BEHAVIORAL GROWTH ENGINEERING
The first conversation. The price question. The anxious wait. The silence after the sale. Those moments run on documented decision science — not on instinct, and not on guru talk. We redesign them, grade every technique we use, and measure the difference honestly.
FIXED SCOPE · FIXED FEE · YOURS TO KEEP WHETHER OR NOT YOU HIRE US AGAIN
THE STANDARD
Anything we cannot grade at Moderate or above, we do not recommend at all.
01
Almost every business believes its revenue is decided by price or by volume. In owner-led service businesses it is usually decided in a handful of conversations, each one a decision made under uncertainty by a person who is slightly afraid.
01
Roughly half of buyers seriously consider only one provider, and most never request a second quote. Shopping rises with confidence, not with price sensitivity. The first competent-feeling conversation usually ends the search — which means the discovery call is not a step in the process, it is the process.
02
A price is never judged in isolation. It is judged against a reference point the customer brought with them — an old quote, a competitor's teaser, a number from three years ago. Whoever sets the reference point first is usually negotiating from inside the customer's head.
03
Between the yes and the delivery, the customer is alone with their doubt. What they remember afterwards is not the average of the experience — it is the peak and the ending. The worst moment, handled well, becomes the story they tell about you.
04
Customers who should obviously come back frequently don't, and reminders barely move them. The documented blocker is suspicion of motive: if this were good for me, why are you pushing it? The cure is a demonstrated willingness to advise against the transaction.
WHY THE NAME
In the mountains where this practice was founded there is a building called the Casa de Fundição.
For most of the eighteenth century, no gold left the region until it had passed through it. Ore arrived as a claim — this is gold — and the claim, on its own, meant nothing. The metal was melted, tested for fineness, weighed, taxed, and only then stamped. The stamp did not say the gold was good. It said how good, in parts per thousand, on the record, signed by someone who could be found afterwards.
Business advice has never had that building.
An idea arrives as a claim — this works — and it travels on the confidence of whoever is carrying it. Nobody asks what it is made of. Nobody asks how much of it is replicated science and how much is a story that closed well in a room once. The industry that sells influence for a living has, with some irony, no chain of custody for its own ideas.
Assay is the building.
02
Every finding we deliver is written in the same four parts, so that you can check us. Nothing arrives as an assertion.
01 · OBSERVATION
Your past-client database receives rate alerts whenever the market moves in their favor, and almost nobody responds. The borrowers you closed at 6.5–7% are sitting on money, the offer is genuinely good, and recapture still lands in the low twenties. Nothing in the outreach was wrong. Nothing in it was answered either.
02 · MECHANISM
Two processes, not one. Loss aversion and status-quo bias create the inertia — the borrower treats their existing loan as the reference point and any change as a risk to be avoided. Then suspicion of the messenger blocks the cure: 'if this were good for me, why are they pushing it?' More alerts feed the second mechanism. What defeats it is demonstrated willingness to advise against the transaction — an annual review that promises, in advance, that most years the answer will be do nothing.
03 · AUTHORITY
Keys, Pope & Pope (Journal of Financial Economics, 2016): roughly 20% of households for whom refinancing was unambiguously optimal had not refinanced, median forgone value about $11,500 each — and their own field experiment showed a mail reminder moved almost no one. Information alone does not cure inertia. Johnson, Meier & Toubia (Columbia GSB): among borrowers already pre-approved for advantageous refinance offers, about 51% did not apply, and of every behavioral factor tested, suspicion of the lender's motives was the one consistent predictor.
04 · GRADE
LOSS AVERSION · STATUS-QUO BIAS
SUSPICION AS THE BINDING CONSTRAINT · SINGLE FIELD LITERATURE
The underlying biases are consensus. The claim that distrust — not rate indifference — is what stops a pre-approved borrower is validated field evidence, not consensus, and we say so. Boundary condition: this holds for lender-initiated outreach. It does not license the same design where the borrower opened the conversation.
That is one finding, in the format all of them take. Note what the grade does: it tells you where to spend conviction and where to hold some back. A recommendation that cannot be graded is a design judgment, and we label it as one.
03
Each engagement has fixed scope and a fixed fee, and each ends with a go/no-go that you control. This is deliberate — the standard fear about consultants is lock-in, and preserving your exit at every gate is what makes the first step safe to take.
01
2–3 WEEKS
A measured audit of your funnel. You receive eight to fifteen findings in the four-part format, a map of the psychological moments that decide your revenue, the mechanical fixes you can ship the same week, and a modelled intervention portfolio with every assumption exposed. It is written to be worth keeping even if you never hire us again.
GATE: You decide whether anything follows, with the modelled numbers in front of you.
02
6–8 WEEKS
We implement the single highest-return, lowest-adoption-cost intervention from the Diagnostic. That intervention is built out, not sketched: point-of-use materials and job aids your Loan Officers reach for in the conversation, a resource library your team keeps after we leave, and a full Loan Officer Behavioral, Negotiation, and Sales Training — the scripts, the objection-handling, and the live drills that move the technique from the page into the room. Conversion architecture first, because changing the environment requires no behavior change from your staff; the training follows so the change survives the people who carry it. Includes the measurement spine: baseline, one primary metric, one guard-rail metric to catch quality dilution.
GATE: Measured delta against baseline.
03
ONE QUARTER
The flagship. Five layers installed — artifacts at the point of use, rituals your calendar can sustain, named roles, a measurement spine, and a feedback loop that retires what stops working. Plus the training curriculum and the negotiation playbooks for your recurring arenas.
GATE: The day-90 readout, written to be forwarded to your board unedited.
04
MONTHLY
Quarterly readouts, technique retirement and introduction, onboarding new hires into the system, and an evolving artifact library. This buys evolution, not dependence — the system is designed to run without us, and that promise is kept.
GATE: Every quarter, on the readout.
We don't discount. We trade. A reduced fee in exchange for the right to publish your numbers, or for scheduled introductions — never a lower price for the same work. Successive discounts only prove the first price was inflated.
04
The graded technique library that sits behind every recommendation we make. Published openly, including the part most firms would hide.
12 / 12 ENTRIES
Losses loom larger than equivalent gains, so a factual loss frame outperforms an equivalent gain frame.
LIMITS Requires a genuine loss. Manufactured urgency is detected and costs more trust than the frame gains.
Peer evidence from people visibly like the customer moves behavior; self-praise is discounted.
LIMITS Reverses if the proof reads as curated. Live numbers only, nothing edited.
In high-stakes sales, questions that develop the cost of the problem discriminate top performers from average ones.
LIMITS Excess situation questions read as unpreparedness. Research first, ask second.
Naming an emotion out loud measurably reduces its intensity, which is why it defuses hostile calls.
LIMITS Performative or inaccurate labeling backfires hard. Label only palpable emotion, never routinely.
An artificial head-start raises completion rates on multi-step processes.
LIMITS The head-start must be justified by something the customer actually did.
A small, voluntary, public first yes raises the probability of a larger later yes.
LIMITS Only if the first step is genuinely voluntary. Coerced compliance produces no consistency effect.
Separate people from the problem, interests from positions, and anchor on objective criteria.
LIMITS A decision structure, not an empirical finding. It organizes judgement; it does not replace it.
Defaults, sequencing and staged disclosure change what people choose without removing options.
LIMITS Choice-overload effects are heavily moderated and often absent. Test before assuming.
Sales frameworks claiming to target a 'reptilian brain' or to read buying signals from neuroscience.
LIMITS The neuroscience cited does not support the commercial claims. We do not teach it.
Fake countdowns, invented stock limits, artificial deadlines.
LIMITS Fails on detection, destroys the scarcity effect for genuine cases, and is illegal in most regulated verticals.
Rapport techniques from neuro-linguistic programming.
LIMITS Fails replication consistently. Where it appears to work, ordinary attentiveness explains the result.
An unrequested, personal, useful favor creates real obligation.
LIMITS Collapses entirely if perceived as tactical. A gift given to trigger the effect does not trigger the effect.
05
Each of these is refused for a stated reason, not for taste. They are in the Field Manual, in the Do Not Use tier, where anyone can check them.
06
We are a behavioral growth engineering practice. Most of a company's revenue is decided in a handful of human moments made under uncertainty, and those moments are governed by decision science that has been studied for fifty years. We redesign them — the funnel, the first conversation, the price discussion, the follow-up that never happens — and we install the artifacts, rituals and measurement that keep the redesign alive after we go.
Every technique we use carries an evidence grade: replicated science, large-scale field practice, real-but-bounded, or discarded. The discarded list is published. Anything that only works while the customer can't see it, we neither recommend nor implement — that clause is in the engagement letter.
THE METHOD, PUBLISHED IN FULL
An Evidence-Based Professional Manual
The methodology is not proprietary and it is not withheld. Every technique in it carries how well it actually replicates — including the popular ones we refuse to teach because they don't. Read it before hiring us.
Read a sample chapterNot a pitch. A working session on your data: we walk your funnel with you live and show you the first two or three findings in your own numbers rather than in our summary of them. If the answer is no, that is a complete answer, and you will not hear from us again unless you write first.
lsolmucci@gmail.com