.png)
CAC is rising. Paid channels are delivering less. And many of your highest-intent prospects are researching, comparing, and shortlisting financial products before they ever enter your funnel.
But the bigger shift is this: banks no longer control the consideration phase the way they used to.
Consumers are increasingly making decisions through AI search, Reddit threads, YouTube reviews, comparison sites, and peer recommendations long before they visit a bank's website. Acquisition is no longer just a media-buying problem. It's now a discoverability, trust, and friction problem.
That was the core theme of a recent MoneyLIVE webcast featuring leaders from Hancock Whitney Bank, TD Bank, Fifth Third Bank, and MoneyLion. The conversation moved past AI hype and focused on something more practical: what's actually changing in consumer acquisition, where banks are losing efficiency, and how leading institutions are adapting.
Your funnel didn't disappear. Customers just bypassed the entrance.
One speaker described it as a shift in where trust gets built: not on your website, but through what customers, third parties, and the internet are saying about you. Your homepage is no longer the top of your funnel. Your reputation is.
To frame it in a slightly different way, Zachary Burnett, General Manager of Loans at Engine by Gen says: "What the financial institutions face today is the ability to start that journey in the various touchpoints where your brand might be exposed."
Historically, banks optimized for owned-channel discovery: SEO, paid search, landing pages, and conversion flows. But AI-generated answers are increasingly becoming a new acquisition layer—one banks don't fully control.
Consumers now ask: "What's the best bank for small businesses?" "Which banks have the best digital experience?" "What are people saying about this bank?" And increasingly, the answers come from sources outside the bank itself.
AI didn't eliminate the funnel. It changed who controls it.
Intent matters more than funnel stages
Traditional acquisition funnels assume orderly progression: awareness → consideration → conversion. But banking decisions rarely work that neatly.
One panelist reframed the issue around trigger moments instead of funnel stages. Financial decisions are often tied to real-life events: a move, a new job, a first paycheck, a growing business, a major purchase. The research window is short. The decision follows quickly.
That means demographic targeting alone is becoming less effective than understanding context and intent. The question is no longer just: "Who fits our audience profile?" It's: "Who is experiencing a moment where this product becomes immediately relevant?"
If your content wouldn't answer a real question, it's not working
The shift in how consumers discover financial products isn't just a distribution problem. It's a content problem.
AI-generated answers are trained on content that is structured, specific, and genuinely useful. If your content library is built around brand messaging and product pages, you're increasingly invisible to the engine now driving early-stage research. Over 80% of searches end without a click—and the broad educational content that once served as your brand's entry point is exactly what AI now answers directly, without ever sending someone to your site.
That doesn't mean content matters less. It means the bar for what actually earns a place in the consideration set is higher.
Burnett put it this way: to show up in the answers AI is giving consumers, you have to optimize for AI discoverability—content that is snackable, well-structured, and specific enough for a model to accurately represent your brand in a response. The institutions doing this well aren't necessarily producing more content. They're producing content that works harder.
One speaker's team uses a practical filter: every piece of content should be answerable as a response to a real question a real person is actually asking. If it wouldn't make sense as a search result, it's probably serving the brand—not the customer. Campaign-style content that promotes your products fails this test. Decision-support content that helps someone navigate a genuine financial moment does not.
The volume play is over. Quality and the structure is what earns discoverability now. None of this means traditional marketing is dead. But it does mean you can't keep running campaigns and hoping the funnel self-corrects. The institutions gaining ground are finding the balance and AI is what makes that balance achievable at speed.
Most acquisition waste happens after the click
One statistic from the webcast captured the scale of the problem: 95% of customers who abandon an application never come back to complete it.
That means most acquisition waste doesn't happen at the ad level. It happens inside the experience itself.
The panel repeatedly came back to the same themes: qualification uncertainty, unnecessary friction, redundant information requests, poor expectation-setting.
And importantly, less friction doesn't always mean fewer screens. One panelist shared that their bank actually saw account openings increase after adding a simple upfront screen explaining how long the application would take, what customers needed to complete it, and what to expect next. Sometimes reducing friction means reducing uncertainty.
AI's real value: compressing the learning loop
Most banks still can't trace a conversion all the way back to its original touchpoint. Running a campaign without that closed loop is like playing golf and not keeping score—you can enjoy it, but you can't improve. The data usually exists. The challenge is connecting it across the digital and institutional silos that keep it apart.
People can enter the funnel at different places. Some come in through an AI-generated answer. Some through a comparison site. Some through a branch referral. The content, the qualification, the experience—all of it has to work across entry points. And none of it holds if you can't measure and optimize fast enough to keep up.
That's where AI's practical value becomes clearest. Before, campaign optimization took weeks or months. Now, AI-native tooling is compressing that cycle into days.
Teams are now using micro-segmentation, dynamic creative testing, AI-assisted audience matching, and rapid feedback loops to identify which audiences convert, which messaging resonates, where customers drop off, and which channels actually drive value. Burnett's team built a feature matrix with over 400 variables running tests faster than any human-led process could, pulling data back in near real time, and continuously adjusting.
The result isn't just better targeting. It's faster learning.
And the panel made another point clear: most banks don't need to build all of this internally. The institutions moving fastest are often partnering with AI-native vendors rather than trying to develop every capability in-house. The competitive advantage isn't owning the technology. It's improving the speed and quality of decision-making.
What leading banks are actually doing differently
Across the discussion, a few clear patterns emerged. The institutions adapting fastest are investing in closed-loop attribution, optimizing for AI discoverability rather than just SEO, and designing around trigger moments instead of static personas. They're moving qualification higher in the funnel, producing less content but content that actually answers the question a customer is asking, and using AI to accelerate how quickly they learn what's working.
Most importantly, they've stopped assuming the funnel starts when someone lands on the website. Because increasingly, it doesn't.
The banks gaining ground right now aren't necessarily the ones spending the most on media. They're the ones redesigning acquisition around how consumers actually make decisions—through AI-assisted research, peer validation, trigger moments, trust signals, and frictionless digital experiences. Acquisition starts long before someone clicks on your website. And optimization means far more than improving click-through rates.
The funnel didn't disappear. It just moved somewhere else.
Watch the full MoneyLIVE webcast recording here.