The Vault Door Opened. Nobody Walked In.

The vault door opened. Nobody walked in#

Swaystack, a fintech product built to fix what banks call account activation, this week took strategic backing from the Btech Consortium Fund, and the pitch is blunt: banks are brilliant at getting someone to open an account and mediocre at getting them to actually use it. The company's own numbers, published when it signed its first cohort of banks and credit unions, put a figure on the leak. Customer acquisition for a new checking account runs well over $400, and nearly half of the accounts that clear that bar go dormant inside twelve months. Eleven community banks and credit unions signed on in Swaystack's first year, betting that the fix sits closer to the front door than the marketing budget does.

The backer is worth a beat of its own. Btech Consortium Fund is a partnership between Elizabeth Park Capital Management, Strandview Capital and a bench of community banks holding combined assets over $180 billion, at least sixteen named institutions acting as both investors and prospective customers. A consortium of banks does not fund a vitamin. It funds the thing its own members already feel the cost of, which means the dormancy problem is not one bank's operations quirk but a shared, quantified pain across a sixteen-bank sounding board.

That is the quiet scandal sitting under most fintech customer onboarding: the money is spent correctly and still wasted. A prospect sees an ad, compares three neobanks, fills out a form, passes a Know Your Customer check, and gets a welcome email. Every one of those steps is a marketing win. None of them is a funded account. The gap between application and activation is where a genuinely good acquisition campaign quietly rots, and it rarely shows up on the dashboard that reports cost per acquisition, because that dashboard usually stops counting the moment the account opens.

Swaystack's fix is deliberately unglamorous: nudges, prompts and a funding path that meets a new customer where the drop-off actually happens, rather than a redesigned homepage or another paid channel. It is worth sitting with why a marketer should care about a product this far downstream of the media plan, because the honest answer is that the work has quietly become downstream-shaped too. A brand that only measures itself on cost per click is measuring half the funnel and reporting the healthy half. Bank account activation is not a back-office metric. It is the second half of the campaign nobody budgeted for.

Set this against the industry's real progress on the front door. The FDIC's own household survey put the unbanked rate at 4.2 percent in 2023, roughly 5.6 million households, the lowest reading since the survey began in 2009. Getting a first account into someone's hands has never been easier or cheaper to prove. What the Swaystack numbers say, in effect, is that the industry solved the hard problem and left the easy one to rot: the account exists, the paperwork cleared, and half the time nobody funds it.

Why customer onboarding is the real CAC#

Customer onboarding gets filed under product, or under operations, or under compliance, and almost never under marketing, which is exactly the mistake. The acquisition budget does not stop working when someone clicks submit on an application. It keeps spending, silently, for as long as that account sits open and unfunded, because every unfunded account is a marketing dollar with no revenue attached and a clock running against it.

The client onboarding process most banks actually run#

Ask most banks to describe their client onboarding process and you get a compliance answer: identity verification, a funding prompt, a welcome packet, done. Ask what happens to the customer who does not fund the account in week one, and the answer thins out fast. There is rarely a second nudge, rarely a reason given for why funding matters, and rarely anyone accountable for the gap between opened and active. The account exists in the ledger and nowhere else.

Cornerstone Advisors and Alkami's 2026 Digital Banking Performance Metrics study, now in its seventh year and drawing on roughly 150 banks and credit unions, found 3.36 digital account applications abandoned for every one that gets completed. That figure sits earlier in the funnel than Swaystack's dormancy number, at the application itself, which means the leak folkfox clients worry about most (the paid click that never becomes a customer) is only the first of two leaks. The second one, dormancy after opening, is bigger and less visible, because it happens after the marketing team has already logged the conversion and moved on.

Regulators have their own name for the moment a bank must be plain with a new customer about what an account is and what it costs. Under Regulation E, an electronic funds transfer carries its own disclosure and error-resolution duties, and the National Credit Union Administration supervises exactly the kind of member-owned institution that signed with Swaystack in its first cohort. None of that machinery was built to chase a customer who never funds an account. It assumes the account is live and simply governs how it behaves once it is, which leaves the funding gap sitting entirely outside anyone's compliance job description and squarely inside marketing's.

The data behind the drop-off#

Put the two numbers side by side and the shape of the problem gets sharper. Cornerstone's abandonment figure covers the application itself; Swaystack's dormancy figure covers what happens after the account is technically open. A prospect who survives both gauntlets is genuinely rare, and every one of them cost the full acquisition price to reach.

The reason best customer onboarding practices keep circling back to the same short list, a funding nudge, a reason given, a second and third prompt, a human off-ramp when the automated one stalls, is that the failure mode is nearly always the same failure mode. Nobody told the customer what to do next, or told them once and never again. That is a solvable problem, and it is solvable by people who already think in funnels, sequences and prompts, which is to say it is a marketing problem that has been sitting in the wrong department.

The regulatory backdrop makes the case sharper still. Truth in Savings disclosures, governed under the CFPB's Regulation DD, already require a bank to tell a new customer plainly what an account costs and what it pays. A funding nudge sent inside that same disclosure window is not a growth hack bolted onto compliance copy; it is the compliance copy doing a second job it was always capable of doing.

How to build a client onboarding process that does not leak#

None of this requires ripping up a core banking platform. It requires treating the week after signup with the same rigour a growth team already applies to the week before it.

Start with the moment funding actually happens#

Map exactly when a customer moves money into a new account, not when they open it. That single data point separates a live customer from a line in the ledger, and most banks do not track it as a distinct event at all.

Give the account a reason to be funded, not just a prompt#

A push notification that says fund your account performs worse than one that says your first paycheque could land here in three days. Customer onboarding process examples that work tend to name a specific, near-term benefit rather than repeating the instruction.

Hunt for the drop-off point, then camp there#

Most banks can name their overall dormancy rate but cannot say which day in the sequence loses the most accounts. That is backwards. A fox does not prowl an entire hedgerow when it already has the scent of where the quarry actually goes to ground; it follows the trail to the exact gap in the brush and waits there. The equivalent here is instrumenting each day of the first fortnight separately, finding the single day where the largest share of accounts stop moving, and putting the heaviest resource against that one day rather than spreading a generic nudge evenly across thirty.

A fox does not abandon a den because the first attempt at the entrance collapsed. It digs a second way in. A client onboarding process that only has one route to activation, and gives up quietly when that route fails, is a den with one entrance and no fallback, and every bank running one is paying full acquisition price for a customer it has already half lost.

The commercial case for fixing it first#

Every agency worth hiring for bank account activation already knows how to lower cost per click. Fewer know how to prove that the click survived contact with a login screen, and that is the pitch that actually moves a chief marketing officer's budget, because it reframes onboarding from an operations cost centre into the highest-leverage line on the media plan.

The Swaystack story is small in dollar terms, a strategic investment and eleven institutional clients, but the argument it makes is not small. Customer onboarding sits exactly where marketing spend and product experience meet, and almost nobody owns that seam. A fox that stakes out the seam between two territories usually eats better than one that only hunts the middle of its own patch, and the same is true for a growth team willing to follow the customer past the point where the acquisition dashboard stops looking.

It also reframes what a bank should ask a prospective agency before signing. Cost per click and cost per application are easy numbers to report because they arrive quickly and flatter everyone. Cost per funded, active account arrives slower and flatters almost nobody, which is exactly why it is the number worth asking for. The American Bankers Association's own retail-banking guidance treats account growth and account usage as related but separate goals for precisely this reason: a bank can hit one and miss the other for years without anyone in the marketing chain noticing.

For a folkfox client weighing where to spend the next quarter's budget, the honest answer is often not another channel. It is a hard look at the client onboarding process the last campaign fed into, and whether the accounts it won are still open.

Frequently asked questions#