By Bri Lees
Mortgage Marketing Thought Leader & Fractional CMO
Main Facts: The Synthetic Sea of Modern Mortgage Marketing
Log out of ChatGPT. Not forever. Just for the length of this analysis. You can log back in at the end, and by then, I’ll tell you what to do differently.
Consider this real-world scenario: One of the most-followed executives in the mortgage industry recently paid a top-tier LinkedIn "voice" several thousand dollars for an intensive, eight-week masterclass. The sole objective? Learning how to leverage artificial intelligence to produce 100% of his executive content. He couldn’t wait to brag about it.
Across the financial services sector, a quiet revolution—or rather, a homogenization—is underway. Leaders are trading human nuance for algorithmic speed. Traditional ghostwriters, coaches, and boutique monthly writing services—long responsible for shaping half of the recognizable executive brands in mortgage—are being unceremoniously replaced by large language models (LLMs).
The result is what industry experts call "the great averaging." Because foundational AI models are trained on virtually identical corpuses of financial and business knowledge, they inherently produce essentially the same outputs. Lenders utilizing half a dozen different models to build websites or draft thought-leadership pieces ultimately arrive at the same visual and linguistic destination.
When executives hand their personal and corporate brands over to a bot, they are systematically averaging themselves out. They are merging into a homogenous sea of digital sameness, trading authentic thought leadership for hollow, mechanical efficiency.
Chronology: From Human Extraction to Algorithmic Deception
To understand how the mortgage industry reached this precipice of synthetic uniformity, we have to look at how executive communication evolved over the last two decades.
- The Pre-Crisis Era (Pre-2008): Executive voices were largely unpolished, direct, and driven by localized relationship-building. Leaders spoke off-the-cuff, and marketing was secondary to operational reputation.
- The Post-Crisis Consolidation (2008–2015): The fallout of the subprime mortgage crisis traumatized the industry. Lenders learned a harsh, enduring lesson: outspokenness carries massive risk. The era of the "safe statement" was born. Marketing departments quietly adopted a corporate mandate of don’t upset anyone.
- The Rise of the Ghostwriter (2015–2022): Recognizing that executives needed to build digital footprints on platforms like LinkedIn without exposing the company to regulatory or public relations missteps, the industry professionalized content creation. Boutique services and ghostwriters paired with executives, conducting deep-dive interviews to extract authentic, lived experiences and translate them into compelling thought leadership.
- The Generative AI Boom (2023–Present): With the advent of accessible LLMs, the human interview became a bottleneck. Executives began bypassing the painstaking extraction of real thinking. Why spend an hour talking to a writer when a prompt can generate fifty "clean" LinkedIn posts in sixty seconds? The professional ghostwriter was swapped for a bot, marking the dawn of the great averaging.
Supporting Data & Industry Insights: The Anatomy of "Fine"
The shift toward algorithmic content has not gone unnoticed by those engineering the future of mortgage technology.
Tela Mathias runs PhoenixTeam, the organization that built and operates the Mortgage Bankers Association’s AI Mortgage Change Champion program. When asked what is going wrong with the industry’s rapid adoption of unvarnished generative AI, Mathias points directly to the core flaw of foundational models.
"We call that phenomenon the great averaging," Mathias explains. "All of the foundation model providers are trained on essentially the same body of knowledge. They will produce, all things being equal, basically the same result. A lender could build the same website with six different models, and the result will look the same."
This technological reality exacerbates a long-standing weakness in mortgage marketing: prioritizing assets over outcomes. For years, marketing teams have measured success by the volume of content produced rather than the behavioral changes or business outcomes driven. AI has merely accelerated this dysfunctional assembly line.
The traditional ghostwriter was paid to dig for diamonds—to extract raw, uncomfortable, highly specific thinking from a busy executive’s brain. The bot, by contrast, skips the interrogation entirely. What it hands back is a statistical average of what everyone else on the internet asked it to write.
To combat this, modern leaders must build a brand backbone. That backbone consists of three non-negotiable elements: Taste, Guts, and Receipts.
Official Responses and Expert Analysis: The Three Pillars of a True Brand
If mortgage executives want to survive the impending flood of synthetic noise, they must consciously audit their content against three rigorous standards.
Part 1: Taste
Taste is knowing precisely what sounds like you, and having the ruthless discipline to throw out everything that doesn’t—even if the discarded drafts are genuinely well-written.
AI will hand you fifty grammatically pristine drafts in under a minute. Every single one of them will be fine. But think about your daily digital consumption: you have read a thousand "fine" LinkedIn posts this year, and you cannot repeat a single one back to me right now.
When you read client social feeds, competitor posts, or industry commentary, the hallmark of AI-generated text is unmistakable. Even when scrubbed by someone who thinks they hid the tracks, the rhythm evens out. The word choices flatten into a monotonous corporate drone.
Consider this ubiquitous, AI-generated trope that has flooded feeds across the sector:
"Rates just dropped by 25 bps. Here’s what that really means for your buyers. In this market, speed wins. This isn’t just a rate change. It’s an opportunity. Who’s ready?"
You have likely posted some version of that yourself this year. And it is entirely forgettable. Hold every draft to a harder test: Tomorrow morning, without looking back at it, try to quote one line from the draft you wrote today. If you can’t, kill it before you write the next one.
Part 2: Guts
Guts is infinitely harder to cultivate because the mortgage industry spent a decade training it out of its leadership. We watched what happened to the loud voices during the 2008 financial crisis and learned the survival lesson too well: round yourself down, say the safe thing.
Many marketing meetings operate under an unspoken, universally understood strategy: don’t upset anyone. It works wonderfully—nobody gets upset. But crucially, nobody remembers you, either.
Being careful is actively costing lenders money today.
- The high-performing loan officer you want to recruit is reading your LinkedIn profile the night before her discovery call with you, and your feed sounds identical to the four other companies currently courting her.
- The Realtor partner you desperately need scrolled past nine identical rate-drop posts this morning and could not tell you who posted a single one of them.
Worse yet, that same Realtor is now asking ChatGPT about you. Your brand used to be what people said about you when you left the room. Today, your brand is also what the algorithmic model says about you when someone asks it for a recommendation. If everything you have ever published is average, the model has nothing distinct to tell her. You are effectively invisible at the exact moment she is conducting due diligence on your firm.
Most people mistakenly believe they already possess guts. They write clichés like "It’s about relationships, not rates" or "People do business with people" and feel briefly courageous. Because nobody disagrees with those statements, they feel safe. But that lack of friction is the tell.
A real position requires a face attached to a disagreement. If you cannot name the specific person in your market who would aggressively push back on your sentence, you haven’t written a thought-leadership piece—you’ve written a platitude. Guts means having at least one sentence you would eagerly defend in a room full of peers who vehemently disagree with you.
Part 3: Receipts
Finally, there are receipts—the one element of brand building that cannot be faked.
In a digital feed saturated with synthetic, frictionless content, radical specificity is the only reliable proof that a human being is behind the keyboard. It requires a specific date, a precise dollar amount, or the messy details of a deal that blew up at 4:00 PM on a Friday.
Publish average, AI-generated content long enough, and you systematically launder yourself out of your own professional record. Eventually, even a great ghostwriter—human or machine—has nothing authentic left to extract from you because you have stopped making distinct professional choices.
A receipt is defined by a decision and a cost. If a statement lacks both, it is merely an opinion.
- The Cost: Holding the line on quality. For example, a fractional CMO turning back a clean, technically on-brand draft to a client with the mandate: "You can’t publish this; it took one brain cell to write." The financial cost is real—a faster "yes" would make the client temporarily happier and the retainer easier to justify. But holding the line protects the client’s ultimate reputation, because their name is the one on the dotted line.
The ultimate test for a receipt is simple: Can you point to something you said or did before the rest of the room agreed with you? If you cannot, what you are running is not a thought-leadership campaign—it is merely an automated posting schedule.
Implications: Finding the Sentence the Model Couldn’t Write
So, let’s return to the exercise proposed at the outset. Log back into ChatGPT.
Ask the model to write the exact post that every executive in the mortgage industry will publish regarding today’s market news. It will generate it instantly, flawlessly, and blandly.
That is your banned list.
Take that output, point the "great averaging" directly at itself, and then write entirely around it. The digital feeds are already overflowing with posts that could apply to any housing market, in any year, written by anyone with an internet connection. Your brand shouldn’t be one of them.
As for the mortgage executive who paid thousands of dollars to skip the only step of content creation that was ever actually worth paying for—the messy, uncomfortable human extraction of real thought—one can only hope he manages to get a refund.
Bri Lees is a fractional CMO and mortgage marketing thought leader. This column does not necessarily reflect the opinion of HousingWire’s editorial department and its owners.