline this generous offer of funding and carry on with our own
     resources.  Upon hearing this decision, Jack Stuppin said that if
     we didn't take this deal, he did not wish to be a shareholder and
     wanted us to buy him out.  Not wishing to deplete our treasury,
     we declined. In not accepting our terms, which differed from his
     original proposal primarily in issues of philosophy, not money,
     Frank Chambers chose to forgo an investment of $500,000 which, if
     held until the stock price hit its 1987 high, would have
     appreciated to more than $37 million.
 
     Peter
     Barnett drew this geological strata illustration in 1984 to
     demonstrate the multiple dot and dash line types introduced in
     AutoCAD 2.0.  The drawing doesn't represent any real geological
     formation—it was made up out of thin air, not hard rock.  It
     has been used as a sample drawing from AutoCAD 2.0 to date.
