The Journey Ahead — Part 5: Getting Paid

At the end of the previous chapter, I mentioned something that sounds almost embarrassingly simple: getting paid.

I had assumed this would be one of the easier parts of building NEXO Timepieces. People have been accepting credit cards on websites for decades, after all. You create the website, connect a payment processor, pay a fee, and a customer can buy a watch. That was approximately the extent of my understanding when I started looking into it, and, as has happened more than once during this journey, the interesting problem turned out not to be the one I thought I was solving.

For a small business selling expensive watches, accepting a payment is not particularly difficult. The difficult part is creating a transaction in which two people who may never have met can both be reasonably confident that they will still have their money—or their watch—when everything is over.

Imagine the transaction from the buyer's perspective. You find a watch online for $10,000. The photographs look good, the description is convincing, the seller appears knowledgeable and legitimate, and perhaps you have spent some time on the website and learning who the seller is. There is only one uncomfortable detail: you have never met this person.

A wire transfer is relatively inexpensive and, once completed, extremely difficult to reverse. From the seller's perspective, that is not a problem. From the buyer's perspective, “irreversible” has a somewhat different emotional quality when $10,000 is about to disappear from a bank account and travel to someone known primarily through a website.

A credit card changes that equation. Suddenly there is a familiar institution standing between buyer and seller, along with fraud protections and a mechanism for disputing a transaction. Whether those protections are ultimately needed is almost beside the point. Their existence gives the buyer additional confidence, and that confidence has value.

The interesting part is that the risk has not disappeared. It has simply moved.

From my side of the transaction, I may now ship a $10,000 watch and receive confirmation that it was delivered, while the possibility of a dispute continues after the watch has left my possession. Perhaps the card was stolen. Perhaps the buyer claims the package never arrived. Perhaps the watch is alleged to be different from what was advertised. The dispute may be entirely legitimate, or it may not be, but either way the transaction has acquired a life beyond the simple exchange of money for a watch.

That was when I began to understand why expensive watches occupy such an uncomfortable corner of e-commerce. They are small, portable, valuable and relatively easy to resell. A single transaction can involve thousands or tens of thousands of dollars, and for a small seller one fraudulent transaction is not an irritating percentage buried somewhere in an annual report. It can be a meaningful event. Payment processors understand this as well, which explains why the innocent question “How do I accept credit cards?” quickly led me toward considerably less innocent questions about fraud, verification, chargebacks, proof of delivery and who ultimately carries the risk when something goes wrong.

There was another question hidden among them that I found even more interesting: how much friction can I introduce before the protections designed to make a transaction safer begin making legitimate customers uncomfortable?

A customer spending a substantial amount of money should not feel as though he is being interrogated simply because he wants to buy a watch. At the same time, it would be foolish for me to treat a five-figure Internet transaction as though someone were buying a T-shirt. Luxury purchases depend heavily on confidence, so the process has to protect both parties without communicating suspicion to either of them.

Thinking about the problem this way gradually changed how I looked at payment methods themselves. I had thought of cash, wires, credit cards, marketplaces and escrow primarily as different ways of moving money. I now see them more as different ways of allocating trust and risk.

With cash, most of the uncertainty is concentrated at the moment of exchange. Is the money genuine? Is the watch genuine? Once both parties are satisfied, the transaction is essentially finished. A bank wire gives the seller considerable certainty once the funds have been received and verified, but asks the buyer to accept much more uncertainty before sending them. A credit card restores some protection to the buyer, while extending some of the seller's uncertainty beyond delivery. Marketplaces and escrow services introduce another party into the transaction and, with it, another set of protections, procedures, costs and rules.

None of these methods is inherently the right one for every transaction. They solve different problems, and that realization changed the way I began thinking about the NEXO website as well.

Originally, I viewed accepting credit cards as a feature. Customers expect businesses to accept cards, therefore the website should accept cards. Simple enough. But NEXO is not a supermarket, and there is no particular reason why every transaction has to happen in exactly the same way simply because the technology makes it possible.

The website can serve a much broader purpose. It can establish identity, show inventory, provide detailed photographs, explain condition, preserve documentation, state policies and, through the articles I write, give someone a surprisingly large window into how I think about watches and the business surrounding them. Someone on the other side of the country can discover NEXO at two in the morning and spend as much time as he wants deciding whether this is a business—and a person—with whom he would feel comfortable dealing.

Once that confidence exists, the eventual transaction can use whatever method makes sense for that particular buyer, that particular watch and that particular situation. It might be a credit card. It might be a wire. It might happen through an established marketplace, or face-to-face at a jeweler while everyone involved drinks considerably too much coffee. The important thing is not forcing every transaction through the same pipe merely because I spent time building the pipe.

Of course, there is another aspect of credit cards that requires considerably less philosophical analysis: they cost money.

A few percentage points sounds trivial until the item being sold costs $10,000. Then the fee becomes several hundred dollars. At $20,000 it becomes something else again, and for a small business operating on margins that are nowhere near what outsiders sometimes imagine luxury-watch margins to be, that is real money.

The obvious temptation is to look at the fee purely as an expense. Avoid the card, accept a wire, protect the margin. But I am no longer convinced that is the right calculation. If accepting a card gives a first-time buyer enough confidence to transact with a business he has never dealt with before, then the fee has purchased something. It has not merely processed a payment; it has helped remove an obstacle to the transaction.

Once again, I find myself back at uncertainty.

I began the previous chapter thinking about paperwork and ended up thinking about institutional memory. I began this one thinking about credit-card processing and somehow ended up thinking about the architecture of trust. Apparently, this is becoming a pattern.

It also reinforces something I am slowly learning about building NEXO. I am becoming less interested in asking what other watch dealers do and more interested in understanding why they do it. What problem does a particular practice solve? What risk does it address? Does that risk apply equally to me? And, perhaps most importantly, am I adopting a process because it genuinely improves the business or simply because that is what a watch business is supposed to look like?

I don't yet know what the eventual answer will be for every NEXO transaction, and I am increasingly comfortable with that. Perhaps someday most transactions will happen directly through the website. Perhaps they won't. The business is young enough that pretending to know exactly what it will look like several years from now would be rather silly.

For now, flexibility seems more valuable than uniformity. I want buyers to have ways of purchasing that make them comfortable, while retaining enough protection that one bad transaction cannot undo a great many good ones. I also want the process to remain personal enough that somebody buying an expensive watch never feels as though he has simply passed through a checkout algorithm.

Because that may be the strangest thing about selling luxury watches online. The final click may be digital, the payment may travel electronically and the shipping label may be generated automatically, but the decision behind all of it remains profoundly human. Somewhere along the way, one person has decided that another person can be trusted with a significant amount of money.

The payment method is simply the machinery that allows that trust to become a transaction.

Looking Ahead

Of course, there is an obvious problem with everything I have just written. Before anyone worries about how to pay you $10,000 for a watch, you first have to convince them that you are someone to whom they might actually want to pay $10,000.

A website can display watches. Building one that inspires that kind of confidence is a rather different problem.

And that, apparently, is next.