The Problem

Perfect Wheels existed because of a problem I had as a national caliber mountain bike downhill racer, and the problem has a date. In July 1994 I entered my first race ever – ever – the UCI World Cup and NORBA National Downhill Series Championships held on the infamous Kamikaze downhill at Mammoth Mountain.

Why start at the top? Because I already knew I was faster than anybody in my zip code and the neighboring trail systems. A local “win” at the bottom of an unsanctioned trailhead tells you nothing. If you want to find out what you actually are, you go play with the big kids in the deep end of the pool and use a clock this time. So that’s where I went.

The bike was assembled the night before and I had never ridden it. It was an 18-inch frame; I’m 6’3″ and leggy and ride a 26-inch custom – racing that thing was like racing a BMX frame. I never rode the course in practice. I just went for it, and took eighth in my class at a NORBA National. The certificate – 07/08/94, Mark Nichols, eighth, 6:10.14 – is below at bottom of this webpage.

The next spring I blew up a front wheel, and the sport showed me its dirty secret: a race-quality replacement was thirty days out, custom from a wheel guru – because hand-building a true race wheel was a mystic craft, and in the entire Bay Area there were maybe three guys with that kind of Jedi talent.

Race wheels did not sit on shelves or hang in racks in any bike shop in America. Race wheels were bespoke, period, and the industry assumption was that racers would only ride custom. I said: that’s bullshit. I will fix that. I cannot be the only person getting disrupted by this. I ain’t got time to play that kind of game – so I went and fixed all of it.

The archived close-out list below is the receipt on that bet: an in-stock inventory of prebuilt race wheelsets at fixed prices, selling to the public and to dealers, with one-day turnaround on customs. The market answered.

The Genesis

Understand how new this was: commerce on the internet only became a legal option in 1995. Until April 30 of that year, the NSFNET Acceptable Use Policy barred commercial traffic from the internet’s backbone – you could not lawfully run a store on it. Only when NSFNET was decommissioned and the backbone spun off to commercial carriers like MCI and Sprint did “ecommerce” become possible at all. Amazon shipped its first book that July. eBay launched that September. That is why so few ecommerce sites existed – not because merchants lacked imagination, but because until April 30, 1995, the thing itself was against the rules of the network. Perfect Wheels launched into that same first year.

Fresh from Sprint data/internet training, I put up perfectwheels.com – an ecommerce website selling custom bicycle wheels and components, one of roughly 7,000 non-adult websites on an internet of about 23,000, most of which were adult-oriented. But Perfect Wheels wasn’t the point. It was the genesis: the enterprise that proved the point. (perfectwheels.com as first captured by the Internet Archive, April 15, 1997)

Perfect Wheels had two addresses, and the difference between them is the story of the business. It started at my home at 1127 Livorna Road in Alamo, in my four-car carport, which I wrapped and doored into a working garage. A carport wrapped into a garage has no heat and no air conditioning: winters, we built wheels around a quartz heater; summers, a twenty-inch floor fan pushed the air around. As orders grew I hired employees – two or three of them, building wheels in that garage, at a house that also held my five kids. Using a bathroom became a lottery. In the second year, when we started making decent money, I pushed the operation down the road to the closest retail location to my house – 3006 Danville Blvd – and we had civilization: walls, plumbing, climate control, a storefront. That storefront is the address frozen on every archived page below, because by the time the Internet Archive’s crawler first found the site in April 1997, the garage operation had already outgrown the garage. The capture documents the second act; the first act ran on a quartz heater. The internet of 1995 let a garage present itself to the entire world as a business. What it could not yet do was carry the world’s payments back to that garage – and that gap is what this page is about.

And the first act has its own receipt. Preserved on the Wayback Machine is the original Perfect Wheels business card: PERFECT WHEELS – Mark Nichols, Principal – ofc 510.944.3573, fax 510.944.0803 – 1127 Livorna Road, Alamo, CA 94507. The garage era – name, title, and home address – archived by a crawler that never knew what it was saving. The logo was a wheel drawn so lean it reads as a crosshair centered on the axle. And the tagline was “True Professionals” – mine, registered, and defended: when Wenzel Wheels tried to copy it, I enforced my rights and made it stop. True story, from the True Professionals.

Look at the archived catalog pages below – three manufacturer lines, hand-coded, photos, specs down to the gram. Each page was a grapefruit, and in 1995-96 it had to be sucked through a straw: twisted-pair copper at 28.8 kbps, and for most of America the straw ran through AOL. A customer waiting half a minute per product photo is not a customer completing a transaction.

Why the Website Existed

Everyone had the idea. When I went to Mavic, DT, White Industries, Nukeproof, TNT, Phil Wood, AC, all of them, for distributor pricing, not dealer pricing, I heard the same line twenty times easy: “yeah, I thought of that idea years ago too.” Sure you did. But nobody pulled the hundred grand to launch it. Distributor pricing was a buy-in: Mavic and DT alone each wanted a minimum $25,000 initial pre-order, then roughly $10,000 in hubs per manufacturer to unlock distro pricing. That was the cost of the ticket. The idea slows nobody down; the capital does. I wrote the checks. Remember that lesson, because it comes back at Digital Island: a global IPLC network is the same story at a bigger number, and that ticket needed Cliff Higgerson’s wallet.

Now do the retail math. My market was every bike shop in America, roughly 3,000 potential dealers, plus retail riders. The trade paper was Bicycle Retailer and Industry News, and I could not afford a display ad in it. Not on a $60 gross profit per wheelset; that kind of dough was waste. So I bought the smallest space they sold: a one-inch by two-inch classified in the back pages, and all it did was promote the website. That is why the website existed. The web was the only place in 1995 where a garage operation could publish a full national catalog, every hub, every rim, every spoke, every weight, for the cost of a phone line, and a tiny classified could point 3,000 shops at it. That is also why the component pages priced everything “See Your Dealer”: the site was selling to shops as much as to riders. The website was not a novelty. It was the cheapest national storefront that had ever existed, and the classified ad was the doorbell.

And let me make this as clear as I can, because I have been saying it for thirty years (sorry, no document proves this one; it is testimony): I would not have started Perfect Wheels at all if not for the internet. Not for ecommerce. For advertising. The distributor buy-in I could fund. The wheels I could build. But a national business you cannot afford to tell anyone about is not a business, and national advertising in the trade press cost more than the margins could ever carry. The internet was the first affordable national advertising medium a garage operation could buy, and that, not the transactions, is what made the company worth starting. Which was, of course, sorta kinda a new thing.

How “Ecommerce” Actually Worked in 1995

Here is what nobody tells you about ecommerce in 1995: the website was the catalog, not the transaction. The site’s own ordering page printed the transaction channel in its header, verbatim: “To Order Call 1-510-944-3573 or Fax 1-510-944-0803.” That was ecommerce. Note everything missing from that header: no email, because in 1995 even email wasn’t trusted enough to carry an order. No text. No chat. No cart. No checkout button. The transaction didn’t just leave the web; it left the internet entirely, all the way back to a ringing phone and a warming fax machine. Customers faxed in orders with their full credit card information sitting on the fax paper, or read their card number out loud over the phone, and I punched it into a credit card terminal on my desk. Card-not-present. No encryption. No verification. Full card data on paper in a tray, in every merchant’s office in America doing “internet commerce.”

It was credit card fraud dreamland. And running a real storefront on that network is how I learned exactly where the internet broke – not in theory, in lost sales and naked card numbers.

That is why I invented merchant transport: put the transaction itself in the browser under SSL, and carry it over IPLC CBR – international private leased circuits at constant bit rate – so the transaction was encrypted end to end, private, deterministic, and fast enough that a merchant and a customer could both bet on it. The problem Perfect Wheels exposed in 1995 is the problem Digital Island was built to solve in 1996.

The Evidence: The Archived Catalog

The Internet Archive captured perfectwheels.com 75 times between April 15, 1997 and 2026. The earliest captures preserve the catalog as it ran – including one page still showing “Blah Blah … x g” placeholder text. The Archive crawled the site mid-build, frozen exactly as it was. Nobody backfills fake history with unfinished placeholders; that is a one-man shop hand-coding HTML catalog pages at night.

The Business: The Close-Out List

The archived close-out list is the commerce itself – built wheelsets priced and ready, sold “to the public and dealers at the Perfect Wheels assembly site located at 3006 Danville Blvd, Alamo, CA 94507.” The page carried its own warranty policy, verbatim: “NOTICE: All wheel warranties are to be done by the component manufacturer. No exceptions. Remember, we only assemble them. We don’t make the parts.”

Custom Builds: One-Day Turnaround

Beyond the close-out stock, the site ran a full customization program. The upgrade page, verbatim: “To fully customize your wheel order, substitute the products on this page in place of the ‘Standard Wheel Components.’ … The typical turn around time on all custom builds is just one day! … you can send us your own hub and we will build it complete for $50 per wheel. … If you want something that’s not listed… just call us and we’ll get it.”

The component catalog behind that program ran deep: spokes from DT (Competition, Revolution, Champion, Alpine, and Aero blades), DT/Ritchey Logic, Wheelsmith, and Marwi titanium in natural, blue, purple, green, gold, Ti-Dyed, and bladed versions; brass, alloy, and Spline Drive nipples in any color; and rims from Mavic (217, 220, 121, 238 – including ceramic and hard-anodized), Ritchey, Sun, Syncros, and WTB, every one specced by weight in grams and drilling count. Component pricing was listed as “See Your Dealer” – Perfect Wheels sold to the public and through a dealer channel.

The Mavic Connection: Proving the Prebuilt Wheel

Look back at the close-out list and count the Mavic rims – 217s, 220s, Reflexes, CXP-30s on nearly every build. That’s not a preference, that’s a position: I held the USA distributorship for Mavic rims, and that distributorship became worth more than the entire wheel-building service. But the rims are the smaller half of the Mavic story.

Perfect Wheels existed because of a problem I had as a racer, and the problem has a date. In July 1994 I entered my first race ever – ever – the Kamikaze downhill at Mammoth Mountain. The bike was assembled the night before and I had never ridden it. It was an 18-inch frame; I’m 6’3″ and leggy and ride a 26-inch custom – racing that thing was like racing a BMX frame. I never rode the course in practice. I just went for it, and took eighth in my class at a NORBA National. The certificate – 07/08/94, Mark Nichols, eighth, 6:10.14 – is below.

The next spring I blew up a front wheel, and the sport showed me its dirty secret: a race-quality replacement was thirty days out, custom from a wheel guru – because hand-building a true race wheel was a mystic craft, and in the entire Bay Area there were maybe three guys with that kind of Jedi talent. Race wheels did not sit on shelves or hang in racks in any bike shop in America. Race wheels were bespoke, period, and the industry assumption was that racers would only ride custom. I said: that’s bullshit. I will fix that. I cannot be the only person getting disrupted by this. I ain’t got time to play that kind of game – so I went and fixed all of it. The archived close-out list above is the receipt on that bet: an in-stock inventory of prebuilt race wheelsets at fixed prices, selling to the public and to dealers, with one-day turnaround on customs. The market answered.

And the story closes where it opened: on a mountain. In 1999, on a bike that actually fit me, riding wheels from the world I’d rebuilt, I won my class at the NORBA Nationals at Deer Valley. The champion’s jersey and the gold medal are below. Five years from first race ever, on a wrong-size bike I’d never ridden, to class champion. And the company born from the blown wheel in between was sold that same year.

One more truth for the record books. The only reason I had the time to train and race that national was that Digital Island fired me that May, three weeks after the S-1 was filed and weeks before the IPO I’d helped build toward since 1996. They took my unvested shares. The calendar they handed back bought me a summer on a mountain and a gold medal. That is finding a positive in a negative, for the record books.

Then Mavic – the company whose rims anchored my inventory and whose USA rim distribution ran through my shop – came out with the Crossmax: the factory-prebuilt performance wheel. The model Perfect Wheels proved at retail scale became, in Mavic’s hands, the industry standard, and prebuilt wheelsets are how virtually all performance wheels have been sold ever since. I was the test market for whether racers would buy prebuilt. They would. Ask anyone who has bought a bicycle wheel in the last twenty-five years.

And Mavic wasn’t alone. Look at the spoke column on the close-out list – DT on nearly every build. DT Swiss ran the same recipe, standing up their own race-wheel operation out of Grand Junction after Perfect Wheels was moving $100,000 a month in dealer wholesale. That wholesale channel was its own proof of the model: Jenson USA, BikePro, Universal Cycles, Cambria, Supergo, and Mail Order Sports in Petaluma were all Perfect Wheels customers. The two suppliers whose components anchored my builds each watched the prebuilt model work through my sales, then built it themselves. And when Perfect Wheels sold in 1999, it sold to one of those wholesale customers – Universal Cycles, the largest of them. After that sale, the rest of the bench followed: White Industries, DT, Nukeproof, Phil Wood, the manufacturers whose hubs filled my catalog, all began selling complete wheels direct. The model born from a blown front wheel and a thirty-day wait became how the industry sells wheels, period.

Why It Matters

Perfect Wheels is the genesis of merchant transport. Every design decision that followed – SSL in the browser as the transaction surface, IPLC CBR as the carriage, deterministic sub-300ms round trips as the standard – traces back to a credit card terminal on a desk in a converted carport and a fax machine full of card numbers. I didn’t theorize that the 1995 internet couldn’t carry commerce. I ran a store on it – garage assembly, retail floor, web catalog, phone and fax transactions, the whole stack – and it told me. A year later, the answer to what it told me was Digital Island.

And Perfect Wheels itself? It ended the way a real business ends: in 1999, I sold it to my largest customer, Universal Cycles. Mark the year. That same summer, weeks apart: Digital Island – the company built on the lesson PerfectWheels taught – went public on NASDAQ, and I won my class at the NORBA Nationals at Deer Valley. The experiment that started with a blown front wheel and a carport on a quartz heater ran four years, moved to a storefront, changed how an industry sells wheels, and closed out in a single season as an acquisition, an IPO, and a gold medal – while the lesson it taught was already carrying transactions around the world.

Source: Internet Archive, perfectwheels.com, first capture April 15, 1997 – captures from 1997 through 1999 document the business under my ownership, through its sale to Universal Cycles; www.perfectwheels.com/index.html captured 4 times, April 15, 1997 – February 19, 1998.

Historical Webpage and Competition Archives