Cotribute, the Anaheim, California vendor that sells account opening software to credit unions and community banks, published a Harris Poll study on 17 November 2026 in which 61% of US adults said they had quit an online financial application at some point, and only 10% of those leavers tried again with the institution they left. Fintechbits checked Cotribute’s own 20 October 2026 release and found that credit unions running its business account opening on Corelation KeyStone reported that 74% of business applicants who began an application submitted it, a client figure drawn from a different population than the new poll.
The study measures application abandonment across two surveys of roughly 2,000 adults each, and it puts most of the leakage at the funding and document steps rather than the wait for a decision.
What the Cotribute application abandonment study measured
The Harris Poll fielded two online surveys for Cotribute. The first ran from 22 to 24 October 2026 among 2,001 adults, the second from 11 to 13 November 2026 among 2,006. The stated sampling precision is plus or minus 2.7 percentage points at a 95% confidence level, and wider for subgroups. Cotribute’s landing page for the report says its three headline figures come from the October wave and describe people, not a share of applications.
That caveat changes how the 61% should be read. It counts adults who have ever quit an application, not the share of applications that fail at any one bank. Among adults who have applied online, the figure rises to 78%. Gen Z (78%) and Millennials (75%) quit far more often than Gen X (58%) and Baby Boomers (39%).
The release also carries two figures that look alike. Of those who quit, 36% went on to seek the same kind of product from a different provider, and 29% did so inside 30 days. The report’s title uses the shorter window, so the 29% is the safer number to quote. Gen Z leavers were over twice as likely as Boomer leavers to go elsewhere, 41% against 19%. The full report sits behind a form asking for a name, institution and email, so subgroup sample sizes could not be checked.
Where application abandonment happens: funding, documents and branch handoffs
Funding was the top exit. Some 46% of leavers stopped when asked to put money in, 42% stopped at identity document upload, and 35% quit on being sent to a branch, a phone line or the post. Just 22% left during the wait for an answer. Some questions allowed more than one answer, so shares may not add to 100%.
Speed expectations are tight. About 58% would accept 10 minutes or under for a checking decision, 37% want an instant one, and 12% would tolerate next-day or slower. Existing customers leak too: 14% moved a new account to another institution because the switch was simpler than dealing with their own. One in four (25%) people who opened a business account began online but had to go to a branch, phone or send paperwork to complete it.
Identity friction is partly mandatory. Banks must run a customer identification programme under 31 CFR 1020.220, which requires procedures that let the bank form a reasonable belief that it knows each customer’s true identity. The question for lenders is how that check is delivered, not whether it happens. Fintechbits covered one vendor attacking the problem from the fraud side, Accertify’s connector for Ping Identity’s DaVinci, which scores signup and login events inside the same flow.
Fintechbits analysis
Why application abandonment is a funnel design problem, not a speed problem
The useful finding is where people quit. Only 22% of leavers left while waiting for a decision, so faster approvals alone would not close the leak. The exits cluster at the steps that ask for money and proof: funding, document upload and branch handoffs. A bank can shorten decisioning and still lose the applicant at the next screen.
Paul Schaus of CCG Catalyst, quoted in the release, says of exits at those steps: “that is not customer impatience, that is a process design issue.” That reading suits Cotribute, whose chief executive Philip Paul says “Abandonment does not end at the application.” Both speakers have a stake: CCG Catalyst is a management consultancy for banks, credit unions and fintechs, and Cotribute sells the software.
The poll helps rivals as much as banks. A Gen Z applicant who is twice as likely to try elsewhere is the customer a neobank or a fintech with instant funding wants. Regulators are unlikely to act on a vendor poll, though the identification rule shows why document steps stay.
Treat the results as a prompt to measure, not a benchmark. Respondents recall their own behaviour, and a Zip survey of 1,015 US BNPL users over three days that Fintechbits wrote up on 3 November 2026 showed how stated intent can run ahead of what shoppers do. Cotribute’s poll has about twice that sample and a published margin, which is better, but the sponsor still sells the remedy.
Frequently asked questions
What does the Cotribute application abandonment study find? It finds that 61% of US adults have quit an online financial application. Among those who quit, 29% sought the same type of product elsewhere inside 30 days and 10% went back to the same institution.
Who ran the survey and how large was it? The Harris Poll ran two online surveys on behalf of Cotribute, with 2,001 adults in October 2026 and 2,006 in November 2026. The margin is plus or minus 2.7 percentage points at 95% confidence, and larger for subgroups.
Where do applicants leave most often? The most common exit is the request to fund the account, named by 46% of leavers. Identity document upload follows at 42%, and a referral to a branch, phone or mail at 35%.
What to Watch Next
Cotribute offers an Abandonment Benchmark that compares an institution’s application funnel, stage by stage, with the report’s findings. Watch whether it publishes funnel data from named clients beyond its 20 October 2026 business account figures, since a poll of consumers cannot show how many applications a given bank loses. Also watch whether the 23-page report opens without a form, so subgroup sizes and question wording can be checked. A repeat of the Harris fieldwork would show whether the 10% retry rate moves.



