How to Start a Fintech Startup Without VC (2026 Playbook)

Short answer: To start a fintech startup without VC, build software that sits on top of licensed partners and never holds or moves customer money yourself. Use read-only bank data from an aggregator such as Plaid or Teller, take payments through Stripe or GoCardless, charge a subscription from the first customer, and leave licences, cards and stored balances until revenue (not investors) can pay for them.

That is the whole strategy in one breath. Most of what makes fintech look like a venture-only industry comes from being the regulated entity: licences, capital requirements, compliance teams and bank relationships. You can skip almost all of it if you choose the right shape of product. This playbook covers which shapes work, what the rules actually say, what the infrastructure costs today, real bootstrapped examples, a worked unit-economics example, a stage plan and a 12-month timeline. Every price, rule and quote was checked on the source's own page on 11 Nov 2026. The advice in between is ours and is marked as such.

TL;DR#

• The expensive part is holding or moving money. US money transmitters are licensed by states "through the Nationwide Multistate Licensing System (NMLS)" (CSBS) and must also register with FinCEN (FinCEN). A UK e-money institution needs initial capital of "at least 350,000 euro" (legislation.gov.uk).
• Pick a regulation-light shape. Budgeting apps on read-only data, invoicing, reconciliation, spend analytics, accountant tools and embedded payments through Stripe Connect all run on partners' licences.
• Infrastructure is cheap to start. Teller's developer tier includes "100 live connections" free (Teller), Plaid's Sandbox is "free and fully-featured" (Plaid docs), and Stripe charges 2.9% + $0.30 per standard US card payment (Stripe support).
• Bootstrapped fintech exists. Lunch Money calls itself "proudly bootstrapped and 100% customer-funded" (Lunch Money); Tiller says it is "funded exclusively by customer subscriptions" (Tiller).
• Subscription first, payment revenue later. Interchange, payment markups and fees on funds held are real levers (Stripe guide), but they need volume. A subscription works from customer one.
• Trust is the product. Read-only access, 2FA, clear data deletion and a short list of vendors matter more in finance than in almost any other SaaS category.

Disclosure: ZeroRaised publishes this guide and runs a launch site for bootstrapped products. ZeroRaised has no commercial relationship with Stripe, Plaid, Teller, GoCardless or any other company named here, and no links in this post are affiliate links. This is general information, not legal, tax or financial advice. Financial regulation changes often and depends on your exact product and location, so speak to a qualified fintech lawyer before you launch anything that touches money.

Why fintech looks like a VC-only game (and when it is not)#

Ask founders why fintech needs venture capital and you will hear three reasons: licences take years, regulators require capital, and banks will not talk to tiny companies. All three are true, but they apply to a specific kind of fintech: the kind that becomes the money business itself.

There are roughly two families of fintech:

1. Money businesses. You hold customer balances, move funds between people, issue cards, lend, or offer an account. Your revenue often comes from the money itself: interchange, interest, float, transfer fees. You are, or you partner closely with, a regulated institution.
2. Software businesses that happen to be about money. You read data, organise it, automate a workflow, or help a business get paid. Money moves through someone else's licensed rails. Your revenue is a subscription, maybe with a small payments margin later.

The first family is capital-heavy by design. Regulators want a buffer between your mistakes and your customers' money, and that buffer is cash you cannot spend on product. The second family is a normal software business with extra care around security and trust. It is the family most bootstrapped fintech companies sit in, and it is the one this playbook is about.

Our read: the trap is drifting from family two into family one without noticing. "We just hold the money for a day before paying the contractor" is a money transmission question. "Users can keep a balance in the app" is an e-money question. Decide early which side of the line you are on and design the product so it stays there.

What the regulations actually say#

You do not need to become an expert in financial regulation to build a bootstrapped fintech product. You do need to know enough to recognise when a feature idea would put you into a regulated activity, so you can route it through a partner instead. Here is the short version for the US, UK and EU, from the regulators' own pages.

United States: state money transmitter licences and FinCEN#

In the US, moving money for others is regulated at two levels.

State licensing. The Conference of State Bank Supervisors (CSBS) explains that "State regulators license money transmitters through the Nationwide Multistate Licensing System (NMLS)" (CSBS). Licensing is state by state, which is why a national money transmission business can mean dozens of separate applications. CSBS has been pushing states toward a common model law, the Money Transmission Modernization Act, and says "28 states have adopted the MTMA", with some others adopting parts of it. The same page explains the model law includes "requirements for tangible net worth (similar to bank capital)" and "surety bonds (similar to insurance)". In other words: money you have to set aside or guarantee before you transmit a dollar.

Federal registration. On top of state licences, FinCEN says that "With few exceptions, each money services business (MSB) must register with the Department of the Treasury", that registration must be "filed within 180 days after the date on which the MSB is established", and that "Registration must be renewed every two years" (FinCEN MSB registration). There is no small-business exemption for money transmitters: FinCEN states that "No activity threshold applies to the definition of money transmitter" (FinCEN MSB definition). An MSB also takes on anti-money-laundering obligations: a compliance programme, reporting and record keeping.

What this means for you: if your product never takes possession of customer funds, these regimes are generally your partners' problem, not yours. Stripe, a bank, or a licensed payments company holds and moves the money; you send instructions through their API. That is the whole reason the software-on-partners model works. Whether a particular flow counts as money transmission is a legal question with real edge cases (marketplaces, escrow-like features, "pay on behalf of" flows), so get a written opinion before you build one of those.

United Kingdom: payment institutions, e-money and the registered AISP route#

The FCA has separate routes depending on what you do.

• Small payment institution (SPI). To qualify you must not "Have average monthly payment transactions in the preceding 12 months that exceed €3m". New firms use projected figures, which "mustn't exceed €3m" (FCA payment institutions).
• Authorised payment institution (API). The fuller licence, with more requirements. Firms providing account information or payment initiation services must "Hold adequate indemnity cover" (same FCA page).
• E-money institution (EMI). If customers keep a stored balance with you, you are in e-money territory. The FCA page asks applicants to "Hold adequate initial capital" (FCA EMI), and the Electronic Money Regulations set the number: "An applicant for authorisation as an electronic money institution must hold an amount of initial capital of at least 350,000 euro" (legislation.gov.uk). A small EMI route exists, capped at average outstanding e-money of €5m and average monthly payment transactions of €3m (FCA EMI).
• Registered account information service provider (RAISP). This is the lightest route, for firms where "the only payment service your firm can provide is account information services (AIS)". You still need to "hold relevant professional indemnity insurance or a comparable guarantee" (FCA RAISP).

What this means for you: a UK budgeting or cash-flow app that reads bank data is doing account information services. You can either register as an AISP yourself or, much more commonly at the start, use a regulated aggregator and operate under its permissions or as its agent, depending on the provider's model. Our read: start with an aggregator and only consider your own registration once you have revenue and a clear reason (cost, control, or a feature the aggregator will not support).

European Union: PSD2 in one paragraph#

PSD2 is the EU directive that created the open banking rules much of Europe and the UK still build on. The European Commission's summary is that "The new services that PSD2 regulates are payment initiation services (PIS) and account information services (AIS)" and that "'Strong customer authentication' will be the new security standard" (European Commission). The SCA requirement came into force on 13 September 2019 (European Commission, payment services). For a bootstrapper this cuts both ways: it is why European bank data is accessible through standard APIs at all, and it is why reading that data is itself a regulated service that you will usually access through a licensed provider.

Why the Synapse collapse matters to small founders#

Banking-as-a-service (BaaS) chains look simple from the outside: an app on top of a middleware company on top of a bank. The CFPB's record of Synapse Financial Technologies shows what happens when the middle breaks. According to the CFPB, "On April 22, 2024, Synapse filed for chapter 11 bankruptcy protection", and there was "a shortfall of between $60 and $90 million". The Bureau's 2025 action ended in a stipulated judgment with "a $1 civil money penalty", which gave the Bureau access to its civil penalty fund to compensate harmed consumers (CFPB).

Our read: the lesson for a bootstrapper is not "never use BaaS". It is that the more of your customers' money sits in a chain you do not control, the more your company's survival depends on other people's ledgers. A read-only or software-only product cannot lose anyone's deposits. That is a real advantage when you are small and cannot afford an incident.

The bootstrappable fintech shapes#

Here are the product shapes that a solo founder or small team can realistically build without becoming a regulated money business. All of them rely on partners for anything that touches funds.

The pattern: the first seven rows are software businesses. The last row is a money business. You can move toward it once you have revenue, customers who trust you, and enough margin to pay for compliance, but you do not have to start there.

Our read on where to start: B2B finance tooling is usually the easier bootstrapped entry. Businesses pay more per account, they need fewer bank connections per paying customer than a consumer budgeting app, and the buyer (a finance lead, a bookkeeper, an agency owner) can tell you exactly what they would pay to save three hours at month end. Consumer personal finance works too, and several of the examples below prove it, but it is a crowded market where retention depends on habit and price sensitivity is higher.

If you are building payments into a vertical product, our vertical SaaS without VC playbook covers why embedded payments are often the second revenue line in that model.

Infrastructure: what it costs today#

This is the stack a bootstrapped fintech typically needs, with prices from each vendor's own pages as checked on 11 Nov 2026. Several vendors publish no pricing at all; we say so rather than guess.

Payments: Stripe#

Stripe is the default for card and bank payments in a software-on-partners fintech.

• Cards: "Standard card payments are 2.9% + 0.30 USD in the US" (Stripe support).
• ACH Direct Debit: "ACH payments on Stripe cost 0.80%, capped at $5" (Stripe). Stripe's own example: a $100 payment costs $0.80 and anything above $625 costs $5.
• Connect (for platforms): if you let Stripe handle pricing for your users, "there are no fees for your platform". If you handle pricing yourself, Connect fees are "$2 per monthly active account" plus "0.25% + 25¢" per payout (Stripe support).
• Identity verification: Stripe Identity's page, served to us in the UK, shows £1.25 per verification and says "Explore Stripe Identity with your first 50 verifications free" (Stripe Identity). We could not load the US price from this box, so check your local page.
• Financial Connections (Stripe's bank data product): on the UK pricing page we saw instant bank account verification at £1.25 per successful call, balances at £0.10 per call and transactions at £0.25 (Stripe pricing, UK). US prices did not load for us.
• Issuing: "Virtual cards are 0.10 USD each in the US, 0.10 GBP in the UK and 0,10 EUR in the EU" (Stripe docs). Commercial issuing is available in the US, Canada, the UK and supported EEA countries; consumer issuing is US only (Stripe Issuing docs).
• Treasury / financial accounts: Stripe lists the US and UK as "Public preview", with "no monthly fees or minimum balance requirements", and notes "Stripe is not a bank" (Stripe Treasury docs).

If you are only selling your own software (rather than moving money for your customers), a merchant of record can take tax and compliance off your plate. Our guide to merchant of record vs payment processor explains the trade-off. For fintech features where your customers get paid through your product, you will usually need a processor with a platform product such as Connect.

Bank data aggregators#

Our read: in the US, Teller is the easiest aggregator for a bootstrapper to model because the prices are on the homepage. Plaid has wider coverage and is what several of the examples below use, but you will need to talk to them before you know your unit cost at scale. In the UK and EU, the market is mostly "contact sales", so build your prototype on one provider's sandbox and request pricing before you commit to a business model that depends on it.

Payments by bank in the UK: GoCardless#

For UK recurring payments by Direct Debit, GoCardless publishes per-transaction prices: Standard is "1% + 20p", "Capped at £4"; Advanced is 1.25% + 20p capped at £5; Pro is 1.4% + 20p capped at £5.60, all ex VAT (GoCardless pricing). International payments on Standard are 2% + 20p. For an invoicing tool aimed at UK businesses with larger invoices, the cap matters: on a £2,000 invoice, Standard costs £4 rather than £20.

Banking-as-a-service and accounts#

If you eventually need accounts or cards for your customers, the US BaaS layer includes Unit ("Unit is a financial technology company and not a bank"), Treasury Prime ("the largest bank network for embedded finance") and Column ("The only nationally chartered bank designed to help you build and fund new financial products"). None of the three publishes pricing (Unit, Treasury Prime, Column). Wise Platform offers its payment infrastructure to "banks, financial institutions and enterprises", also without public pricing (Wise Platform).

Our read on bank partnerships: treat this layer as a stage-three decision. Partners in this space tend to be selective about which programmes they take on, onboarding and compliance reviews take time, and you will be asked who your customers are, how you will monitor them and who is responsible for what. Arriving with paying customers, a clean product and a clear compliance owner puts you in a much better position than arriving with a pitch deck. Until then, Stripe's preview-stage products and plain old "send customers to their own bank" are fine.

Your own business bank account#

Separate from all of the above, you need an ordinary business account for your own company's money. Our comparison of the best business bank accounts for bootstrapped startups covers that. Keep it boring.

Compliance basics you cannot skip#

Even a software-only fintech has more security and compliance work than a typical SaaS. Here is the minimum, in order.

PCI DSS: stay in SAQ A#

If you take card payments, keep card data off your servers entirely. The PCI Security Standards Council's SAQ A covers merchants "with account data functions completely outsourced to PCI DSS validated and compliant third parties", and for e-commerce requires that "All elements of the payment page(s)/form(s) delivered to the customer's browser originate only and directly from a PCI DSS compliant TPSP/payment processor" (PCI SSC SAQ A). In practice that means Stripe Checkout, Payment Links or hosted payment elements, not a card form you build yourself. Under PCI DSS v4.0.1, SAQ A merchants must also "confirm their site is not susceptible to attacks from scripts that could affect the merchant's e-commerce system(s)" (PCI SSC blog). Keep third-party scripts off your payment pages.

SOC 2: when a B2B buyer asks#

Business customers in finance will ask for a SOC 2 report sooner than in most categories. You do not need one on day one, but you should build as if you will: access control, logging, encryption, vendor list, written policies. Our guide to SOC 2 for bootstrapped startups covers the cheapest route, so we will not repeat it here.

GDPR and data protection#

Financial transaction data reveals a lot about a person: where they live, what they buy, sometimes their health or religion. Under GDPR that raises the stakes on minimisation, retention and access requests. Our GDPR checklist for solo SaaS founders is the starting point; for fintech, add a clear retention period for transaction data and a one-click way to disconnect accounts and delete everything.

KYC and AML#

If you only read data or sell software, you will usually not run your own KYC programme; your payment partner verifies the people it pays out to. If you build anything where you onboard people who will move money through your product, you will be asked about identity checks. Stripe Identity (above) is the cheapest identity check we could price publicly today. Other vendors in this space did not show prices we could verify, so we have left them out.

Insurance#

The FCA's AIS routes require professional indemnity cover (see the regulation section above), and even outside the UK, a finance product should carry cyber and professional liability (errors and omissions) insurance before it has meaningful customer data. Our read: get quotes in your first quarter of revenue. We have not quoted prices because they vary with your data, revenue and location, and we could not verify a public rate.

Trust and security are the product#

In most SaaS categories, security is a cost centre. In fintech it is a feature people choose you for. Things that work well for small teams (our read):

• Read-only by default. Ask only for the data scopes you need. If your product never initiates payments, say so plainly on the connect screen and the homepage.
• 2FA on every account, from launch. Offer authenticator apps or passkeys, not only SMS.
• Show the data path. Name your aggregator and payment processor on a security page. Users trust a named, regulated provider more than "bank-grade security".
• Easy exit. A visible "disconnect bank" button and "delete my data" flow, with a stated timeframe.
• Fewer vendors. Every sub-processor that touches financial data is one more thing to secure and list. Resist adding analytics and chat widgets inside logged-in finance screens.
• Incident plan on paper. Who gets told, in what order, within how long. Write it before you need it.

Actual Budget is a good illustration of trust as positioning. Its homepage describes it as "a local app, plain and simple" (Actual Budget), which turns privacy into the headline feature.

Real bootstrapped fintech companies#

These are companies that, in their own words or a founder's own writing, describe themselves as bootstrapped, self-funded, customer-funded or independent. We checked each on 11 Nov 2026 and left out anything we could not verify.

Lunch Money (personal finance, solo founder)#

Lunch Money describes itself as "proudly bootstrapped and 100% customer-funded", with "a deliberate focus on sustainability over outside investors or data monetization" (Lunch Money). It was founded in 2019 by Jen Yip, and the about page says "For its first five years, Jen ran Lunch Money as a one-person company." Pricing is $10 per month or $100 per year, with the annual plan marked "Set your own price!" and a 30-day trial (Lunch Money pricing). Bank data comes through Plaid, plus CSV and PDF upload (Lunch Money import).

What to learn: a solo founder ran a consumer fintech for five years by building software on top of an aggregator, charging from the start, and making "we do not sell your data" part of the pitch.

Tiller (spreadsheet finance feeds)#

Tiller says: "We're proudly funded exclusively by customer subscriptions. No shareholders, no venture capital. Just you." (Tiller). That is its position today rather than its whole history: in 2022 GeekWire reported that Tiller "has raised an undisclosed amount of funding from people whom Polson described as 'seasoned technology investors'" and had reached breakeven (GeekWire, 2022). Our read: angel-backed early, customer-funded now, never venture-scale. It was founded by Peter Polson and is based in Seattle. Pricing is $99 per year after a 30-day trial (Tiller pricing). "Tiller uses Yodlee and Plaid as data partners", and bank data flows into the customer's spreadsheet (Tiller).

What to learn: Tiller does not try to replace the user's tools; it feeds data into a spreadsheet they already use. That is a narrow, defensible fintech shape with minimal UI to build.

Invoice Ninja (invoicing and payments)#

Invoice Ninja's story page calls it "100% bootstrapped (self-funded, no VC overlords!)" and "Bootstrapped, founder-run, and proud of it." Hillel Coren and Shalom Stark started discussing the idea in late 2013, and "By March 2014, Invoice Ninja was live" (Invoice Ninja). Pricing includes a free plan for up to 5 clients, Ninja Pro at $14 per month or $140 per year, and Enterprise from $18 per month (Invoice Ninja pricing). It also offers self-hosting.

What to learn: invoicing is fintech without the licence. Invoice Ninja sells the software that sends the invoice rather than becoming the payment company, and a free tier plus a self-hosted option widens the top of the funnel at almost no cost.

Actual Budget (local-first, now open source)#

James Long wrote in 2017 that he was "Working on Actual and bootstrapping it", and planned a bank sync feature that "will use a third-party service and will be available for a monthly fee" (James Long). Actual is now "100% free and open-source" under the MIT licence (GitHub), and its site says it "has built in support for bank syncing using goCardless (EU/UK) and SimpleFIN (US/Canada)" (Actual Budget).

What to learn: the bank sync plan from 2017 is exactly the software-on-partners model, priced as a paid add-on because the third-party data costs money. Also a reminder that bootstrapped fintech can end in many ways, including handing the project to a community.

YNAB and PocketSmith (long-running independents)#

YNAB was founded by Jesse Mecham (YNAB) and costs $14.99 per month or $109 per year after a 34-day free trial (YNAB pricing). YNAB's own pages do not discuss funding; the founder's interview on the thoughtbot podcast is titled "22 Years of Bootstrapping with Jesse Mecham" (thoughtbot podcast).

PocketSmith, from New Zealand, was founded in June 2008, and co-founder Jason Leong describes it as "Independent and founder-owned" (PocketSmith blog). It has a free plan, and paid plans start with Foundation at $14.95 per month (or $9.99 per month billed annually) (PocketSmith plans).

What to learn from all six: every one is a software business about money, not a money business. None of them is a bank. The paid products charge subscriptions. Several lean on a single named aggregator and say so publicly.

How to make money: the fintech revenue levers#

Stripe's guide on building a fintech company lists the revenue streams fintechs use under the heading "Diversify your revenue streams" (Stripe):

Our read: subscription is the bootstrapper's lever, because it is the only one whose revenue does not depend on transaction volume you do not yet have. Payment markups come next for B2B products that already move money through Stripe or GoCardless. Interchange, float and lending are where VC-backed fintechs make their money, and they are exactly the parts that need licences, partners and capital. Leave them for later or for never.

Price for value, not for cost. A B2B reconciliation tool that saves a bookkeeper a day each month can charge more than a consumer budgeting app, even though the data costs are similar.

Worked example: unit economics of a small fintech#

This is an illustrative example, our arithmetic using the public prices above. Your numbers will differ.

Example A: US consumer budgeting app on Teller and Stripe#

Assumptions: $8 per month subscription paid by card; each user links 2 accounts; you pull transactions through Teller and make 4 balance calls per account per month.

Two things jump out. First, balance calls cost more than transactions in this example, so cache balances and avoid refreshing on every app open. Second, the $0.30 Stripe fixed fee is a big share of a small monthly charge: an annual plan at $80 would cost 2.9% x $80 + $0.30 = $2.62 in card fees for the year, about $0.22 a month instead of $0.53. Push annual billing.

At 300 paying users, contribution is about $1,821 per month before hosting, insurance, tools and your time. That is not a salary, but it pays for the stack and proves the model. Remember Teller's developer tier covers 100 live connections at no charge, so your earliest test users cost nothing in data while you validate.

Example B: B2B invoicing add-on with Stripe Connect#

Assumptions: you sell an invoicing tool to small agencies for $39 per month, and you use Connect so each agency gets paid through your product. You choose to handle pricing yourself and charge a 0.5% application fee on invoice volume. Each agency collects $10,000 per month and receives 4 payouts.

This shows why "you handle pricing" on Connect needs care: the 0.25% payout fee takes about half of a 0.5% application fee. Stripe's alternative, letting Stripe handle pricing, has "no fees for your platform" (Stripe support), so compare both before you design your pricing. Card processing fees on the agencies' own invoices are not in this table; how those are charged depends on which Connect setup you choose. Also note that a payments margin only exists when your customers are actually collecting money through you, which is why the subscription comes first.

The stage plan: what to build when#

This is our framework. Each stage has a gate: do not move on until you pass it.

Most bootstrapped fintech companies, including several in the examples above, live happily at stage 2 or 3. Stage 4 and 5 are options, not destinations.

For stage 0, our guide on how to validate a SaaS idea before building applies almost unchanged. The fintech twist: ask people to show you the spreadsheet or workaround they use today. If there is no spreadsheet, the pain may not be real.

A 12-month plan#

This is our suggested timeline for a solo founder building a B2B or consumer finance tool on partners, working with a side income or savings.

Months 1 to 2: pick the niche and validate.

• Choose one group you can reach: bookkeepers for a specific trade, freelancers in one country, households using one budgeting method.
• Run 15 to 20 interviews. Collect their current spreadsheets and workarounds.
• Write down, in one sentence, whether your product will ever hold or move money. If yes, book an hour with a fintech lawyer now.
• Sell 3 paid pilots or pre-orders, even at a discount.

Months 3 to 4: build the read-only MVP.

• Build on an aggregator sandbox (Plaid Sandbox or Teller's developer tier in the US; request access and pricing from a UK/EU provider early, since new signups at one provider are currently closed).
• Use Stripe Checkout or Payment Links for your own billing so you stay in SAQ A.
• Ship 2FA, a security page naming your providers, and a disconnect-and-delete flow before the first live user.

Months 5 to 6: first paying customers.

• Move pilot users to paid plans. Push annual billing.
• Track a handful of numbers: weekly active accounts, data cost per user, churn reasons.
• Launch publicly. Our product launch checklist for bootstrapped makers covers the sequence.

Months 7 to 9: harden.

• Write your incident plan, vendor list, retention policy and access controls. This is the SOC 2 groundwork, even if you do not buy an audit yet.
• Get insurance quotes (cyber and professional liability).
• Renegotiate or re-check aggregator pricing now that you have real usage numbers.

Months 10 to 12: decide on the payments layer.

• If customers are asking to get paid or pay through you, add Stripe Connect or GoCardless with a legal review of the flow.
• If not, double down on the subscription: add the features your best customers keep asking for.
• Revisit pricing with a year of data. Fintech users who trust you are usually less price sensitive than you fear.

If you need some outside money along the way without giving up control, our list of ways to fund a product without venture capital covers revenue-based options, grants and pre-sales.

Common mistakes#

1. Holding money "just for a moment". Escrow-style features, wallets and "we pay the contractor for you" flows can turn a software company into a money transmitter or e-money issuer. Route them through a licensed partner or drop them.
2. Building your own card form. It moves you out of SAQ A and into much heavier PCI scope. Use hosted payment pages.
3. Ignoring per-user data costs. A consumer app with frequent balance refreshes can spend more on data than it earns. Model it with real prices before you set your price.
4. Picking an aggregator without checking signups and coverage. One major UK/EU provider has disabled new signups to its bank data product. Confirm you can actually get production access in your target countries.
5. Starting with interchange as the business model. It needs a card programme, a partner and volume. Start with a subscription.
6. Leaving security for later. In finance, the first incident can be the last. Ship 2FA, least-privilege scopes and deletion with the MVP.
7. Vague marketing claims. "Bank-grade security" and "your money is safe" invite questions you may not be able to answer. Say exactly what you do and which regulated partners you use.
8. Going too broad. "Finance for small businesses" is a VC pitch. "Month-end reconciliation for UK physiotherapy clinics on Xero" is a bootstrapped business.

Checklist before you launch#

• I can say in one sentence whether my product holds or moves customer money, and the answer is "no", or a licensed partner does it.
• A fintech lawyer has reviewed any flow where money moves.
• My aggregator and payment providers are named on a public security page.
• I request only the data scopes I need, and read-only where possible.
• Card payments use a hosted page (SAQ A); no third-party scripts on payment pages.
• 2FA is available to every user.
• Users can disconnect accounts and delete their data, with a stated timeframe.
• I have a written retention policy for transaction data and a GDPR process for access requests.
• I have modelled data cost per user with today's prices and set prices with at least 3x headroom (our rule of thumb).
• I have an incident plan and insurance quotes.
• I am charging from the first customer.

Frequently asked questions#

Can you start a fintech startup without VC funding?#

Yes, if you choose a product that does not hold or move customer money itself. Budgeting apps, invoicing tools, reconciliation and reporting add-ons, spend analytics and tools for accountants can all run on licensed partners such as Stripe, Plaid, Teller or GoCardless. Lunch Money, Tiller, Invoice Ninja and Actual Budget are examples that describe themselves as bootstrapped or customer-funded. What usually needs VC-scale money is becoming the regulated entity: a money transmitter, an e-money institution or a lender.

Do I need a licence to start a fintech app?#

It depends on what the app does with money. If you hold customer funds or transfer them, you may need state money transmitter licences in the US (issued through NMLS) and FinCEN registration as a money services business, or FCA authorisation as a payment or e-money institution in the UK. If your app only reads account data, in the UK that is an account information service, which has its own lighter registered AISP route, though most early products use a regulated aggregator instead. Pure software that never touches funds, such as invoicing on Stripe, usually avoids these regimes. This is not legal advice; check with a fintech lawyer before launch.

How much does it cost to start a fintech startup?#

A software-on-partners fintech can start for the cost of a normal SaaS. Plaid's Sandbox and Teller's developer tier (100 live connections) are free, and Stripe charges per transaction (2.9% + $0.30 per standard US card payment). The costs that grow are per-user data fees, such as Teller's $0.30 per enrollment per month for transactions, plus insurance, security work and legal review. Becoming licensed yourself is a different order of cost: a UK e-money institution alone needs initial capital of at least EUR 350,000.

What is the best fintech idea for a solo founder?#

Our read: a narrow B2B finance tool for a group you already understand, such as reconciliation for a specific accounting workflow, cash flow reporting for a type of agency, or invoicing for one trade. B2B buyers pay more per seat, need fewer bank connections per account than consumer apps, and accept read-only data. Consumer budgeting works too, as Lunch Money and YNAB show, but competition is heavier and retention depends on habit.

What happened with Synapse and why does it matter for fintech founders?#

Synapse Financial Technologies was a middleware company between fintech apps and partner banks. According to the CFPB, it filed for chapter 11 bankruptcy on 22 April 2024, and there was a shortfall of between $60 and $90 million in customer funds. The lesson for small founders is that when you rely on a banking-as-a-service chain, your customers' money is only as safe as the weakest record-keeper in that chain, so read-only and software-only products carry far less of this risk.

Launch your fintech on ZeroRaised#

ZeroRaised is a launch directory for products built without institutional VC. If you have built a finance tool on customers' money rather than investors', submit your product for free and get it in front of makers and early adopters. For more playbooks on building without investors, join the ZeroRaised newsletter.

Sources#

All sources last checked 11 Nov 2026.

• CSBS: money transmission under state supervision, last checked 11 Nov 2026
• FinCEN: MSB registration, last checked 11 Nov 2026
• FinCEN: MSB definition, last checked 11 Nov 2026
• FCA: payment institutions, last checked 11 Nov 2026
• FCA: e-money institutions, last checked 11 Nov 2026
• FCA: registered account information service providers, last checked 11 Nov 2026
• Electronic Money Regulations 2011, Schedule 2, last checked 11 Nov 2026
• European Commission: PSD2 explainer, last checked 11 Nov 2026
• European Commission: payment services, last checked 11 Nov 2026
• CFPB: Synapse Financial Technologies enforcement action, last checked 11 Nov 2026
• Stripe support: standard card payment pricing, last checked 11 Nov 2026
• Stripe: ACH payments 101, last checked 11 Nov 2026
• Stripe support: monetizing payments with Connect, last checked 11 Nov 2026
• Stripe Identity, last checked 11 Nov 2026
• Stripe pricing (UK page), last checked 11 Nov 2026
• Stripe docs: Issuing and virtual cards, last checked 11 Nov 2026
• Stripe docs: Treasury, last checked 11 Nov 2026
• Stripe guide: building a fintech company, last checked 11 Nov 2026
• PCI SSC: SAQ A for PCI DSS v4.0, last checked 11 Nov 2026
• PCI SSC blog: SAQ A updates, last checked 11 Nov 2026
• Plaid pricing and Plaid Sandbox docs, last checked 11 Nov 2026
• Teller, last checked 11 Nov 2026
• GoCardless pricing, Bank Account Data signups and Bank Account Data docs, last checked 11 Nov 2026
• TrueLayer, Yapily pricing, last checked 11 Nov 2026
• Unit, Treasury Prime, Column, Wise Platform, last checked 11 Nov 2026
• Lunch Money about, pricing and import, last checked 11 Nov 2026
• Tiller about, pricing and bank connections, last checked 11 Nov 2026
• GeekWire: Tiller personal finance startup finds growth formula in Microsoft Excel (2022), last checked 11 Nov 2026
• Invoice Ninja story and pricing, last checked 11 Nov 2026
• James Long: Why I'm building Actual, Actual Budget and Actual on GitHub, last checked 11 Nov 2026
• YNAB about, YNAB pricing and thoughtbot podcast episode 618, last checked 11 Nov 2026
• PocketSmith blog and PocketSmith plans, last checked 11 Nov 2026