Community banks are heading into budget season with more money for growth than they've had in years. Nearly half of bank executives expect technology budgets to rise by 40% or more this year, and 73% of community bank leaders are prioritizing organic growth in their local markets, with core deposit growth named as the greatest threat to those prospects. The spending is real, and so is the intent behind it.

However, the results are not keeping pace. Curinos puts the cost of acquiring a checking customer at $559, double the 2018 figure, and reports that in 2025 only one in ten financial institutions grew both customers and deposits by 3%, while half saw no growth in one or both. It doesn’t make sense. More budget, more channels, more offers, but for most banks no more growth.

The reason sits between the applicant and the account. A bank pays to bring a prospect to its door and the digital funnel then completes 22.9% of the applications it starts. Every dollar spent before that point is spent on behalf of whichever institution converts the other 77%.

For years the industry assumed that lost applicants ended up at a national bank or a neobank. Bank Director’s 2026 Technology Survey says otherwise. Asked to name their greatest competitive threats, 58% of bank executives and directors cited fintechs, such as Block and PayPal, ahead of the biggest banks at 48%.

That changes what a leaking funnel costs an institution. The applicant who abandons a checking application doesn’t disappear. That person now has myriad options, including crypto and stablecoin platforms, and earning the deposits is more chess than the checkers game it was in the past. When your investment may be supporting your objectives but is definitely also supporting others' goals, it's not as simple a problem as a leaking funnel.

Cornerstone Advisors and Alkami’s 2026 Digital Banking Performance Metrics found 3.36 applications abandoned for every account opened online, a ratio that has not improved as digital volume has grown. Set that next to the $559 acquisition figure, and it's clear that a bank is buying more than four applicants to open a single account before retention takes its share.

And retention takes a large share. Curinos finds that roughly half of digitally opened accounts never fund in the first month, against 5% in the branch, and that after twelve months just 41% of digitally acquired relationships remain on the books, compared with three in four opened at a branch. Digital reached 40% of new relationships in 2025 and will pass 55% by 2027, so the thinner channel is becoming the main one. The report also puts the marginal cost of the next customer, beyond the ones who arrive without marketing spend, above $1,400. At that price, a bank that keeps four relationships in ten is paying roughly $3,400 for each one that survives its first year.

None of that appears in a cost-per-acquisition report, which is why it rarely reaches a board.

The Bank Director survey explains that a full 76% of bank executives and directors see technology primarily as an engine for efficiency, and 24% see it as a means of generating growth. Under that view, account opening belongs to operations, marketing owns the top of the funnel, and the conversion rate between them belongs to nobody. Abandonment basically stays invisible until somebody instruments it.

The friction adds up where risk enters. When fraud tools can't separate a real applicant from a synthetic one in real time, the fallback is manual review, which is slow enough to lose good applicants. And it's porous enough to admit bad ones. Automation is less common than the budgets suggest, since only 12.5% of community banks in the CSBS survey reported automated loan underwriting of any kind. For business customers the funnel often does not exist at all: 17% of institutions in the Cornerstone benchmark offer digital business account opening and 20% can originate a business loan online, which means the most valuable relationships a community bank can win are still being asked to visit a branch during business hours.

Existing customers receive the quietest version of the same treatment. A customer opening a second product is typically asked to re-key data the bank already holds and re-verify an identity it verified years ago, processed as though they had walked in off the street. The bank has already paid the $559 for that customer. Losing them at the second product wastes it twice.

Three metrics demonstrate where the subsidy is flowing, and they're crucial for helping set a course to increased retention.

First, banks must measure applications started against applications completed, by product and by channel, then measure again for existing customers against new ones. The gap between those two populations is the existing-customer leak, quantified.

Second, institutions should check the number of accounts opened against accounts funded within 30 days, reported by channel so the branch number cannot conceal the digital one. Month-one funding is the earliest retention signal a bank has.

Finally, banks should look at the quantity of products per household at 12 months versus the book-wide average. If the 12-month number sits well below the average, cross-sell is stopping after year one and nobody is watching it stop.

A community bank that closes its own leak stops subsidizing the field of choice. During budget season, when strategic initiatives for filling the funnel are prioritized, use these three measures to assess where the funnel may be a sieve and how to close the gaps for organic growth.