Thank you for supporting The Generalist.
You get the sense that Zach Perret doesn't mind slipping under the radar. On a week addled by market turbulence and manipulation, as the WallStreetBets Reddit community stripped short-sellers and denuded Robinhood, the Plaid CEO issued a rather coy tweet:
"This week is on track to bring 2m+ new retail investors online."
In the hands of a more bombastic, promotional leader, a second sentence might have read: "And almost all of them will be using Plaid." As investors deserted Robinhood for Public and WeBull or bounded over to the balmy stability of Coinbase's crypto exchange, one element remained constant; they authorized access to their bank account via Perret's platform.
Less than a month after Plaid announced its decision to stay independent, rebutting the interest of Visa, the company must have had one of its best weeks ever.
But what will Plaid do with its new lease on life?
In studying the messaging from Plaid's leadership, the Justice Department's case to block Visa's acquisition on anti-trust grounds, and new product rollouts, we can begin to map the road ahead. Across geographies and technologies, Plaid and Perret look set to complete one of tech's all-time end runs.
In the piece below, we'll explore...
- Plaid's insane defensibility
- The geographies the company is targeting (and where they should go next)
- Product laddering toward an end run
- Zelle and the threat of coordination
Plaid: A metaphor
What is Plaid?
Describing it as an API that allows users to permit financial institutions to pass data to fintech applications may not mean much to some.
Let's put it in more intuitive terms:
Congratulations, you have just been conscripted into Mi6. Your first mission is to extract critical dossiers from the Unholy Hall of COBOL.
Upon arrival at the Unholy Hall, you see six doors: three on your left and three on your right. In front of each entry is a guard. You know that on the other side of each door are the dossiers you need. To complete the mission, you need to gain entrance to all six rooms, gather the information, and pass it off to your American counterpart, Finn Tek.
You walk up to the first guard.
"Please, I would like to enter the room," you say to Guard 1, ignoring your extensive stealth training but remembering your mother telling you that you don't get what you don't ask for.
"Password," Guard 1 responds brusquely.
Oh. You haven't prepared for this. Ok, well, that was a bit rude, but perhaps the next guard will be nicer. You stroll over to Guard 2 and repeat your request.
"Mot de passe," Guard 2 says, coquettishly.
You move on to Guard 3, who asks for the password in German, then to Guard 4, who, you could swear, makes his demand in Akkadian. When Guard 5 opens his mouth, a stream of numbers and shapes emerge and flutter into the darkness like bats, and Guard 6 emits a high-pitched scream.
Frustrated, you leave the Unholy Hall, knowing that Finn Tek will laugh and gloat at the Super Single Spy Mixer that evening.
"What the hell happened?" Finn says at the mixer, slapping your shoulder so hard you spill your Covertini. "I was waiting for an hour.”
"Look," you respond. "I tried! But when I got there, the guards said I needed a password. At least, I think they did. Most of them I couldn't even understand. Who speaks Akkadian, anyway?"
"Assyrian," Finn says smugly. "Didn't you bring back-up?"
You shake your head.
"Well, there's your problem. Gotta have a specialist. I know a guy. Just the man for this kind of thing. They call him El Tartán."
The name alone sends a shiver down your spine.
When you arrive the next morning at nine AM, your new partner has come. El Tartán wears a thick tweed suit and nods as you walk up. Wordlessly, the pair of you enter the Unholy Hall.
Nothing has changed from yesterday — the same doors and the same guards. El Tartán takes the lead.
"Password," Guard 1 says.
No sooner has the name left his lips than the door springs open, and Guard 1 steps aside.
Without any prelude, El Tartán turns to Guard 2. "Bouton de rose." Again, the door flies back, and the Guard clears the entrance.
In a string of sound, El Tartán turns to the remaining Guards throwing together syllables and sounds and images until the Hallway fills with a deep, energizing basso and all four doors unlatch. You and El Tartán scurry from room to room, rounding up the dossiers. Outside, Finn Tek receives your hand-off. Mission completed.
This is the value that Plaid brings to the table. At the behest of the user, the Tartán of fintech infrastructure serves as translator and code-breaker, linking out-of-date systems used by financial institutions with modern fintech applications like Robinhood, Chime, and Venmo. In making that connection, Plaid opens the door to dossiers full of useful information, including bank account details, transaction data, and more. Once unlocked that data can be used to build better financial products.
It's hard to overestimate just how powerful this capability is. Instead of a hallway with six doors, Plaid operates in a corridor of tens of thousands. Though estimates vary, some suggest there are 32,000 financial institutions globally; as the first-mover in its space, Plaid has built up a virtually unassailable lead, at least in the United States. The Justice Department's anti-trust complaint contended that Plaid connected to 11,000 US financial institutions and more than 200 million consumer bank accounts. (A brief caveat here: a number of minor institutions are powered by shared infrastructure, meaning that with one integration, Plaid can connect multiple banks.)
The extensiveness of the long tail is critical to Plaid's value proposition. If there were only 200 financial institutions in the US, or 20, insurgents would find it significantly easier to present a challenge. In comparatively little time, a new company could emulate Plaid's product and provide equivalent access. As it stands, a direct attempt would represent an exercise in futility: Plaid has devoted seven years and over $300 million to reach its current status.
Each integration, in and of itself, is often far from trivial. An illustration: banks still use COBOL, a coding language developed in 1959, to run critical functions. While Plaid probably doesn’t have to touch anything quite that antiquated, it undoubtedly does battle with some unwieldy tongues. Finding engineers fluent in such esotericisms is tricky, and doing the work to "open the door" is meaningful.
The enviability of Plaid's position is worth mentioning because of the flexibility it affords the business. From its perch, the company can extend its advantage in several directions. It can continue to scale domestically, further discouraging local challengers; it can expand geographically, using its dominant position in the most valuable market to subsidize a push elsewhere; and it can ladder up its product to compete with its own clients.
Plaid has already chosen its next frontier: Europe. In 2019, the company established a London base before extending to France, Ireland, Spain, and the Netherlands. Like so many others after a break-up, Plaid used the dumping of Visa as an opportunity to confirm its desire to spend more time abroad. Keith Grose, head of international, noted the company plans to "aggressively scale" on the continent, with staff set to double.
Grose explained the decision by highlighting the high demand for Plaid in Europe. That may be true, but the choice to focus on Europe may represent less of an offensive and more of a defensive maneuver.
In a recent podcast, Plaid's head of engineering, Jean-Denis Greze, outlines the differences between the US and European ecosystem:
In Europe, you have something called 'open banking,' which I think was implemented to a variety [sic] of degrees of goodness by different banks and different players. But it effectively means that if your core bank is 'A' and you want to use an app that relies on some of the data that that bank has, there's a mechanism for you to share data from one bank to another. In America, that doesn't exist.
That Plaid emerged in the US is no coincidence. While America has left the free-market to solve the problem of arcane bank infrastructure, Europe has placed the burden on regulators. Rules like the "Payment Services Directive," known as PSD2, have simplified the process of gathering financial information. To return to our metaphor, Europe's open banking gives you access to the information you need by opening just one or two doors rather than six. And instead of needing to speak any number of different languages, you can use English.
By reducing the pain of integration, open banking has made it considerably simpler to provide a Plaid-like authorization system, but less time and capital invested. That's resulted in the emergence of reasonably well-funded competitors such as Tink, Truelayer, and Bud.
From one perspective, Plaid's European invasion subsequently makes little sense. Why invest in a competitive market with weak defensibility?
As a defensive move, it adds up. The only way a competitor could challenge in the US market is if they had sufficient funds to devote to catching up. With its established fintech ecosystem and comparatively affluent consumer base, Europe could provide the platform to sustain such an assault. Furthermore, the continent houses several large financial institutions that Plaid will want to keep from competitors. Already, BNP Paribus has chosen to partner with Tink to bring "multi-bank features'' to its app. By moving aggressively now, Plaid can lock-up partnerships and neutralize growth.
Plaid will want to focus on regions with an emerging fintech ecosystem, a long tail of financial institutions, and either weak regulation or a strong preference for a free-market approach in planning its steps beyond Europe.
The latter issue complicates an Asian expedition. China is hostile to US tech. India is rolling out a national payments system. Hong Kong is embracing an open banking approach titled the "Commercial Data Exchange." Singapore's Monetary Authority already operates several similar tech initiatives. Entering Asia via acquisitions may make most sense, given the importance of local regulatory know-how. At present, Brankas and Brick seem the most viable targets.
Latin America, and Brazil, in particular, may represent the most attractive opportunity. As of 2018, the country has the 14th most banks globally, little regulatory momentum (unlike Mexico), and high-consumer willingness. A recent guide reported that 73% of Brazilian consumers were willing to bank exclusively online. Intriguingly, in October of last year, Visa announced it was acquiring Yellowpepper, a competitor with offices in Brazil, Mexico, Colombia, Peru, Ecuador, and Bolivia. It may not be long before Plaid follows suit.
The end run
On Thursday, Plaid unveiled a new product: direct deposit switching. Though only in beta, the new service ("Deposit Switch") allows fintechs to provide a new feature to users: the ability to change the account that receives their income quickly. Traditionally, consumers have had to file paperwork to make such a switch. Often, the transition interrupts service and leaves users unsure of where and when, exactly, their next paycheck will arrive.
It's a relatively small addition to the platform but one that's likely to please users. More than that, it's a potential indication that Plaid is laddering towards its end run.
In November's complaint, the Justice Department outlined how Visa's purchase of Plaid would harm consumers. The argument focused on how Plaid might leverage its unique position to threaten the acquirer's debit monopoly, relying on the statements of Visa leadership. In a now-famous statement, one Visa VP described Plaid as an island "volcano" with only "the tip showing above the water." The inference was that Plaid's potential far outstripped its current capabilities. The executive concluded that Plaid possessed a "massive opportunity — one that threatens Visa."
The complaint alleges that Plaid could leverage its broad distribution to provide "pay by bank" capabilities, a process that would challenge traditional online debit payments. This system, not uncommon in other countries, allows both consumers and merchants to skip a debit network like Visa.
By going direct, the transaction cost is greatly diminished, with the Justice Department estimating as much as a 95% reduction. These savings impact merchants directly, who, in turn, pass them down to the customer. With its ample bank connections, few may be as well-equipped as Plaid to facilitate this direct relationship. If the company were to succeed in this maneuver, the prize would make Plaid's initial acquisition price of $5.3 billion looks like a rounding error.
Visa's rebuttal raised several valid points, chief among them the fact that, to date, Plaid has not engaged in payment processing. Moreover, Plaid does not have an obvious relationship with consumers or merchants. These are justified clarifications, but there's a reason that same Visa executive warned, "I don't want to be IBM to their Microsoft." Returning to our earlier discussion — few companies have as critical an integration and firm a position as Plaid.
This framework clarifies the path ahead for the company. Its primary tasks will be addressing the gaps Visa identified by edging closer to consumers, de-risking the challenges of a "pay by bank" solution, and securing its future by deepening integrations and growing the long tail.
In announcing the Deposit Switch, Perret noted:
"Digitizing financial services has been a huge step forward for consumers, but we're still in the early stages. Very excited to launch our first product (of a few!) focused on giving consumers more options further upstream."
I expect those forays upstream to fall into three categories:
- Risk scoring and fraud detection
- Banking-as-a-Service (BaaS)
The most recent release represents a step towards capturing payroll. The Deposit Switch page illustrates that Plaid now connects with providers like ADP, Gusto, and Justworks, alongside gig economy platforms like DoorDash and Postmates.
Plaid would directly compete with upstarts like Puzzl, Finch, Pinwheel, and Argyle by moving in this direction. It would also set the foundation for handling payments and gathering identity information — vital features of a full-stack payroll API. Moreover, Plaid can recalibrate its relationship with the consumer through initiatives like Deposit Switch — not only is the platform granting access to information, it is actively solving a user's problem.
The additional identity information gleaned may set the stage for a push into risk scoring and fraud detection. Plaid has all the necessary access to conduct sophisticated risk scoring as it stands. Adding employment data would bolster the capability, as would further integrations into credit bureaus and other third-party sources.
A move like this would be something of a double-edged sword. On one side, Plaid would be making it considerably easier to provide consumer loans. On the other, the company would encroach on the territory of its clients. Right now, lenders like Upstart and Blend use Plaid as a data source for their own loan decisioning; would they want to have part of their USP eroded by a service provider?
"Plaid Credit" would compete with businesses like Bloom and would significantly de-risk a "pay by bank" solution, allowing the company to better evaluate users' ability to pay.
Finally, Plaid needs to ensure it remains at the vanguard of the banking-as-a-service (BaaS) space. Though its core API fundamentally altered the industry, new businesses have emerged offering additional banking services. This includes startups like Unit, Moov, Treasury Prime, Modern Treasury, and others. These BaaS platforms are less concerned with simplifying the process of accessing a bank than building one. For example, Unit helps companies embed financial capabilities like account creation, card issuing, payments, and lending into existing products. (Intriguing aside: Operator Partners, the venture vehicle associated with Plaid’s founders, invested in Unit.)
As it stands, these businesses are compliments. New financial offerings will almost certainly integrate with Plaid so that customers can connect their account to other financial apps (think Venmo). That’s all to Plaid’s benefit, but there are greater rewards on offer. By building a broader BaaS offering, Plaid can serve new institutions more deeply. Rather than relying on the API for data alone, new products would lean on Plaid for essential financial services. This opens the path for greater monetization, and strengthens Plaid’s market position. By making it easier for insurgents to build financial products, Plaid accelerates the growth of the long tail. This makes Plaid even more difficult to disrupt — instead of serving 11,000 financial institutions, it might serve 50,000, with products embedded across marketplaces, social networks, SaaS apps, and publishers. When it came time to “pay by bank,” no company would have better range or deeper relationships.
Perhaps this is what Perret meant when he tweeted, “Most of building a technology company is figuring out how to scale the long tail…”
The threat of coordination
If this week demonstrated anything, it was perhaps the dysfunction of the US financial system. But if Plaid has an existential threat, it's this: cooperation.
Should they choose to do so, financial institutions could adopt voluntary open banking protocols. These would reduce Plaid's value proposition, making the US environment more similar to that of Europe. Mounting an insurgency would become significantly more manageable, and competition would likely increase.
Still, would the creation of such protocols be enough? Could thousands of financial institutions find the motivation to voluntarily undertake extra work for which they may not be prepared or appropriately staffed?
A more credible threat might come from incumbents joining forces to mount a challenger. In 2017, a consortium of banks released a peer-to-peer payment service called Zelle. Like Venmo, Zelle allows you to send and request payments online, and in particular, via your phone. Despite being late to the party and offering an inferior product, Zelle quickly accumulated market share, overtaking Venmo and Square Cash in terms of total payments processed. In the first half of 2020, Zelle processed almost double the volume of Venmo.
The Zelle phenomenon recasts Plaid. Though it's tempting to see the company as the big fish in the pond, Zelle’s ascent hints at the great beasts lurking at lower depths. If Venmo could see its share cut so sharply by a banking JV, couldn’t Plaid?
In 2019, I expressed my belief that such a move would make sense.
It remains to be seen whether Early Warning Systems (EWS), the parent company of Zelle, has designs on the space. Beyond peer-to-peer payments, EWS offers account validation and risk scoring services, implying broader capabilities. Most significantly though, it has done the hard work of aligning incentives across America's largest banks — more than 30 are involved with Bank of America, BB&T, Capital One, JPMorgan Chase, PNC Bank, U.S. Bank and Wells Fargo sharing ownership.
EWS is Plaid’s most viable competitor. But the consortium would be foolish to underestimate the will and skill it has taken for Perret and Company to reach this point.
After news of the Visa deal's breakdown emerged, I noted that I believed Plaid was a "minimum $20B business in today's market, with a shot to become a +$100B one." Given the company has countenanced private financing valuing the company at $15 billion, and investors have waved off secondary sales at 4x the Visa price, it seems others may agree.
Plaid's success doesn't hinge on its ability to make this end run. Even without touching payments, Perret can steward the business to an outcome an order of magnitude larger than that offered by Visa. But the blueprint is there. By expanding geographically, Plaid grows its network of banks, further insinuating itself into the financial system's heart. By adding to its product, the company can more directly communicate with consumers, de-risk "pay by bank," and ensure that as new financial products proliferate, Plaid remains essential. In a dynamic, volatile market, Plaid is the quiet constant.