What open banking actually enables
Open banking, in practical terms, means that banks must share customer financial data via APIs when the customer consents. In the US, the Consumer Financial Protection Bureau's Section 1033 rule has formalized this, following the UK and EU's earlier Open Banking and PSD2 frameworks. The result is a set of infrastructure providers (Plaid, Teller, MX, Finicity) that aggregate bank account data and expose it cleanly. What remains underbuilt is the application layer: the products that use this data to solve specific financial problems for consumers and businesses. Building on top of existing infrastructure rather than trying to become a bank is how most successful fintech companies of the next decade will be built.
Account aggregation for specific verticals
A personal finance app using Plaid is a solved problem at the horizontal level. The unsolved version is vertical-specific: a financial dashboard for freelancers that understands self-employment income patterns, a cash flow tool for restaurant owners that integrates POS data with bank data, or an expense categorisation product for real estate investors that knows what a cap ex versus an operating expense looks like. Vertical fintech products built on open banking data command $30 to $150 per month and have much higher retention than horizontal tools because switching means losing the context the product has built up about your specific financial situation.
Embedded lending for SaaS products
Every SaaS company that processes payments for its users or sees their revenue data has an opportunity to offer embedded financing: revenue-based advances, invoice financing, or equipment loans funded by a bank partner. Shopify Capital and Stripe Capital proved the model. The software problem is the underwriting layer: building the models that assess creditworthiness from operational data (transaction volume, churn rate, inventory turns) rather than credit scores. A white-label embedded lending API that SaaS companies can integrate in two weeks, with you handling the bank partnership and underwriting, charges a spread on loan originations and scales with your clients' transaction volume.
Banking-as-a-service for niche communities
The baseline BaaS stack (Unit, Synctera, Treasury Prime) makes it possible to launch a neobank in weeks. The gap is in the communities that the major banks do not serve well: immigrant communities who need remittance-integrated banking, gig workers who need income smoothing tools, or small business owners in specific verticals who want banking integrated with their industry software. A neobank built for restaurant owners, for example, could combine the business checking account with the POS integration and the food cost analytics in one place. The bank partnership is the hard part; once you have it, the software layer is relatively straightforward.
Cash flow forecasting for SMBs
Most small businesses run out of cash not because they are unprofitable but because they cannot see a cash shortfall coming three weeks in advance. A cash flow forecasting tool that connects to a business's bank accounts and accounting software, models the next 90 days based on recurring payment patterns, flags upcoming shortfalls, and suggests timing adjustments (pay this invoice later, send this invoice sooner) is worth $50 to $200 per month to any business owner who has ever missed payroll or maxed out a credit line unexpectedly.
KYC and AML automation for fintech builders
Every fintech startup that handles money must implement Know Your Customer and Anti-Money Laundering checks. The incumbents (Jumio, Socure, Sardine) are well funded and well integrated at the enterprise level. The gap is at the startup level: a simple, well-documented KYC and fraud API with transparent per-check pricing, a developer-friendly sandbox, and a compliance ops dashboard that surfaces flagged accounts without requiring a dedicated compliance team. At $0.50 to $2.00 per verification with a $500/month platform fee, you need 200 active fintech clients to reach $1M ARR.
What to build first
Pick one vertical and one data problem. The easiest entry point is the cash flow forecasting tool for a specific business type: restaurants, law firms, or e-commerce sellers all have predictable cash flow pain and are willing to pay for visibility. Connect to Plaid, build the 90-day forecast model, and charge $79 per month. Once you have 50 paying customers, you will know what the next feature should be based on what they keep asking for. Use the LTV Calculator to model retention assumptions, and read SaaS pricing models explained before you set your price.
The competitive landscape
Plaid owns the data layer. Unit and Synctera own the BaaS layer. The application layer is wide open. The best opportunities are narrow verticals where you can build a feedback loop between the financial data and the operational context, and where the buyer is already spending money on a workaround. Read TAM, SAM, and SOM explained to scope the market before you commit.