Startups: Live and Learn.

I Find Karma (
Thu, 5 Sep 96 06:20:42 PDT

It's from like a year ago, but there are some real gems in this
interview with Jerry Kaplan. For example:

> What entrepreneurship is about is looking at the world and saying, "I
> think things could be different and better, and I think I could
> assemble the resources necessary to make that happen."

By definition, Rohit is an entrepreneur. But, be cautious, grasshopper...

> I've learned not to mistake a lot of noise and smoke for real markets. I
> think we created so much noise that everybody thought there was a market
> there, when in fact it proved not to exist in any large quantities at
> that time. We were too early, and that's a very common problem for
> startups. But it would seem like a reasonable thing to do at the time.

However, this is no reason to despair:

> Start a company; it's a matter of trying it and if it works, great,
> and if it doesn't, oops.

And while we're on the topic, a startup is itself a biological being:

> The startup is a unique institution, and a different one. It's designed
> to accelerate the pace at which business evolves and mutates into ever
> more efficient and effective forms.

So, beware the pitfall that befell pen computing:

> The main thing you have to be is realistic, both about your own
> capabilities and about the true potential of what you're trying to do.
> It may be a great idea, but most of the companies die because they
> fail to address some supporting element around that idea, just to get
> enough focus to make it happen.

And in the end, it's either champagne dreams or nothing but ether:

> you either wind up with the right share for what you've done or you
> wind up with nothing because there was some imbalance in distribution
> of rewards that would cause the entire operation to collapse.

Full article follows.


Startups: Live And Learn
Jerry Kaplan On Ventures Past And Present
By Mary Eisenhart
Copyright 1995 by Mary Eisenhart and MicroTimes. All rights reserved.

When we last visited with Jerry Kaplan, he was on the cover of issue #81
with his colleagues Bill Campbell and Robert Carr. They were all very
upbeat about the prospects of GO, the startup Kaplan had founded to
pursue the vision of pen-based computing.

Kaplan recounts his experiences of GO's roller-coaster fortunes in his
current book Startup: A Silicon Valley Adventure[ISBN: 0-395-71133-9].
Described by its author as "a racy novel," it's a lively chronicle of
the industry's inner workings, from lofty ideals to darkest skulduggery.

And with the benefit of all that experience, Kaplan is now doing another
startup. This one's called OnSale, , dedicated to
some novel approaches to Internet-based marketing. Its motto, as we
discovered while visiting the Mountain View headquarters: Pursue the

> So Jer, after all these terrible experiences you had, why would anyone
> in their right mind want to do a startup? [laughter]

Well, Startups R Us. That's who I am.

And while the experience at GO didn't turn out positively in the end, I
wouldn't trade it for anything. I felt honored to have the opportunity
to work with the people that I did, and a chance to take my turn at bat
in creating a new arm of the industry.

There is a certain affinity that some people feel for that act of
creation. I think that every entrepreneur secretly is out to prove one
simple thing, which is that an individual can make a difference. And
once you've been there on the edge, it's difficult to imagine becoming
Vice President of Bull at some large company.

[laughs at look on my face] Is this what you're looking for?

> might be the first entrepreneur I've ever talked to who has
> said in so many words that the driving force is proving that an
> individual can make a difference. There's quite a number of them that
> I can't imagine making such a statement.

No, but I think in their heart they believe it. If you didn't believe
that, you wouldn't be an entrepreneur. What entrepreneurship is about is
looking at the world and saying, "I think things could be different and
better, and I think I could assemble the resources necessary to make
that happen."

Implicitly what you're saying is you don't always have to be on the
receiving end, simply consuming what others have created, or dealt out;
but rather you can change the way people work, change the way people
play, and change the way people think. If you don't have that passion,
you're probably not in it for the right reasons.

> So raw greed, the traditional motive, doesn't count from your
> standpoint?

For the most successful entrepreneurs I know--obvious people like Bill
Gates or Mitch Kapor and others--money is at best secondary in what it
is that they do. The old expression goes that money is merely a way to
keep score in the game.

I don't think that they did what they did because greed was the
motivation. I think they did what they did because they believe in
creating great product, and they understand the charge you get out of
seeing somebody's life improve through your hard work.

> So what have you learned from the GO experience?

I've learned a lot of things. I've learned to perhaps be a little more
cautious. I'm also a little bit older.

I've learned a lot about the way the industry works, the economics and
dynamics of the industry, which is a lot of what the book's about,

I've learned not to mistake a lot of noise and smoke for real markets. I
think we created so much noise that everybody thought there was a market
there, when in fact it proved not to exist in any large quantities at
that time. We were too early, and that's a very common problem for
startups. But it would seem like a reasonable thing to do at the time.

It was a wonderful experience. I don't regret it for a minute. It was so
wonderful I felt it was worth capturing for posterity by writing a book
about it.

I also wanted the world to see what it's like to be in the middle of a
startup. Because I think that it's not well understood--the excitement,
the personal commitment, and the challenges that one faces.

> One of the things the book does extremely well is document the human
> cost of the Silicon Valley startup culture--from relationships going
> kablooey, to people turning into werewolves from stress, to having a few
> hundred people lose their jobs because the grand vision didn't pan out
> and the VCs pulled the plug. It is a multi-box-of-Kleenex book in that
> sense.

[Kaplan cracks up] Hopefully it's also entertaining!

> Well, it is. But there's a part at the beginning where you're talking
> to your father about business, and you say that you could never have
> started this company forty years ago because the culture has changed,
> and it's no longer a matter of putting yourself on the line when you
> start a company; it's a matter of trying it and if it works, great, and
> if it doesn't, oops.
> I'm not convinced this is completely benign.

Well, it isn't completely benign.

The startup is a unique institution, and a different one. It's designed
to accelerate the pace at which business evolves and mutates into ever
more efficient and effective forms.

To accomplish that it is necessary, I think, to create a separation
between the concepts of personal fortunes and corporate fortunes. The
companies die, but they don't take the people out back and shoot
'em--the ideas live on, and the lessons live on through the know-how of
those people as they move from one institution to another. That's a
unique and, I think, extremely valuable form of cultural good.

I think we're very lucky to have developed the startup, and to be at a
place in time when those types of bets and experiments can get made at a
limited, though nonzero, human cost.

> What pitfalls would you advise the earnest young startup to avoid?

Oh, there's so many.

The first is, don't listen to me. [laughs] Make your own life.

I think you have to believe in what you're doing, but I think there are
a number of mistaken ideas that many young starting entrepreneurs have.

> There's such a thing as believing in your vision too much.

Yes, and that's one of the dangers. The main thing you have to be is
realistic, both about your own capabilities and about the true potential
of what you're trying to do. It may be a great idea, but most of the
companies die because they fail to address some supporting element
around that idea, just to get enough focus to make it happen.

Pen computing is a great example of that. There was a whole series of
things that had to happen around it in order to make it possible, and
some of them simply didn't occur. And if you're not there watching and
listening and being realistic, you're doomed.

Because startups exist in the ultimate reality. There's no way to live
in a fantasy world when you're living in a startup.

> Do you think the current pen platforms, Newton and Magic Cap, will
> survive?

I hope they do. Of course I have no stake and no special insight. I
think both of those groups have done an excellent job of articulating
and realizing a particular vision of the potential of mobile pen
computing. Both of them face significant challenges in their respective

I hope they can survive long enough to reap the fruits of all the hard
work of not only their own teams, but some of the earlier people who
were not able to hold out that long.

> It seems like General Magic-type startups need a lot more alliances
> and supporting infrastructure than something like Lotus would
> have required fifteen years ago.

Yes. It's a much more mature industry, and one that requires larger
amounts of investment today, in many of these things. But I think that
the characteristics of the early PC market are just now being recreated
in terms of businesses based around the Internet. While that may not
last long--it may only have a two- or three-year honeymoon--at the
moment there is once again that unbridled sense of opportunity, low
entry cost, and fertile ground for new ideas.

> Is there a reality-distortion field quality about working on a product
> that's several years, at best, from realization, while at the same
> time incredible amounts of money are pouring in to finance it? I can
> see where the sense of obligation would become crushing.

It wasn't crushing for one simple reason. Everybody involved sincerely
believed in what we were doing.

> But as you've said, belief isn't necessarily enough.

No, no. But we just didn't know. For many years, we always thought that
the next version of the product had a good opportunity to make it in the
market. And as I said in the book, in the end we had ample financing
opportunities, and yet really what occurred was more a failure of will,
because the management team, after so many turns of the crank--

> There's a feeling of disintegration toward the end.

Which I was trying to capture. Because what really happened, despite all
of the surface activities, was that the management team began to lose
faith that in fact we were close to finding a real market for the
product. That faith is what sustains the organization through difficult
times, and allows it to float across the chasm, as it were.

It was the loss of that faith, through the desire of the management team
to know the truth and to be telling the truth, that I think in some
sense was critical to the collapse of the whole enterprise.

At the time, we were working very hard. It was very exciting; it was
hard work but exhilarating, probably like climbing a mountain.

But we hardly felt pressure of a sort that you describe. We thought that
the market was just around the corner, throughout most of the life of
the enterprise.

> So you didn't get sleepless nights from the sense of investors
> breathing down your neck?

No. I have nothing but kind words for the investors. We were lucky
beyond any right to have the patience and the support of investors like
John Doerr and Vinod Khosla [of VC firm Kleiner Perkins Caufield &
Byers]. They stuck with it, and were supportive throughout the entire
process--not just the Kleiner Perkins folks, a lot of the others as
well. Were it not for that we would have been dead many, many years
earlier, and we would have made even more terrible mistakes than we did.

And I hope you'll include that in the article!

> There's no sense of Vulture Capital in the book.

No, it's really not the case.

Now, they're in business to make money. Venture capitalists often have
it in their interest to increase the risk-return profile in a venture,
while for the people inside, it may be in their interest to reduce the
risk-return profile. But my experience with our investors was that they
were as passionate and as committed as anybody else involved in the
project, if not more so. They were tireless cheerleaders throughout the
process, and are the antithesis of every negative stereotype one can
think of about financiers.

> What should somebody with a gleam in their eye and no money look for
> in a VC?

Well, the first thing to look for in a VC is money. [laughs] That may
sound funny, but they do run out of funds, or sometimes they're between

But I do have one very important piece of advice for people. People who
think they're raising money from VCs may relate it to the experience of,
say, getting a loan at the bank, or getting a mortgage. Nothing could be
further from the truth.

Your backers are your partners. You should choose them as carefully as
you would choose a spouse. If you don't like them, or your values are
not the same, you are guaranteed to have a messy divorce, if not fail.

It's important that you view it in that way, and that you interview them
and research their particular style and their approach, and understand
who is going to be working on your case, just as carefully as you would
in any close personal relationship.

> So should the bright young garage guys just go to Sand Hill Road and
> hang out and hope for the best?

Well, personally, I think they should get on the Net and Pursue the
Irresistible! [laughs]

> Well, okay, after they've hung out on your Web site and spent money they
> don't have because they're not funded yet!

Well, what should they do?

The first thing is to be realistic about the value of your idea. The VCs
will tell you about the value of your idea. You need to listen to the
people around you. I find a lot of potential entrepreneurs are so in
love with their idea that they're blinded to their true risks and

Others really are in it because they like to prove that they can run a
big company, and if you're not prepared to step aside or bring in the
critical talents that are needed when they're needed, you will not

A third factor is that a lot of people think that they don't want to
give up a lot of equity. But the truth is, even when you take in a
minority investor, you effectively give up control of your company.
Because once you have an investor, you have a legal and moral obligation
to manage the company in the interests of the stockholders as a whole,
and not your own interests. A lot of people think that having control
has to do with owning more than 50% of the company, and that's not true.
The only way to retain control is to own 100% of your company.

The other thing to recognize is that in building a company you're really
assembling a collection of resources and stirring them in a pot, all of
which are necessary and all of which will be paid for or compensated in
proportion to their natural value to your organization. And that there
are perfectly legitimate times when a person might have an idea, go to a
VC, and give up 90% of their company, because the value that's provided
by the VC may be 90% of the value.

Ultimately, as I learned from Dave Liddle, one of the most insightful
and brilliant people in the industry, you either wind up with the right
share for what you've done or you wind up with nothing because there was
some imbalance in distribution of rewards that would cause the entire
operation to collapse.

You witness the history of such bad deals as the Visicorp-Software Arts
deal--their inability to restructure a poorly crafted agreement really
was one of the major reasons for the success of Lotus Development

> Because its competitor was in disarray at a strategic time?

Right, and because the partners were working at cross purposes.

> Are there any things you would have done differently, knowing what you
> know now?

Oh, absolutely. I would have turned left when I turned right, I would
have swung when I ducked... There were lots of things I would have done

I think, perhaps to my detriment, that at the ripe old age of 43, I
would not have had the energy nor the, perhaps, recklessness to
undertake something like the establishment of the pen computing

A friend of mine, one of my colleagues at GO, used to say that software
is a young man's game. While I think I've chosen a very different kind
of battle this time, I think there's something to be said for his point
of view.

> I've also heard that startups are a young man's game...

I'm beginning to believe it. I haven't gotten much sleep lately. It's busy.

> So tell us about OnSale.

Well, the story of OnSale is a very interesting one. We're right now
writing the most fascinating chapter, whose outcome is yet to be fully

My partners are Razi Mohiuddin and Alan Fisher. Alan and Razi run a
product development and marketing shop called Software Partners.
They've been in business for about eight years; they've done a number of
very successful products. They're direct marketers of their products,
for the most part; the most visible is probably Street Smart, which was
not directly marketed, but is the Windows and Macintosh front end to the
Charles Schwab trading system. They're very, very good classical
application developers, and they now have a team of about twenty people
over here.

I talked to them about a year ago--actually this is in the book. It's
funny, it's coming true exactly the way we talked about it. They were
interested in where else they might apply this technology that they'd
developed in Street Smart, and we began to talk about electronic
markets, why they worked and why they didn't work.

We came to a certain point of view about them, particularly with respect
to the Internet, that each new medium, which I believe the Internet is,
like radio or television or print media, creates some new characteristic
way of selling that exploits its unique advantages, sells to a very
particular demographic, and is best for certain classes of goods.

To date on the Internet, most of what I was seeing was people simply
taking catalog content and repurposing that online. It only takes a few
minutes to discover just how boring that is. It has few advantages over
its paper counterpart, and numerous disadvantages. As a result, selling
on the Internet, or selling electronically, has gotten a lackluster

I sat down with my partners to think about what unique ways of selling
might be done on the Internet, and I believe we've identified some--we
have at least a reasonable hypothesis about how you might use that
medium in a much more effective way.

Every form of retailing is embedded in some kind of entertaining or
educational experience that creates a positive emotional experience for
the buyer. That's what brings people back, and that's why they buy.
Somebody recently said to me that if everybody who went to the store
simply bought what they thought they were going there to buy, retailing
would be dead and there'd be no malls at all.

So people obviously had their interest sparked by having a positive
shopping experience.

> What are these elements that are important on the Internet?

Well, first of all, the fact that prices and availability can change
instantly. So we designed a series of techniques to create a fast-action
sort of entertaining financial market, with a couple of formats that we
selected initially that allow people to play what amounts to a game in
the course of buying their goods. I don't mean like a video game; I mean
like a poker game, a game where they have to decide whether to buy now,
at what price they want to buy it, what strategy to take.

We've carved out an area where we have limited-quantity goods--they may
be collectibles such as fine wines, fine watches; we have cruises we're
selling; tickets to events; posters; or closeouts--right now we have
Apple computers that we're closing out. The common characteristic of
those goods is that they are available only in limited quantity.

Then we put them up in this fast-action market format, where people can
see how many we're selling, they can gauge the interest of other people,
how many other people have bought. Then when they bid or buy, their
initials, city, and state or country come up on the screen.

It creates a sense of community, which as you know is a key advantage of
electronic media. People have the sense that they're right here, right

> And they WANT that gold-plated Elvis and they want it now!

Damned if that doesn't happen. I got an email last week from a fellow in
Japan. He was bidding on a 1939 bubble-back Rolex watch. I believe it
was the first self-winding watch. This is a classic. It actually went
for $3100 on Friday--just to give you an idea what's going on here.

This guy wrote to me and said, "God help me, I must own this watch! But
every time I bid, MK of Weston, Massachusetts, immediately tops my bid.
What can I do?"

The perfect OnSale customer! Our tag line is, "Pursue the Irresistible."
That's what this is about.

It is plausible that we have created a place that people go, where they
are entertained, where it's fun, and where they become enticed into
purchasing special kinds of items that are of interest to them.

The second thing that we did that's different is instead of running
around to the usual retailers--Land's End, Tower Records, Sony Music and
Tapes, whatever, that you see everywhere all over the Net--we have
targeted the demographic on the Net with a very particular class of
goods: items of discretionary interest, primarily, to relatively
well-off technical males.

We're seeing a surprising amount of activity in high-ticket items,
which, like the gentleman in Japan, people just simply cannot resist.

> How long have you been up, and how many hits have you had?

We've been up for three weeks; the first week was a test, with a charity
auction for the Computer Museum. During that week, we had probably about
twenty-five thousand visitors. That's not hits; a lot of people quote
hits, and that's kind of a phony figure.

Most Web page retrievals require multiple "hits"--independent requests
to the server for information. Commonly, there is one hit for the main
text of the page, which may embed other references (commonly images),
each of which is a separate hit. Depending on how you design your pages,
what the customer thinks of as one retrieval may be ten or more hits.

Furthermore, the number of pages at a site visited by a given individual
may vary greatly.

It's easy to mislead the uninitiated by stating that your site gets
"five hundred thousand hits a day," which could mean fifty thousand page
retrievals, and two thousand people visiting, if they each look at
twenty-five pages on average. Two thousand doesn't sound as good as half
a million...

We had about twenty-five thousand visitors during that week, and we
raised about $30,000 for the museum.

Now I don't really want to say how much traffic and what's been
happening since then, but we are very pleased with the sales and the
visits we've been getting.

> Explain how this works. Suppose my great-uncle leaves me Stan Musial's
> cap--

We actually have a Stan Musial signed baseball right now!

> --and as it happens I'd rather sell it. What do I do?

Well, actually, you happened to pick the situation where I'd politely
say, "I'm sorry, but I can't deal with you."

The reason is, we want to deal with established, reputable merchants who
are capable of doing shipping, fulfillment, and customer service,
without creating ill-will or other problems for us.

There's a series of issues with respect to dealing with individuals. The
first is customer service; the second is quality control of the goods.
When you want to buy a new watch, you don't usually check the
classifieds; some people do, but you understand the kind of risks you
take when you don't deal with an established merchant.

Now I think when people are placing orders for significant amounts of
money over the Internet, one of their legitimate concerns is the
credibility of the source--are they for real, or is this just somebody
who's trying to get their credit card number and run? That's a real

I feel that dealing with reputable merchants is the best way to address
that issue. Not only would dealing with individuals be difficult from a
standpoint of getting them set up to sell their one little item--because
our costs are proportional to putting things up on the system--but I
think it would create a customer service nightmare.

> So you vet your suppliers pretty thoroughly.

Yes. We do. We've got a collection of suppliers who are of established

This is a very high customer service challenge, a very high customer
service business. I want everybody who buys to be happy, and most of
our merchants offer significant warranties or return policies.

Like, that watch was from 1939; it comes with a one-year warranty. An
individual could never do that, but I'm dealing with a specialty watch
merchant who's a great guy, and he's willing to provide that kind of
guarantee to the people who buy the watch.

One of the other aspects of the merchandising is that we have experts
online who will tell stories or explain their philosophy of collecting.
So, for example, Richard Paige, who's our watch guy--you get a picture
of him online, you can look at the items that he's selling, and you can
understand what sort of things you should look for in a fine watch.

We field a significant number of inquiries for these merchants, people
who just have questions. A wine guy, for instance--somebody writes in,
"I went to a restaurant and they pulled the cork and handed it to me.
What should I have done?" We collect those kinds of things, and
hopefully we'll post them online at some point.

The idea is to create a friendly, entertaining, fast-moving environment.
The mechanics are, we make deals with merchants, who have unusual items
that are likely to have discretionary appeal to relatively high-income
technical males...

> How do you find them?

The technical males?

> No, the vendors.

I sniff them out!

I ask around, I talk to people, and now that we're online, they write
in. That's what I do. We've collected a group of them, and I'm very,
very pleased with who we've got.

Richard Paige is a great example. He's local here in San Francisco; he
runs a shop on Union Street called Paris 1925. He's a fourth-generation
watch dealer, and a collector of jewelry, watches, and pens; he sells to
a very specialized clientele. We're able to give him a much broader
reach than he gets with foot traffic on Union Street--he's now able to
sell watches all over the world. We've done that now in the past two
weeks. He's incredibly excited about this.

Another one that's local is Ed Lehrman, who is the Director of
Adventure, as he calls himself, at the Passport Wine Company. Passport
is a local company that sells unusual wines, and that's also been doing
very well. Last week we sold three bottles of a 1987 Opus 1--$800
apiece. This will give you an idea of the size of some of those

The conventional wisdom is you can't sell on the Internet, nobody buys.
My belief is that you can sell; it's all in finding the right

> One of the reasons that's cited for not selling on the Internet is
> that there is no secure way of exchanging money at the moment. How
> are you handling that?

It is a big concern for the customers.

First of all, we happen to be using the Netscape commercial server, and
a lot of people have Netscape browsers, and that will encrypt your
credit card when it is transmitted.

My opinion, while I respect the contrary view, is that there is only the
remotest chance that your credit card floating across the Net is going
to get surreptitiously stolen. You would take a far greater risk buying
a pizza at the corner store than you do ordering on the Net.

But if people are uncomfortable about that, and some are, they simply
phone us up and I just literally sit there and put the order in as if I
were at their computer. I would say that probably 1% of the orders are
coming in that way.

> How do you make your money?

We receive a percentage of the sale, depending on the arrangements with
the particular merchant. Sometimes we get referral fees or other
arrangements; because we sell a broad variety of types of goods, those
arrangements do vary depending on the class of goods we're selling.

> What are your plans for the future of the company?

Well, we set out initially to demonstrate the proposition that people
will buy goods if they're offered in this kind of exciting and
entertaining format. We're three weeks into it, and I think the evidence
shows that that proposition is correct. We have lots of customers, and
the feedback we're getting from people using the service is they really
enjoy it, it's a lot of fun.

The question is where to go from here, and frankly I'm not entirely sure
of the answer to that. Firstly, we have a lot of work left to do in
terms of focusing and owning the particular way in which we merchandise.
We've already collected tremendous data on what motivates people to buy,
what sorts of things they will buy, how sensitive they are to price
points, and it's been fascinating to see that.

I think we want to tune up the model, and then we want to turn up the
volume. So far we're not too well known, being open for two weeks, and I
think that we have tremendous room for growth, because the netizens
themselves are growing, because the major services are opening their
floodgates to the Internet, and also because we can continue to improve
our merchandise selection and our merchandising methods.

Where we go from here corporate-wise--I frankly haven't been too focused
on that. I've been more focused on trying to ensure that the proposition
works, that we're offering good values, and that our customers are

> So you don't see yourself as going on the endless quest for funding
> you describe yourself going through with GO? It seems like a
> pretty different business.

It is, for a couple of interesting reasons. Firstly, in contrast to GO,
this has required a relatively small amount of capital, so my partners
and I have been able to finance it ourselves. That's not to say we won't
consider taking in financing in the future, but it's a very different
picture when you have a demonstrated ability to market than it is when
you're trotting around with an exciting but unproven pen computer.

If I had to guess, this selling proposition will interest several large
potential partners for us. I'm not averse to the possibility of forming
the right kind of alliance with one or more of those major partners.

> Where do you see them coming from?

There's a whole series of different places they could come from. We're
really creating a new distribution channel, and learning what the
characteristics of that channel are. A new distribution channel is
likely to have significant value for some large outfits, be they in
retailing, networking, the electronic publishing business, whatever.
We're open to discussion at this point. Of course, it's a very early

We're trying to be to the Internet what QVC is to television.

> How many of your customers so far are US vs. non-US?

On a three-week sample, I'd say it's 95% US.

> Right now, fulfillment to non-US customers is not a problem?

No, merchants have to deal with that, not me!

> But it's a factor in your business model.

It's true. I don't see any immediate difficulty with selling
internationally, with the exceptions of shipping and customs, and each
merchant is dealing with that as they see fit.

It's even a problem in the US, because you can't sell wines in certain

> Is it necessary to subscribe to your service?

No, not at all. Anybody can come in and browse, and only when you bid or
buy do you have to register.

> And at that point you give a credit card number?

Right. We check the card for validity, and that's the way it works.

> So how does this work? You announce that so-and-so's baseball emporium
> is offering a Mickey Mantle baseball for sale to the highest bidder,
> and bidding starts at $200 and is open for X period of time?

That's one of the formats.

We're using three initial formats. One is a standard auction, but the
standard auction has a twist. You actually tell us what you're willing
to pay, and then we bid for you. It's called proxy bidding. So I don't
put up your top bid right away. And you're guaranteed to get it at the
second bidder's price plus one increment.

So you might say to that Mickey Mantle baseball, I'm willing to pay 250
bucks. You'd tell me that, if we were selling one at a standard auction.
Somebody bids $100, you come in at $105. When that happens, an
electronic mail message goes out to the person who just got beat, and
says, "You've been beat, here's the story; you can respond by email." d
quantity up for sale, let's say five. There is a fixed time. During that
time period people bid, and the bids are right there on the screen.

At the end of the time period, the top five bidders get it, at the
lowest successful bidder's price, which in this case would be the
fifth-highest bidder.

> Now, think about the strategy that you need to have there. What do you
> bid if you want one of those?

You wouldn't want to bid really high or really low?

But you may want to ensure that you get it, and the price that you bid
is not the price that you pay unless you're the low bidder. The amount
you're willing to bid is really more a statement about how badly you
want the item, than it is about what you think it's worth.

And people immediately understand the dynamics of this, and have, I
think, behaved in very sophisticated gaming ways in order to deal with
these factors.

The markdowns are fun. We have a fixed amount of goods, and we have an
algorithm. Each day the price moves in response to how many we've sold.
We just sold one hundred refurbished Apple laser printers last
week--they went at ridiculous prices. The people at the bottom got 'em
for two hundred bucks.

If we sold more than fifteen in any given day, the price went up $25. If
we sold ten to fifteen, it stayed the same. If we sold less than ten, it
went down $25. So the price just floats until it finds the market.
That's the markdown way.

> So what kinds of stuff do you have up now?

We've got all kinds of stuff. We've got wines, cruises, event tickets,
movie posters, rock 'n' roll posters. We're about to add rock 'n' roll
memorabilia--I think that's going to be a big hot seller. Sports
memorabilia, and Apple computers, and laser printers, at the moment.
This is the current mix of goods.

> How do you plan to tell people that you exist? 95% of the population
> doesn't go where you are.


On the other hand, we can easily reach the 5% who do.

There are basically two means--one is standard publicity, the sort that
I'm doing with you now; the other is online publicity, which is also
very effective--people can click their way to us from different sites
and sources.

But the single most important thing, which has already been the case
even in just two weeks, is word of mouth. People enjoy it, they like it,
they come back. We're getting tremendous amounts of repeat traffic,
which to me is the most gratifying thing about the system.

Once you check through the Spiegel catalog, you're not going back next
week, because you know it's the same catalog. With us, it's a constantly
changing array of goods and prices and buyers. "Let's see what's
OnSale!" That was the theory behind the name.

> Are your VC pals attracted by this?

I think they're very much attracted by it in general. Many friends have
told me that they're investing in "Internet ventures." But it's like the
early days of the PC, it's so unstable and so uncertain.

It's very difficult for a seasoned investor to get comfortable until
some kinds of boundaries are set up, and some kinds of markets are
defined. So I think that while there's great interest, and a lot of
learning going on on the parts of venture capitalists, I think there's
relatively little in the way of actual investment at this time.

There's investment in companies like Netscape.

Tool companies are a lot easier to understand in this arena, which is
true in a lot of industries when they first occur. There was always a
market for tools for pen computing, because you're selling to

You're selling the pans to the miners, and that's clearly a business.
You don't have to believe that the way to get rich is to pan for gold in
order to make money selling Levi's.

So that's a clearly defined market that therefore is attracting more
investment from traditional sources.

> Do you see much investment in content rather than tools?

That's even more esoteric. That's less of a typical VC opportunity, but
that's because those are specialized markets, where the knowledge and
expertise of each market is very much distributed.

Where you see content plays being made, it's usually in conjunction with
a partner who owns that content in other domains.

> So it's more like a licensing arrangement, than deciding "I'm going to
> fund the first digital movie."


> Angel-type financing arrangements are more common for content?

That's my sense of it. Or else it's just flat-out owned by content

Before people get discouraged, they need to recognize that it often took
more than a decade for other media to develop a viable economic model
that has endured. The fact that radio is an advertising-sponsored
medium--I think it was twenty years from the time when commercially
plausible radio receivers were available to the point where that model
emerged. It's a fascinating story.

I'm a little less familiar with what happened with television, but I
know that the guy who invented television was asked, "What do you think
this will be used for?"; and he had in mind things like teaching new
surgeons how to perform operations without having them in the operating
rooms where they might contaminate the operation.

He certainly didn't have in mind Gilligan's Island. Nonetheless, the
whole model of Gilligan's Island being the filler to keep the eyes on
the screen between the ads for Oreo cookies is inobvious, and took many,
many years to occur. I think we'll see the same thing with electronic
media--is this an advertising medium? Is this a subscription medium? Is
this sort of a nickel-and-dime shop? Do we need new forms of electronic
currency, or not? These are all issues that will play out over the next
few years.

> For one thing, electronic media are demand-driven rather than based on
> the model of broadcasting information to an essentially passive audience
> of couch potatoes.

I think there's a fundamental difference between radio/television and
interactive online media, and it's precisely what you were saying.

And that's why I don't think televisions and computers are going to
merge. One of them is a passive activity, the other one is an active
activity. They attract different kinds of people, or at least people who
are involved in different kinds of activities.

Different states of mind.

That's a good way to put it, they deal with different states of mind. In
the interactive domain, you're intellectually engaged in what you're
doing. You're a part of what you're doing. So how to deal appropriately
with people whose minds are engaged is a novel challenge for the
traditional media.

> What are you finding works?

One of the shocking things that I've found so far is that people are
very smart.

> Even given that you've got an unusually bright demographic?

What I'm referring to is that in the games that we're creating, the
environments that we're creating--and we do this full-time--we've given
our best thought to how to design them correctly.

And nonetheless I find that customers are extremely sophisticated about
what strategies they ought to employ, when and how to buy. They taught
us a great deal in the first two weeks.

Which shocked me. I thought it would be difficult for people to get
their arms around what this was and what they ought to do. But they
showed us a thing or two right off the bat about what made sense and
what didn't, by buying or not buying. You're voting with your dollars in
a system like this.

We found people to be very knowledgeable and sophisticated in terms of
the goods. Things that were not priced correctly just flat-out didn't

The interesting thing is we get that feedback and that information very
very quickly, in a way that traditional channels do not. So even the
merchants, who in theory know their markets, would suddenly find that a
particular type of goods that historically had had a certain kind of
selling price and pattern may very well have changed rapidly, and now
they have a way to respond to that, where they never could or would
respond previously, because that information wasn't available.

An interesting example of this--the Mickey Mantle baseball that we're
selling this week. At the time of this interview, Mickey Mantle is in
the news, having just taken ill. That has dramatically affected the
value of a signed Mickey Mantle baseball. If we were putting out a
catalog, we could have done nothing about that, but because we're an
interactive medium, we could immediately gauge that demand and adjust
the offerings to the market.

> OnSale doesn't present the same kind of technical challenges GO did,
> though.

No. The current project has no technical risk. It had high market risk,
and that market risk is dropping rapidly.

> Where would you like to be with this in a year?

In a year I'd like to see us having established ourselves as a
significant force in retailing. And I think that if we work hard and
have our reasonable share of luck, and if we treat our customers well,
we have an opportunity to do that.