Pull up your last two capital requests.
One was a branch. Maybe a remodel, maybe a ground-up build. It came with renderings, a site survey, a construction timeline, and a board presentation. Nobody in the room questioned whether the institution should have branches. The debate was about scope and timing.
The other was your website. It came through as a marketing line item, it competed against the sponsorship budget and the direct mail spend, and someone probably asked whether you could get it done for less.
Now put the two numbers side by side, divide each by the number of member interactions it serves, and try to defend the result to your ALCO.
You can't. That's the point of this post.
Financial institutions evaluate every other fixed asset on utilization. Branches get measured on deposits per location, transactions per FTE, and cost per transaction. Your core system gets measured on uptime and per-member cost. The website is the one significant asset in the building that escapes the analysis entirely, because it lives in the marketing budget, and marketing budgets get evaluated on spend, not on throughput.
So let's run it.
The branch side. According to industry construction data, a new freestanding branch in 2026 runs roughly $350 to $600 per square foot. On a 3,000-square-foot prototype, that's $1.05 million to $1.8 million in hard costs before land. Bancography's branch construction surveys have tracked the same trend for two decades: branches keep getting smaller while cost per square foot keeps climbing. A significant remodel lands comfortably in seven figures.
The digital side. CU 2.0, citing year-over-year industry studies, put the average amount a financial brand invests to build a new website at $32,714, and noted there's no meaningful correlation between that number and asset size. Billion-dollar institutions underfund their websites at roughly the same rate as $200 million ones, because in both cases the site is still understood internally as a brochure rather than a growth channel. Agency practitioners who build in this space consistently say that anything under $50,000 buys templates and not much else, and that a realistic starting budget for a full redesign with content strategy is closer to $100,000 to $150,000.
Now the denominator. This is where it stops being a marketing argument and starts being a stewardship argument.
Branch transaction volume has been in structural decline for thirty years. The FMSI Teller Line Study found average monthly teller transaction volume down more than 45% over a twenty-year span. The Bureau of Labor Statistics projects teller employment to fall another 13% between 2024 and 2034. Meanwhile, McKinsey's analysis of the branch-to-mobile transition found the average mobile banking user made 38 app visits per month in 2025, against fewer than two branch visits per year.
Credit union-specific numbers tell the same story. Affinity Plus Federal Credit Union, a $5 billion cooperative in St. Paul, reported 191,530 monthly active digital banking users as of April 2026, about 65% of total membership, averaging 21.37 sessions per active user per month, with mobile driving roughly 90% of activity. Cornerstone Advisors' research found that one in five credit union members now logs into a mobile app every single day.
Take a $200 million credit union with 20,000 members and three branches. These figures are illustrative. The point is the ratio, not the precision.
| Branch | Website & digital | |
|---|---|---|
| Capital invested | $1,500,000 (remodel) | $40,000 (redesign) |
| Annual interactions served | ~60,000 visits | ~300,000+ sessions |
| Capital per annual interaction | ~$25.00 | ~$0.13 |
That's a ratio of roughly 190 to 1, invested in inverse proportion to traffic. If a branch manager proposed spending 190 times more per member visit on the least-used location in the network, the conversation would end quickly. Somehow the same allocation survives when the underused asset is made of brick and the overused one is made of code.
Run your own version of this table before your next budget cycle. You already have both numbers. Nobody has put them in the same slide.
If the website were merely underfunded, it would be a cosmetic issue. It isn't, because the site is now where account opening happens, or fails to.
Cornerstone Advisors' 2026 Digital Banking Performance Metrics report, commissioned by Alkami, found that financial institutions lost an average of 3.36 digital checking account applications for every one they completed in 2025. That translated to roughly 8,904 missed checking accounts per institution. At the 75th percentile, institutions are abandoning more prospective accountholders than they convert. Digital account opening is functioning as a filter rather than a funnel.
Signicat's research, reported by The Financial Brand, puts digital onboarding abandonment at up to 68%, climbing from 63% in 2020. Applications running more than 20 steps see abandonment above 60% on their own.
Set that against your growth picture. Filene Research Institute found only 44% of credit unions grew membership in 2024, with growth concentrated at institutions above $1 billion in assets and smaller credit unions seeing declines of 6% or more. Cornerstone's research shows deposit gathering at the top of the worry list for 69% of credit union executives. And the 2026 ACSI results showed credit union satisfaction slipping specifically on digital experience, while community banks led the segment.
The membership problem and the website problem are the same problem. You are buying traffic into a funnel that leaks two-thirds of it, and then funding the leak repair at $32,714 every five years.
Three reframes make this land with a board that thinks in assets and returns.
1. Move the website out of the marketing budget. As long as a redesign competes against media spend, it will lose to whatever produces a visible campaign this quarter. Capitalized and depreciated as a fixed asset alongside branch improvements, it competes on the terms it should: utilization, useful life, and return.
2. Fund maintenance, not just replacement. The five-year redesign cycle is a facilities mindset applied to software. You don't gut-renovate a branch and then refuse to change the signage for sixty months. Budget continuous optimization: conversion testing, funnel instrumentation, content updates, accessibility remediation.
3. Report the website like a branch. Sessions, unique visitors, applications started, applications funded, cost per funded account with direct deposit attached. Put it in the same board packet as your branch performance report, in the same format. The comparison will make the case without anyone having to argue it.
Your most-visited location has no greeter, hasn't been renovated since a management team ago, converts under a third of the people who try to open an account, and is staffed by a vendor contract nobody has read recently.
If that were a physical branch, it would be on the closure list, or on the capital plan by Friday.
Run the two numbers.Bring the table to your next planning session. The argument makes itself once the comparison is on one page.
Sources: Bancography Branch Construction Surveys; CU 2.0; Cornerstone Advisors 2026 Digital Banking Performance Metrics (commissioned by Alkami); FMSI Teller Line Study; McKinsey; Filene Research Institute; Signicat via The Financial Brand; ACSI 2026; NCUA; U.S. Bureau of Labor Statistics.