redesign | payments: Square’s IPO will be an interesting one to watch

Rakesh Agrawal is a payments expert. He has been widely quoted about the industry in publications such as Bloomberg Businesweek, The New York Times and appeared on CNBC and Bloomberg.

According to Bloomberg, Square has begun the process of an IPO with a confidential filing. (Allowed under the JOBS Act for companies with less than $1 billion in revenue for the previous completed fiscal year.)

Square has been widely considered a hot company. Part of this is due to the celebrity status of founder Jack Dorsey, who also co-founded Twitter. Its eponymous card reader is used by millions of small businesses, according to the company. (This should be discounted because there is no commitment; I’m technically a Square user, even though my Square sits in a drawer.)

There’s no doubt that Square has simplified and demystified the bewildering payments process for small businesses. Instead of dealing with byzantine rates, statement fees and minimums, Square charges a flat 2.75% of each transaction.

But that also creates a business problem for Square: most transactions at coffee shops and food trucks result in a loss to the company due to the relatively high fixed cost of credit and debit transactions.

Square claims $30 billion in processed transactions. If you’re generous and assume all of that volume is at the 2.75% rate, that translates in to $825 million in gross revenue. But the vast majority of that goes straight out the door to payments processors.

In the credit card industry, success is measured in “bps,” or basis points. 100 basis points equals 1 percent. Apple gets 15 bps per transaction for Apple Pay. Square should be getting significantly more. But, unlike Square, Apple likely isn’t on the hook for fraud losses.

The S-1 won’t be available to the public for some time, But when it comes out, here are the key things to look for:

  • Merchant mix. The type of merchant affects Square’s profitability. Food trucks and coffee shops are bad. Doctors, plumbers and contractors are good.
  • The median ticket. The lower the median ticket, the worse things are for Square. A number in the $10-$20 range is disastrous. Above $100 is great.
  • Credit vs. debit mix. Credit transactions are substantially more expensive to process.
  • Transactions per merchant. This will likely show that the vast majority of Square’s “merchants” are people like me with Squares sitting in drawers.
  • Fraud losses. Fraud makes or breaks a payments business. In the early years, PayPal nearly went bankrupt because of fraud.
  • Revenue from non-transaction products. Many of Square’s widely publicized initiatives are failures: Square Wallet, Square in a Box, Square gift cards, Square Cash. For Square to be a business that approaches its valuation, it will need to develop significant revenue from ancillary products.

redesign | payments: Facebook and peer-to-peer payments

Connect buttonFacebook’s upcoming move into peer-to-peer payments brings it into a crowded field. The most obvious competitors are Square Cash, PayPal’s Venmo and PayPal itself. There are also competitors on the bank side, including Popmoney and Chase QuickPay.

Clearly Facebook is in a strong position to dominate the space. But with a near universal user base in the United States, that could be said about any space Facebook enters.

Facebook Payments will leverage the Messenger product and let people send money back-and-forth while messaging each other. Instead of using ACH, the system uses debit cards. (Just like Square Cash.)

Some of the big questions are: Continue reading “redesign | payments: Facebook and peer-to-peer payments”

redesign | payments: Hiring Stan Chudnovsky shows Facebook might be getting serious about payments

Facebook has frustrated me for a long time: they have such huge reach and engagement that I’ve thought that Facebook should be entering more spaces, rather than just limiting itself to ads on the current product. Facebook has billion dollar opportunities in payments, commerce, television and local, just to name a few.

But Facebook has been laser-focused on mobile. (The last time I interviewed at Facebook, I was told that my ideas were “too big” for Facebook.)

With the hiring of Stan Chudnovsky to work on Messenger (and presumably payments within Messenger) with David Marcus, it shows that the company might finally be looking to diversify beyond the core product. (Stan worked for David at PayPal.)

In much of the developing world, an ad-supported business is unrealistic today. There are a number of reasons for this:

  • Purchasing power. If you make $1 a day, it’s hard to develop an ROI to advertise to you.
  • Lack of audience. Much of the audience still isn’t online. (Though Facebook is working heavily on that.)
  • Lack of infrastructure. Without agencies and similar infrastructure, the complex nature of advertising models is hard to pull off. (Not that a new model can’t be invented — it should — but that takes time.)

Payments is a natural fit for Facebook. People all over the world have to pay. And taking even a penny or two on a transaction could easily dwarf revenue from ads.

Facebook’s best play here is international. In the U.S., there are very well established payment systems. Credit cards are ubiquitous and card networks dominate. It’s a hard market to displace, even at Facebook’s scale. But bring payments into a greenfield opportunity and Facebook has a real chance to dominate.

In the U.S., there are social payment opportunities. Venmo, which David acquired for PayPal, has shown an emerging behavior that has huge potential when paired with the Facebook firehose.

Facebook also has a good team; David and Stan are both entrepreneurs. Working at PayPal was (almost) as odd a fit for them as it was for me. It’ll be interesting to see what they can accomplish when they’re given resources within a company that truly values innovation.

Disclosure: I briefly worked for Stan and David at PayPal, until I decided that PayPal would never be able to move at the pace that I wanted it to.

redesignAnswer: My first transaction on Amazon was for $300

Amazon local balance

In testing the new Amazon Local Register yesterday, my first transaction was for $300.

Amazon didn’t make any money on this transaction. (Nor will they make money on a lot of the transactions on Amazon Local Register at the 1.75% rate.)

There was a mistake in the original image of my transaction. (Sort of.) I needed to show the card and the original load value, so I made the card facing forward. In order for it to actually read the card, the mag stripe needs to be facing the camera.


redesignAnswer: The Square/PayPal charge was to cash out balance on PayPal debit card

The purpose of the transaction on Friday was to use up the remaining value of a PayPal debit card that I had purchased for test purposes. The PayPal debit card is a poor value compared with products like Walgreens Balance Financial and American Express’s Bluebird. T-Mobile customers also have a better value with T-Mobile’s prepaid card.

PayPal charges a monthly fee of $5. Rather than paying another fee, I decided to use up the remaining balance by charging myself with Square.

The total transaction amount was $3.75. Square did not make a profit on the transaction. (Come on, you know me! Did you really think I’d let Square make a profit?)

A number of people commented that I made a mistake by leaving my debit card number in plain view. I don’t make such mistakes. My credit card numbers in social media are like sideboob on TV: you think you’re seeing something, but you’re not. If you see a credit or debit card number, it’s a clue. (In this case, a clue that there was no money on the card.)

redesignAnswer: What’s the simplest way to improve this Amazon GC?

This is the kind of question that I love because it draws a lot of answers, many of which are valid.

The answer I was thinking of was to make it easy to scan. Supposedly the Amazon app can scan real world objects. But Amazon requires you to manually enter the claim code. You should be able to scan a bar code or QR code to automatically apply the code to your account after logging in. OCR technology is also good enough that you should be able to scan the text as printed.

Amazon has a wide range of gift card products. Some of the responses focused on emailing gift cards and doing things electronically. Amazon already does that.

But Amazon also offers gift cards at retail and through incentive programs. This particular gift card was received through a credit card rewards program. I’m not exactly sure what the magnetic stripe is for, because I can’t swipe it on my computer. My best guess is that it’s for activation.

This particular card doesn’t have the scratch off; cards sold at retail do. (There used to be gift card fraud where crooks would copy down activation codes and wait for them to be activated.)

Some of the other great answers:

  • Print the card on card stock so it can be recycled or biodegrade. Whole Foods does this with their gift cards. I’ve seen other gift cards made of a plasticky corn based product.
  • Use all alphabetic characters. The intermingling of numbers and letters makes entering the code on mobile devices especially hard because you have to toggle between keyboards.
  • Make the print of the redemption code larger so that older folks or folks with vision issues can more easily read it.

Victor: iTunes already does this, making it easy to redeem Starbucks app and song of the week codes.

redesignAnswer: The most fundamental flaw in @FICO credit scoring

Imagine two borrowers.

Borrower 1

  • Has a $20,000 credit limit on 1 card.
  • Is using $5,000 of that credit limit.
  • Makes $25,000 a year.
  • Has no assets to speak of; essentially a net worth of $0.
  • Has a mortgage and a car loan.

Borrower 2

  • Has a $7,500 credit limit on 1 card.
  • Is using $5,000 of that credit limit.
  • Makes $250,000 a year.
  • Has $300,000 in liquid assets.
  • Paid cash for car; mortgage is paid off.

Who is the better credit risk? Most humans would say borrower 2.

But the FICO scoring model would say borrower 1.

The fundamental flaw in FICO is that it doesn’t take into account the income or assets of the borrower. It focuses on things like credit utilization, recent inquiries, etc. Someone who makes $500,000 looking for a $25,000 line of credit is more likely to be able to repay that loan than someone who makes $30,000.

The FICO model has many, many flaws. Another significant one is that it often creates scenarios where doing the economically optimal thing reduces your FICO score. We’ll cover that in a later quiz.

My favorite answer to this quiz was:

they let Smedley influence credit denials to top tier CEO’s. the lies keep spreading even deep into financial institution models. i bet FICO will still let David Marcus have a First Progress secured credit card – Anuj hasn’t moved that deep!

redesignAnswer: A small change lets @penfed make a lot more money on credit cards

The question: why did PenFed send me a new Visa without asking?

photo (5)


The key to this puzzle is noticing the different words below Visa. In one case, it is “Platinum”. In another it’s “Signature.”

The interchange rates for Signature cards are substantially higher than for Platinum cards. The issuing bank can make a lot more on transactions simply by “upgrading” the customer to Signature. This is virtually no cost, aside from re-issuing the cards. There are a bunch of services that come with Signature status (like concierge), but these are low-use services of marginal value.

For people in the payments industry, this should be an easy-to-moderate puzzle. (1 person in the industry answered correctly.) For people outside of payments, this was a hard puzzle.

A few things that came up a lot in the answers:

  • Because Signature has the ability to charge above the limit people might be inclined to charge more. But the no preset spending limit thing is an attempt to copy AmEx cards. It’s also mostly marketing — you rarely get to charge much more.
  • Some others said that the lack of raised numbers is a security feature and the reason for re-issuance. It’s a marginal benefit and not something that banks would not spend money to re-issue cards en masse for.

I’m joining OLO’s advisory board

I’m pleased to announce that I’m joining the advisory board of OLO. You can see the press release here. Forbes recently had a great write up of OLO.

OLO provides online ordering for companies such as Five Guys, Noodles & Co. and Cold Stone Creamery. Guests can order from their smartphones and pick up their order in store without waiting in line.

I met founder Noah Glass earlier this year and was very impressed by his commitment to helping restaurants incorporate mobile technologies.

Mobile has changed a lot of spaces already; I think there’s a lot of potential to improve the restaurant experience for both consumers and operators:

  • A better consumer experience. We’ve all experienced the lunch rush. Your stomach is grumbling and you just want to eat. With order ahead, you can bypass the line and get straight to your food. Mobile ordering also increases order accuracy because you can see what you’re ordering. I was recently at a bar in Utah and ordered a Hop Notch IPA. The bartender thought I said half nachos. Oops.
  • Increased throughput for restaurants. People who know what they want aren’t turned off by a long line and go elsewhere.
  • Reduced operational costs for restaurants. Miscommunication in ordering and unclaimed orders cost restaurants money.

As I’ve noted repeatedly in my writings, local is a hard slog and OLO has shown the dedication and perseverance necessary to succeed in local. The company started off with mobile ordering by text message in 2005. It is also well financed, having recently received a strategic investment from PayPal as part of a $5 million round.

Noah and team are off to a great start and I look forward to working with them to help restaurants take advantage of mobile.