In a blog post titled Full Stack Startups, Chris Dixon wrote:
“Suppose you develop a new technology that is valuable to some industry. The old approach was to sell or license your technology to existing companies. The new approach is to build a complete, end-to-end product or service that bypasses existing companies. Prominent examples of this full stack approach include Uber, Tesla, and Netflix. Most of these companies had partial stack antecedents that either failed or ended up being relatively small businesses. The problems with the partial stack approach include: 1) bad product experience; 2) cultural resistance to new technologies; and 3) unfavorable economics.
The full stack approach lets you bypass industry incumbents, completely control the customer experience, and capture a greater portion of the economic benefits you provide. The challenge is you need to get good at many different things. The good news is that if you can pull this off, it is very hard for competitors to replicate so many interlocking pieces.”
My take: Chris’ articulation of ‘full stack” versus “partial stack” startups helps explain why today’s fintech startups–many of whom claim to be disruptive–are anything but disruptive. Plain and simple, today’s fintech startups are partial stack companies.
The challenge that many partial stack fintech startups face isn’t one of bad product experience or cultural resistance to new technologies. In fact, these factors fuel fintech startups more than they hold them back.
Unfavorable economics is the primary challenge facing partial stack fintech firms. In a banking context, generating revenue by charging for something on top of all the other things that consumers must pay for to get their banking done, is a significant challenge. If consumers are going to go apoplectic when a bank charges them a $5 monthly fee, how are they going to stomach paying for all the point solutions offered by all the emerging partial stack fintech startups?
As Mr. Dixon put it, “Your slice of the stack might be quite valuable but without control of the end customer it’s very hard to get paid accordingly.”
In banking, the reality just might be that it’s hard to get paid accordingly even with control of the end customer.
Lessons Learned From Tesla and Uber
Tesla is a good example of a full stack startup–it owns the means of design, manufacturing, and distribution of its product. Tesla’s full stack aspirations are built around a technology that fundamentally changes the nature of the product (the car). Designing, building, distributing–and creating the ecosystem to support–an electric car is fundamentally different than designing, building, distributing, and building a support ecosystem (which includes gas stations) for gas-powered cars.
No fintech startup–in the US at least–comes anywhere close to creating a sustainable alternative banking ecosystem.
Uber, on the other hand, has been able to achieve what it has because it attacked a market that was characterized by a very large number of small, geographically-limited players. It didn’t just use technology to create a better user experience, it transformed the structure of the industry. And as it moves into other countries, it is doing the same.
The Uber opportunity simply doesn’t exist in banking.
What Tesla and Uber have done–while pursuing a full stack strategy–is achieve a degree of scope and scale (Uber in particular). The strategic choice of full stack versus partial stack for a fintech startup is similarly about scope and scale. The ability for a fintech startup to reach the level of scope and scale and Tesla and Uber have requires:
- More money than most (if not all) can get their hands on;
- More time to than most (if not all) have to achieve that level of scope and scale;
- More reformulation of a complex industry structure than most (no, make that all) have the ability, let alone foresight and vision, to pull off.
What’s Really Happening in Fintech
In a generic sense, there are four components to an industry’s stack: design, production, distribution, and support. Fintech startups aren’t creating full stack alternatives to the existing infrastructure, nor are they “unbundling” the legacy banks (which has become a popular premise). A more accurate assessment is that fintech startups are rearchitecting the industry’s hierarchy of needs:
1. Security. The most basic need is security–protocols and mechanisms to ensure that account access and transaction execution is secured and protected.
2. Movement. The next level of need is money movement–the ability to move money between accounts and/or participants. This encompasses the movement of money between businesses (B2B), consumers to business (dyslexically referred to as B2C), person to person (misnamed P2P), and between businesses/consumers and government entities (G2C?). The payment networks and ACH are examples of entities that create money movement capabilities.
3. Performance. The highest level of need is where value is created, derived, and optimized. The performance level provides capabilities that impact: a) the speed of money movement; b) the cost of providing capabilities; and (perhaps most importantly) c) the return provided to the customer. Delivering capabilities that provide performance (or value) can–and almost always do–rely on security-level and movement-level capabilities.
US-based fintech startups are not, however, creating full stack solutions that would enable them to bypass industry incumbents. In Europe, it may be very different. The UK’s Mondo, and Germany’s Fidor may be considered examples of full stack bank startups.
Last year’s Innotribe competition at SIBOS provides an illustration of this. There were seven finalists, six of whom fit neatly into the needs hierarchy:
- From a security perspective, Sixscape replaces insecure username/password authentication with certificate-based authentication, while Ensygnia’s technology improves identity verification.
- From a movement perspective, Epiphyte converts currency to bitcoins, moves the coins, then cashes them out on the other end in the local tender. TransferGo enables international money transfer without actually transferring the sender’s money internationally.
- From a performance perspective, Stockspot provides robo-advice to investors, while Lending Robot helps investors looking for lending opportunities on P2P lending platforms like Lending Club.
The 7th finalist in the 2014 Innotribe competition was Standard Treasury. How does it fit in to the hierarchy of needs?
In a number of presentations I’ve done recently, I’ve said that Standard Treasury represented a new layer in the hierarchy of needs–the connectivity layer–which I placed at the bottom of the hierarchy. I was wrong. Standard Treasury is really an entry in the movement layer of the hierarchy.
What makes Standard Treasury different from an Epiphyte or TransferGo, however, is what’s being moved. While Epiphyte and TransferGo move money (and the data about it), Standard Treasury moves capabilities. By creating a set of APIs for industry participants, Standard Treasury makes it possible for a firm to more efficiently create a full stack solution.
But there are plenty of full stack banks out there already. Why do we need any more? Because a full stack fintech startup could, would, and should compile a set of capabilities across the hierarchy of needs that is open (versus proprietary) and more truly best-of-breed than the legacy full stack banks offer.
The ability for a full stack fintech startup to succeed isn’t dependent on building out new security or new performance features (like portfolio optimization, PFM, automated savings, etc.). It’s dependent on the ability to define service or value bundles for specific segments of the population, and then tying the needed capabilities together into the service or value bundle.
In other words, APIs become central to the competitive dynamics of the industry.
Here’s why: In a banking environment that offers a wide range of capabilities to do things like find deals on intended purchases, automate savings, provide advice on investment opportunities, etc., full stack providers must rapidly assess partial stack providers’ offerings, integrate them, and deploy them to their customer base.
If this process takes nine to 12 months (or, heaven forbid, longer) from a technology integration perspective, you’re dead in the water.
If this process requires significant time and resources to negotiate legal matters, revenue sharing, pricing, etc., you’re dead in the water.
API-based competition is about speed, agility, and personalization. (By the way, for all the talk about personalization in banking, nothing we have today comes close to what’s possible in an environment with a robust set of partial stack fintech providers and smart full stack banks.
This, by the way, is why digitization is so important. It has nothing to do with consumers’ channel behaviors or preferences, and everything to do with financial institutions’ ability to deliver products and services. Digital capabilities that construct personalized services could be sold and supported in a branch or over the phone. But with the digital ability to design, construct, and deploy the service, you’ve got nothing to sell.
The Future of Full Stack Banking
Regarding full stack strategies, Mr. Dixon wrote “if you can pull it off, it is very hard for competitors to replicate so many interlocking pieces.” Spot on.
The reality of the banking industry in the US, however, is that there is no startup truly pursuing a full stack strategy. As a result, dire warnings that banks will go by the wayside are totally ridiculous (the US government is way more of a threat to banks than any fintech startup is).
Consumers need a full stack solution. But up until now, the only way to deliver that full stack was with proprietary service offerings and formal relationships between firms that determined who was or wasn’t in the service stack. Revenue-share agreements will make it easier for fintech startups to participate in full stack banks’ solutions, enabling a more open banking environment.
The critical, unanswered questions are: 1) Can and will legacy banks shift their business models to compete in an API-based competitive world? and 2) Can and will some firm emerge as a truly new full stack bank?
Too many people underestimate large firms’ ability to change–in the long run. Yes, large firms are slow to make rapid changes in model and capabilities. But for the longer-term, they have an insurance policy: Money. They can buy their way into the future.
As for some firm emerging as a truly new full stack bank, there are two firms that strike me as having the potential to do so: Amazon and Paypal. I don’t know that either will move in that direction, but both have the resources, existing scope/scale, and perhaps the nerve to do so.