The fintech startup rule book is being rewritten by APIs

For the last several years, conventional wisdom for many startups has been that self-built, proprietary software is the best way to win customers. Recently, without much fanfare, conventional wisdom has moved on. The dawn of the API economy is upon us, and the most efficient way of delivering value has changed. 

For digital startups, standard operating procedure has been to build functionality and technology from scratch — so-called “DIY”. One of the main reasons for seed funding was to recruit a team of dev-ninjas and start building. 

Times change. More and more capabilities are now deliverable by open APIs – lightweight, embeddable code that delivers nearly unlimited functionality. This a radical reversal on the existing model, where DIY building imposes massive capital costs at exactly the time when you don’t have any revenue. 

Despite the trend towards an API economy, many startups are still prioritizing building from-scratch software functionality that is outside of their core focus, absorbing months or years of dev time and millions of their precious seed dollars.

This is a dangerous misallocation of resources that threatens the business case for many of these would-be unicorns. Instead, the capabilities that define your offer need to be built as quickly as possible, before someone else sweeps up the spend available in the chosen space.

Breadth is the new differentiation

The new battleground for startups is breadth of functionality. Successful players like Square started with a focused offer, promising to automate a single aspect of back office drudgery. SME business owners were impressed and signed up, and enterprise users have followed suit. 

What’s happening now, is that competitive new entrants are offering the same automation, delivered via embedded APIs, at a lower cost. And they are going further than that and offering additional automation tools – including procure-to-pay workflows (bill payments, BNPL, etc.) and order-to-cash workflows (invoicing, invoice insurance, cash advance, etc.) – to differentiate themselves. 

Embedded tools eliminate the “cost of figuring it out.” Even if a startup has $100M to spend on development, an embedded finance tool, built and refined by specialists, is going to be a better and quicker option than someone coming at it for the first time.  

The white elephant trap

Despite the ease and availability of embedded APIs today, I still meet CEOs and CTOs who believe build-from-scratch is the logical strategy. The apparent attraction is that startups gain full control over outcomes. They can design and create “unique” functionality that will stand out in increasingly crowded markets. Let’s break that down. 

It’s pretty clear that most development programs struggle to deliver the initial vision in a reasonable timeframe and within anything like the original budget. Speak to other entrepreneurs who are a few years further along the path, or who have now run out of funding, and they will largely tell you that speed to market trumps just about anything else. 

It’s only when you get to market that you find out how your value proposition really lands with prospects. Better to get there quickly with great functionality and refine the offer based on real market feedback, than to second-guess customers’ spending priorities with highly-finessed (and hugely expensive) software that might be a white elephant.  

Is “unique” really what customers want?

Point automation apps were great when they were new and did away with the alternative, manual process. No one particularly enjoyed scanning their receipts every month and filling out the relevant accounts forms, nor cross-checking their timesheets and adding client hours to another accounts form to create an invoice that would need cross-checking several times.  

But we’re past that point now. The software functionality to do those things is not inherently difficult – it just didn’t exist before Expensify, Harvest and the many other apps that now offer similar functionalities. 

Users are now wondering why they need multiple apps, logins, billing cycles, and the rest to do these things. A single provider would be much more convenient enabling them to do multiple things from a single interface —  ideally one that addresses the specific needs of their sector. But creating from scratch each one of those capabilities adds time and cost — things you don’t have if you want to get to market quickly. 

And however niche a sector appears to be, users are very unlikely to have genuinely unique needs — certainly not in numbers large enough to justify the cost of bespoke DIY development. Most will want to do exactly the same things as peers in their sector: Get paid, pay their bills, log and collect expenses, and file taxes to remain compliant. 

Building unique functionality might attract a few more customers at the margins, but not the core market needed to keep investors happy. Pushing a launch out by a year or more and burning through millions, just to attract those marginal extra users, is not a smart use of startup seed funding.  

Context is everything

If this makes it seem like fintechs are engaged in a race, not to the bottom but to reach the widest functional breadth, then there is more to it than that.

In financial services, functionality used to be siloed: Bank accounts, for example, tended to be separate from anything else. Then came the Uber wallet for its drivers. This was highly contextualised and therefore instantly more valuable to those drivers. 

Uber developed its wallet using DIY, because there was no other option at the time. Lyft then responded by DIY-ing its own offer. But that was way too slow. By the time it was ready, Uber was off in the distance. In the API economy, there is no longer any need to wait.

Every company is now a fintech?

Looked at through an API lens, every bank, vertical software company or platform is potentially a fintech. And 90 per cent of them are looking to build integrated experiences into their offer. 

The effectiveness of embedded finance tools shouldn’t be a surprise. According to research from J.P. Morgan, software companies that embed payments into their platforms see a two-five times increase in revenue per customer. Moreover, embedded finance will generate $230bn in revenue by 2025, a 10-fold increase from $22.5bn in 2020.

And yet, not all startups are on the case. It’s not as if they ring a bell somewhere to tell you the game has changed. By the time you notice, it’s probably too late and your competitor has won.

If breadth and contextualisation are the new rules of the game, getting there with API embedded tools saves a ton of money - money you can use to build something amazing.

The views and opinions expressed are not necessarily those of AltFi.