April 21, 2025
The mortgage industry’s current vibe? Somewhere between “adapt or die” and “wait and see.” As mega-lenders soak up market share and deploy AI like it’s going out of style, the rest of the pack is left deciding whether to sprint, jog, or nap. So what are the options—and which one won’t lead to extinction?
Approach 1: The Cautious Strategist
Let’s call them the AI-adjacent adopters. These lenders are staying on top of tech trends (or at least trying), cautiously implementing solutions that pass the sniff test of reliability and ROI. Think methodical AI use for document validation, data integration, or intelligent disclosures—like Brimma’s Vallia suite that actually does something with AI beyond slapping it on the label.
It’s a smart move, theoretically. You pick what’s real. You pick what works. But in practice? This group is stuck in an endless loop of vendor demos and AI buzzword bingo. With the market crawling with “intelligent” tools, how does a mid-sized lender even evaluate which AI is actually intelligent? Especially when your IT team is outnumbered by your underwriters by a lot!
Still, this group embraces the reality that too much change too fast is operational poison. And they’re right: throttling transformation to match internal bandwidth is just good sense.
Approach 2: The Strategic Sleeper
Then there are the “keep the lights on” crew. Their motto? “Let the early adopters figure it out.” While it may feel like waving a white flag, it’s more of a tactical snooze. Wait 6-9 months, see which vendors survive the pilot phase, and then go all in. Is it lazy? Maybe. Is it risk-averse and possibly brilliant? Also yes.
This approach banks on the reality that most lenders aren’t software companies. You’re here to close loans, not debug integrations. Besides, in AI years, nine months is basically a decade. By the time these folks act, the dust may have settled—and someone else has already figured out what actually works.
Approach 3: The Frankenstack Builder
This group is often smaller, scrappier, and has a fondness for duct tape. They try to cobble together best-in-breed tools—AUS from one place, CRM from another, maybe Brimma’s Data Connect to glue it all together. It’s a bold strategy that can work… if you’ve got a tech-savvy team and a strong caffeine supply.
The risk? Integration nightmares and Frankenstein workflows. The reward? A tech stack that’s potentially better than what even the big players are running. Just don’t expect a smooth ride.
Approach 4: The AI Romantic
Ah, the dreamers. These lenders buy into the hype wholesale. “Let’s AI all the things!” they say, while ignoring that a good chunk of AI tools haven’t even hit beta. They sign contracts, deploy bots, and pray for miracles.
Spoiler: this often ends with expensive shelfware and a compliance officer breathing into a paper bag. Unless you’re partnering with a provider who understands the mortgage lifecycle (looking at you, Vallia), you might be investing in sizzle without steak.
AI Smackdown: The Great Lender Debate
Picture it: Four lender personas seated around a sleek, glass conference table. Espresso shots are flowing. The whiteboard is already half-full of acronyms. And the gloves are off.
Cautious Strategist (leans forward, arms crossed):
“Let’s start with the obvious: change without clarity is chaos. I’m not here to gamble on AI solutions that can’t explain their logic or guarantee results. We roll out what’s proven—DocFlow, AUS Sandbox, AI CRM. One step at a time.”
Frankenstack Builder (laughs):
“Proven by who, though? You think the mega-lenders waited for third-party certification before deploying AI at scale? If you’re not experimenting yourself, how do you even know what’s reliable?”
Strategic Sleeper (smirking):
“And what’s the cost of your inefficiency, Frankenstack? Every time your stack breaks, your ops team goes full firefighter mode. I’m not in this game to be a tech company. I’m a lender. We fund loans. I’ll let others burn capital on R&D.”
AI Romantic (arms dramatically spread):
“Okay, but what if rates drop tomorrow? Mini refi boom hits. You gonna handle that volume with interns and duct tape? I’ve got AI ready to scale now. Vallia Voice-to-LoanApp doesn’t take PTO. Neither does Vallia Lead Expeditor. My bots are caffeinated and ready.”
Cautious Strategist:
“But what if your bots hallucinate? What happens when AI misclassifies a document and a bad disclosure goes out? That’s not just a compliance slap—that’s borrower trust down the drain. I need to control for exceptions, not chase edge cases.”
AI Romantic:
“You assume human processes don’t make mistakes. Spoiler: they do. And they cost more. My approach is about velocity and learning. AI doesn’t get tired. And with feedback loops, it gets smarter faster than your team can.”
Strategic Sleeper:
“And what if this AI utopia you’re building crumbles when the vendor folds or gets acquired? What’s your continuity plan?”
Frankenstack Builder (grinning):
“I build mine to avoid that. Vallia Data Connect ensures if a CRM dies, we plug in a new one. You sleep while I build resilience.”
Cautious Strategist (points across the table):
“Is faster your goal? Or is it smarter? Or scalable? Because you rarely get all three at once. I’m picking smart. Then scalable. Then fast. In that order.”
AI Romantic (rolling eyes):
“Meanwhile, the borrower doesn’t care if you’re smart—they care if you’re fast. You can’t deliver a wow experience at a crawl.”
Strategic Sleeper (shrugs):
“Speed is useless if you’re going the wrong direction. Six months from now, I’ll know which tools are hype and which are hero. Then I’ll buy the right ones—discounted, tested, integrated.”
Frankenstack Builder:
“And I’ll already be live with five tools that talk to each other. Will some fail? Sure. But the cost of not trying is falling behind permanently.”
Cautious Strategist:
“You’re assuming AI is the only road forward. Maybe the question isn’t ‘should we use AI,’ but where and why. I’m not slow—I’m surgical.”
AI Romantic:
“Yeah? And when your surgical team is still prepping, I’ll be three loans ahead, served with a smile and a chatbot.”
Whew. Now that’s how you stress test a strategy.
Bottom Line: Choose Your Fighter—But Choose Soon
In the end, there’s no perfect path—just the one that best aligns with your risk appetite, operational maturity, and vision for the future. The debate between our personas isn’t just theatrics—it’s a reflection of the real questions every lender should be asking:
- What’s your true goal? Is it speed? Scalability? Stability?
- Are you willing to absorb early-adoption bumps for long-term competitive edge?
- Can your current tech and team handle another rate dip and a sudden volume spike?
- If the economy contracts, will you regret overcommitting to unproven AI?
- If it expands—or refis roar back—will you be kicking yourself for not having automated more?
Six to nine months from now, will you be:
- Refining your AI playbook?
- Desperately trying to catch up to competitors who already have one?
- Celebrating a brilliantly-timed entry onto a proven platform?
- Or stuck managing manual workarounds in a market that’s already moved on?
You don’t need to bet the farm. But you do need to place a bet. Because the only thing more dangerous than moving too fast… is standing completely still while the world retools around you.
So take a hard look at your roadmap. Take a smarter look at your partners. And decide: what kind of lender do you want to be when the next shift hits?
