What We Look For Before We Ask You to Spend Anything

You have heard the pitch. A firm walks you through a plan, the plan is good, and somewhere behind your eyes a different conversation is happening: this sounds right, I do not have the budget for it, and I am not sure I can justify it to my partner, my board, or my banker.

Nobody says that part out loud. It kills the engagement anyway.

So we do something before we ask you to fund anything. We look for money the business has already earned and is not collecting. The principle is simple enough to put on one line, and we run it on every engagement: find the money first, prove it can be moved, then scale.

We call it the Wedge Layer. It is not a service. It is a search, and it happens during Discovery before any recommendation is made.

The four places money hides

Pricing that drifted

The most common finding and the least examined. Prices set years ago. Discounts that started as exceptions and quietly became defaults. A rate card that has not moved while input costs have.

The leverage here is unlike anything else available to a business. McKinsey's analysis of S&P 1500 economics found that a 1% price improvement, with volume held flat, produces roughly an 8% increase in operating profit — 8.7% for the typical Global 1200 company (McKinsey, The Power of Pricing). No campaign competes with that arithmetic. A price move on something already on your menu or price list drops almost entirely to contribution margin, because the cost is already incurred.

Settings inside your own systems

Not strategy. Configuration.

An upcharge entered in the wrong units. A shipping rule nobody revisited after a carrier change. A discount code with no expiry date. Duplicate line items that split reporting so that nothing looks big enough to be worth fixing.

These never appear on a P&L as a line, which is exactly why they survive. They appear as a margin that is slightly thinner than it should be on a popular item, and everyone has long since stopped noticing.

Demand you already generate and lose

Enquiries that go unanswered. A website converting at a fraction of what it should. A quote process that outlasts the buyer's patience. A form that has been silently failing for months.

This is the cheapest revenue in any business, because you have already paid to create it. In manufacturing, the arithmetic on that last one is stark: quote response speed tracks directly to win rate, with sub-five-minute responses benchmarked closing around 32% of quotes against roughly 7% for responses past 24 hours (Uptool, citing manufacturing benchmarks), while the average general job shop takes about 3.8 days to turn a quote around (Bloomfield). Compressing that costs process work, not media spend.

Relationships you already own

Customers who bought, were happy, and were never contacted again. Accounts you serve in one location and not the other three. A callback programme that worked and quietly stopped when the person running it got busy.

In most businesses this is the largest pool of the four, and almost nobody works it — not because it is hard, but because it is not in anyone's job description.

How far we actually go

A finding here has to survive an owner saying prove it. So the standard we hold ourselves to is this: a recommendation is only included if we can name every competitor used and show exactly where you rank against them.

On a recent pricing engagement, that meant just under 600 priced items examined line by line — not a category-level percentage applied across the board, but each item with its own current price, recommended move, and stretch position. It meant using realised prices rather than menu prices: what customers actually paid, total sales divided by units, across a 24-month window, because a printed price and a collected price are different numbers once discounts and overrides are in play. Every data file had to reconcile to the cent against its own sales report before we would use it; files that did not reconcile were excluded rather than smoothed.

The competitor set was named — 86 operators across four local markets, several thousand captured price lines, each carrying its source type, because a price from an operator's own site and a price from a delivery marketplace are not the same number and blending them produces something nobody can defend.

Each item carried base, optimistic, and pessimistic elasticity, so the model showed what happens if the recommendation is wrong.

And we imposed a ceiling on ourselves. Every recommended move went to the local peer median and no higher — not to what the market would bear, but to the middle of it. That deliberately leaves value on the table in exchange for being unarguable.

Three more things went into that work that most firms would have deleted. The research base behind it carries a section titled Claims I Would Not Make. When our own earlier figures did not survive audit, they were withdrawn inside the deliverable in writing rather than quietly corrected. And the single largest line in the analysis was not yet defensible on the available evidence, so we pulled it out and subtracted it from our own headline number.

Separately, the same engagement's system audit found a premium upcharge configured in cents rather than dollars. Nobody had noticed, because it never presented as a problem — it presented as a slightly thin margin on a popular item. That one setting was worth roughly $16,000 a year, and correcting it cost nothing but the time to find it.

What it looks like in practice

A specialty industrial manufacturer had a product with a natural reorder cycle, which meant every lapsed account represented a reorder that had simply stopped with nobody aware of it. Alongside that, an online store sat behind a website that made its products genuinely hard to find and buy. Neither required a single new customer. Together, reactivating dormant accounts and unblocking direct purchasing was modelled to recover roughly 91% of an entire twelve-month program cost before one net-new customer was counted.

A long-established multi-location restaurant group was serving genuinely premium ingredients and pricing several of them like a coffee shop. The finding was not that everything was underpriced — three of eleven premium categories were already at or above their local median, and we said so. The gap was concentrated in a handful of categories where the low price had never been the reason guests came. The first recommended move cost nothing to implement and was measurable within weeks.

A commercial contractor with four decades of history had roughly a thousand active customer relationships with no contact cadence at all, multi-site customers served in one location but not the others, and a warranty callback programme that had worked and stopped. None of it needed new market demand. It needed a system to work what already existed.

A manufacturer with a superior product had built its own wall. Requesting a sample — the highest-intent action a prospect could take — was not in the main navigation, and a visitor who found it faced roughly eighteen required fields and then a mandatory phone call before anything would ship. Compliance documents buyers need in order to specify a product sat behind a form. The wedge there was not money in an account. It was demand the company was already paying to create, and then blocking on arrival.

And sometimes there is nothing there

This is the part that makes the rest of it credible, so we will be direct about it.

The wedge is not always available.

Some businesses are priced correctly and run tightly. There is no leak worth naming, and looking harder would only mean finding what we wanted to find.

Some are demand-constrained rather than conversion-constrained. There is no captured demand to recover because not enough demand is arriving in the first place — a genuinely different problem requiring a genuinely different answer. We have delivered diagnostics where the honest finding was that the headline product was already priced below market and raising it would have been the wrong move.

And some businesses have economics where a self-funding claim would simply be misleading. For an owner-operated company, the honest financial frame is seller's discretionary earnings rather than EBITDA, and the arithmetic does not hold the same way. We have written analyses that carry no self-funding claim at all, deliberately, because the evidence did not support one.

A fourth case is more common than any of them: the wedge probably exists, but it cannot be identified yet because the data required to find it has not been supplied. In that situation we present the candidates and defer the decision, rather than picking one and defending it later.

We look every time. We do not always find. We never manufacture.

Why our fee does not move

There is a fair objection to everything above, and it deserves a direct answer rather than a careful one.

If they are paid more for finding money, they will find money whether or not it is there.

Our fee is a flat monthly amount, fixed before the work begins and unchanged by what we turn up. No share of recovered margin. No success fee. Nothing that pays us more for producing a bigger number.

That is why we can tell you there is nothing to find. And why, when we do find something, you can believe it.

The sequencing is the point

When there is a wedge, everything changes shape. The cheapest, lowest-risk move goes first and shows up in your own numbers within weeks. The build is funded by those results. The expansion is funded by compounding revenue.

The risk is sequenced to come last. The bet on growth is placed with the house's money — not yours.

And when there is no wedge, you find that out in Discovery rather than in month six.

Where growth breaks in your business is rarely where you think it is. The Outside-In Analysis is how we start looking — six documents, no cost, and nothing required from you but your domain.