How We Price
A fixed price, a metric in the agreement, and 20% of our fee riding on it. Here is the whole mechanism, including the parts that limit it.
Hourly billing is not dishonest, and the firms that use it are not villains. It is a risk-allocation choice: the buyer carries all of the uncertainty. If the work takes twice as long as anyone expected, the invoice doubles, and the firm doing the work is no worse off.
That arrangement makes sense when nobody can scope the job. It makes much less sense for the problems we take on, which we have generally seen some version of before. If we have seen it before, we should be able to price it — and if we're wrong about how long it takes, that should be our problem, not yours. So we quote a fixed number, and we hold it.
Three numbers, every engagement
One line each. They are not the same number, and we show you all three.
Expected value
The full business case — every driver, with our confidence level and the assumptions behind it.
What we measure
The subset we can read from your own systems, against a baseline we capture with you before work starts.
What’s at risk
20% of our fee, tied to those measurements hitting the targets written into the agreement.
Illustrative example — not a client result.
S4 is a new firm. These figures show how the arithmetic works, not something that has happened. Real engagement results will be published on this site as they close.
Illustrative engagement: a 45-person specialty distributor with manual invoice processing
| Driver | Type | Est. annual value | Confidence | Metric | Baseline | Target | Carries fee |
|---|---|---|---|---|---|---|---|
| Reduce invoice handling time | Efficiency | $84,000 | High | Minutes per invoice processed, from ERP timestamps | 14.2 min | 4.0 min | Yes |
| Cut exception rework | Efficiency | $31,000 | Medium | Exceptions per 1,000 invoices, ERP exception queue | 62 | 20 | Yes |
| Audit-ready approval trail | Compliance | $18,000 | Medium | % of invoices with a complete approval chain, ERP audit log | 71% | 99% | Yes |
| Faster payment → early-pay discounts | Revenue | $40,000 | Low | — | — | — | No |
| Better vendor relationships | Competitive advantage | Qualitative | Low | — | — | — | No |
Expected value
$173,000 a year, plus the qualitative items
Measurable value
$133,000
Fixed price
$46,000
At risk
$9,200 (20%)
Guaranteed on delivery
$36,800
Read it in one sentence: we expect this to be worth about $173,000 a year to you; $133,000 of that is measurable from your ERP against a baseline we capture together before we start; our fee is $46,000, and $9,200 of it is contingent on those three metrics hitting the targets in the agreement. If they don't, you don't pay it.
Note that the revenue driver stays in the proposal. It is 23% of the case and it is real. Our fee just isn't exposed to whether your AP team actually takes the early-pay discounts.
Where the money sits
Every engagement uses this shape. Only the numbers change.
| Component | Paid when | Typical share |
|---|---|---|
| Mobilization | On signature | 25–35% |
| Delivery | On acceptance of the deliverables | 45–55% |
| Outcome — at risk | At the close of the measurement window, if the targets are met | 15–20% |
The total is fixed and stated up front. There is no hourly component, no rate card, and no "estimated" total.
Why not 100% at risk
Because you'd pay for it. A firm staking its entire fee on a number it only partly controls prices a risk premium into the quote, and you fund that premium whether or not anything goes wrong. Twenty percent is real money to a firm our size, and it is honest arithmetic. We'd rather show you a smaller number that we mean than a larger one hedged in the price.
What we will and won't stake fee on
This is the part most firms leave out. It is the part that makes the rest credible.
| What we're improving | Fee behind it? | Why |
|---|---|---|
| Efficiency — a process gets faster or cheaper per unit | Yes | Short chain between our work and the number. Usually the core of the engagement. |
| Compliance — a control exists or it doesn’t | Yes | Often binary, and readable from your audit log as control coverage. |
| Risk reduction | Sometimes | Only when it can be stated as a leading indicator we can read — control coverage, incidents per period, mean time to detect. Never as “losses avoided”. |
| Revenue growth | No | Depends on your sales team, your pricing, and your market. We’ll quote the value; we won’t stake fee on it. |
| Competitive advantage | No | Real, and not measurable inside a contract window. We describe it; we assign it no fee-bearing dollar value. |
Here is the full value we expect this to create. Here is the part we can measure precisely enough to put our own fee behind. We only guarantee the part we control.
How we decide whether a metric can carry fee
A number has to pass all three. If it fails one, it stays in the value case and out of the fee.
-
1. We can read it from a system you already run.
ERP, CRM, ticketing, time tracking, a database, a log. If measuring it needs a survey, a judgment call, or someone's recollection, it can't carry fee. And if the number can't be pulled from that system today, at quote time, there is no baseline — and there is no such thing as a retroactive baseline.
-
2. There's almost nothing standing between our work and the number.
Count the independent parties in between. One is safe: quote turnaround time, where the tool we build is the change. Two is negotiable. Four — say, sales win rate, which runs through turnaround, pricing, your reps, and the market — is not a metric we'll stake fee on.
-
3. It's a rate, not a total.
"Minutes per reconciliation", not "hours per month on reconciliation". "Exceptions per 1,000 invoices", not "total exceptions". Totals move when your business moves, in both directions, for reasons that have nothing to do with us — and we're not interested in getting paid for your good quarter any more than you're interested in paying us for your bad one.
Every metric that carries fee gets a row in a measurement schedule — Exhibit A — attached to the statement of work. The row names the metric, the unit, the system it's read from, exactly how it's extracted, the baseline, who attested to the baseline, the target, the measurement window, and the share of our fee riding on it.
We publish a sample of that schedule, filled in, so you can see the level of specificity before you talk to us.
See a sample Exhibit AHow this becomes a contract
Nothing on this page is an offer or a set of terms — it's a description of how we price. Real terms are set during the diagnostic, when we've seen your systems and know which numbers are actually readable. What comes out of that is a proposal with specific metrics, baselines, and a specific fee split, and if you accept it, that's what gets signed. Until then this is just how we think about it, in public, so you can judge it before you spend anything.
What things cost
A page about pricing certainty that contains no prices is not worth much.
| What it is | Price | |
|---|---|---|
| Value Diagnostic | 2–3 weeks. Discovery, a written value model, and a measurement plan for the top three to five opportunities — what's worth fixing, what it's worth, and which parts we could put our fee behind. | $5,000–$9,500, fixed. Credited against an engagement started within 60 days. |
| Outcome Engagement | 6–16 weeks. Implementation against defined outcomes, with the metrics and baselines in the agreement. | $25,000–$75,000 typical, fixed, with 15–20% at risk. |
| Advisory Retainer | Ongoing advisory. Fixed monthly, defined inclusions, no hourly true-up. Offered after an engagement, not before. | Quoted per scope. No outcome fee — retainer scope is deliberately open, so there's nothing honest to measure it against. |
FAQ
What if we don't hit the target?
You don't pay the outcome fee. That's the deal, and it's the whole reason the number is written down before we start.
What if the delay is on our side?
The things we need from you — system access, a decision within a stated number of days, a named counterpart — get written down as preconditions alongside the metrics. If one of them fails and we flag it in writing at the time, the clock pauses or the metric is treated as met. We don't get to invoke that quietly after the fact.
Why only 20% at risk?
Because 100% would mean pricing in a risk premium you'd pay for either way. Twenty percent is real money to a firm our size, and it's arithmetic we can actually stand behind. We'd rather stake a number we mean.
What if the scope changes?
A written change order with the fee and schedule impact stated, agreed before the work proceeds. A fixed price only means something if changing it is deliberate and visible.
You're new — why should we believe your estimates?
You shouldn't, yet. What you can do is check the method: we show you every assumption, every metric, and every baseline before you spend anything past the diagnostic. And we'll publish what we learn when we're wrong, including the engagements where the target wasn't met.
Who is this not for?
If the problem can't be read from a system you already run, or if what you want is a body at an hourly rate, we're the wrong firm and we'll say so on the first call.