Lecture 18: Legal and Accounting Basics for Startups

You don't have to get an attorney with a fancy office on Page Mill Road. But even if you did some big law firm, you would probably get a fee deferral that covers formation, so you're not out of pocket anything anyways.

I don't yet know the details of how Web sites get paid for running ads.

Of course, at venture firm KPCB, the Mary Meeker reports commonly claim that a Web site can get paid $2 per 1000 ads displayed -- additional details are missing. In my arithmetic, I assumed only $1 per 1000.

For Facebook, I don't get it: With some irony, for people who saw the movie, I find the f Web site brutal, excruciatingly brutal: I can't make any sense out of it -- I click and click, the response is really slow, and then the screen jumps around to whatever, why, to what, I have no idea. What the heck the "timeline" is, I have seen no definition, can't make sense out of it without some investigation, and don't want to do that. If Zuck wants to have some obscure UI/UX, good for him, but I'm not impressed. There are lots of goofy icons I can't spell, pronounce, look up in a dictionary, see clearly enough to recognize anything, etc.

For ads on f, I don't get it: To me Zuck is running f as some weird thing with little or no interest in ad revenue. If he is getting paid by the click, then so far he's made $0.000000 from me.

My Web site is all about niches, really just niches, for each user who arrives, as in drill down, zoom in, filter out, focus in on a highly personalized niche, for some quite fundamental reasons likely by a wide margin the most finely grained, personalized, niche-oriented Web site on the Internet so far and for a long time in the future.

And, yes, I have some associated ad targeting ideas.

So far, my UI/UX is just simple, dirt simple, childishly simple, about the simplest possible HTML with minimal CSS simple. So, no pull-downs, no pop-ups, no roll-overs, etc.

And, no icons! Wish I could stand on Mount Everest and shout with transcendental ecstasy louder than the blast from Mount Pinatubo "no icons!". Right, you guessed it; oh how you guessed it; I deeply, profoundly, bitterly hate and despise icons! No icons! At last, at last, free at last, no icons!

The screen never jumps for any reason. All the layout is from, right, just tables, with all sizes fixed and exact in terms of pixels.

The UI/UX is simpler than that of HN. ASP.NET writes a little JavaScript for me, but so far I have yet to write a single line of it and am eager to continue this way.

The UI/UX is so simple that a child of 6 who knows no English could learn to use the site in about three minutes just watching an adult, and an adult who knows no English could guess how to use the site in about two minutes. We're talking much easier than, say, Microsoft Word. And since the site is in English, for people who know English, sure, it's still easier to learn.

Each Web page is just exactly 800 pixels wide with high contrast and large fonts. The site should look fine on any device with a Web browser up to date as of about five years ago -- I'm not counting, maybe 10 years ago.

We're talking simple and, thus, really easy to use.

Each Web page sends for about 400,000 bits so that in a Web browser each page should load and format like "boom" or "pop".

Then, on each Web page, there is a banner ad of the standard size 720 x 90 pixels and down the right side an average of about four ads in the standard size of 300 x 250 pixels. And the ads do not get in the way of the utility or hurt the UI/UX. So, we're talking ballpark five ads per page.

Looks to me, page for page sent, compared with f, my pages are nearly a license to print money.

And, per user, the assumptions I made in the arithmetic should be okay for the users that do like my site. And some users, in the word in the movie, will find the site "addictive" -- something like a new game, and more addictive than, say a slot machine. The usage is highly interactive, especially for the part about drill down, .... One candidate early enhancement is a curious way to calculate a reward for skilled usage!

It looks like the main issue will be, will 2+ billion people like the site?

If people do like the site, then it looks like I have a shot at getting more revenue per user per month than f. Also my UI/UX is much easier to use than f! Heck, f's too tough for me! My hat's off to you, Zuck; I can't understand your Web pages or your site.

But thanks for the info on the $1.60. I've been writing software, nearly ready to start polishing and then go live, and collecting some initial data. Details of ad revenue will come later.

The result is, I've just had to do more of the work, say, serially instead of in parallel. That's not all bad: (1) I understand more of the work, really all of it, because I did all of it (the co-founder and I broke off before he contributed anything). (2) I saved time and effort coordinating with a co-founder.

Sure, a VC might say, "That your candidate co-founder didn't do much suggests that he was not really impressed with the project. In that case, neither am I."

But there's another explanation: Often some people just don't want to get along.

To get someone to work reasonably hard, one approach is to have a big, successful company, pay enough to support a wife and family, and have the wife with two young kids and with one more in the oven. Two guys in a garage don't meet this criterion.

I'm plenty motivated; he wasn't. My view is, for as much as he had to bring, still he had some serious flaws in his character.

So, find another co-founder? Not so easy. Easier just to do the work! For a co-founder, some of the usual recommendations are to select someone with good paper qualifications and have known and worked well with for years. Okay, my list of such people is exhausted. So, net, I'm a solo founder.

Again, the US, border to border, from crossroads, villages, and towns up to the largest cities, is just awash with solo founders of successful Main Street sole proprietorships. So, for a business where the initial work is just typing code into a computer to be used as a Web server, being a solo founder seems fully promising to me, even if VCs don't like it.