Halden & Ross is a demonstration build by Orabeam. The practice is fictional and has no clients, credentials or results. Every statistic on this page comes from Gallup, SHRM or the ICF and is cited.
Gallup puts it at 52%. It also finds that 51% had no conversation about their satisfaction or their future in the three months before they left. That gap is the entire case for this work — and it is arithmetic you can check, not a transformation story you have to take on faith.
Free · No email required · Gallup, SHRM and ICF figures, all dated and linked
Where these numbers come from
The 52% and 51% figures are self-reported by people who had already decided to leave. That is a real limitation and we would rather name it than hide it. Treat them as the size of the addressable problem, not as a guarantee of what coaching recovers.
It is the most quoted number in this industry, and you will find it on almost every competing site. Here is what it actually is, and why we have built the calculator to work without it.
The figure — a median ROI of 7× for companies, 3.44:1 for individuals — comes from the ICF Global Coaching Client Study, whose original data wave is from 2009. It is recited in 2026 marketing as though it were current research. The related “5.7×” Manchester study is older still, from 2001.
The respondents are companies and individuals who bought coaching, asked afterwards how it went. There is no comparison against organisations that did nothing, no randomisation, and a powerful incentive toward post-hoc justification. Nineteen percent reporting a return of “at least 50×” should be read as a fact about survey design, not about coaching.
The honest case is smaller and much harder to argue with. Losing one mid-level person costs roughly 80% of their salary; losing one leader, roughly 200%. Set that against the cost of an engagement and ask how many retained people it takes to break even. Almost always the answer is less than one. That claim survives scrutiny. A 7× claim does not.
The whole case, drawn to one scale. No multiplier, no survey.
A self-reported fifty-fold return is not a measurement. It is an opinion with a number attached to it, offered by someone who has already spent the money.
Set your numbers. Everything updates instantly, itemised, with the study behind each line. Nothing is sent anywhere — the model runs in your browser and every constant is inspectable at window.HaldenRoss.model.
Fees below are published market ranges for executive coaching, not this practice’s rates — a demonstration practice has none. The ICF puts the 2025 North American average at $297 an hour across all coaching types; senior and C-suite work sits well above it.
A named leader, a specific inflection: first ninety days in a new seat, first time carrying a P&L, first time the people who used to be peers report to them. The engagement has an end date because the transition does.
Where the problem is not one person but the space between several. Usually surfaces as decisions that keep getting reopened, or a leadership group that agrees in the room and not afterwards.
For a chief executive who has nobody left to think out loud with. No curriculum, no framework, no worksheets. Monthly, open-ended, and cancellable at any point with no notice period.
On these ranges
Ninety minutes, no charge. What is actually happening, who else has tried to fix it, and what would have to be different in a year. If the answer is that this is a structural problem wearing a coaching costume — a broken incentive, a role that cannot be done as written, a decision the board will not make — we will say so, because coaching does not fix any of those.
Six to eight people around the leader: the person they report to, their direct reports, one or two peers. Confidential, attributed only in aggregate. Almost everything useful in the first three months comes out of these rather than out of an assessment instrument.
Sessions are confidential. The sponsor receives progress against the agreed goals and nothing else — no content, no anecdotes, no characterisation. Written down before the first session, because a leader who is managing what gets reported back is not being coached.
At the halfway mark, an explicit review with the sponsor and the leader: is this working. If it is not, the engagement ends there and the remaining fee is not charged. A coaching relationship that cannot be ended is a subscription, not an engagement.
Honestly: the evidence is weaker than the marketing. What exists is largely self-reported satisfaction and self-reported return from people who bought it, with no control groups. There is a well-documented correlation between coaching and employee engagement — 72% of respondents in the 2023 ICF/HCI work saw a link — but correlation is what it is. What we can say without stretching: Gallup finds half of voluntary departures are described as preventable by the people who left, and half of those people had no conversation about their future beforehand. Coaching is one way to make that conversation happen reliably. That is a defensible claim. “Transformation” is not.
Because the direction of the bias matters. A higher assumed replacement cost makes the programme look better — fewer retained people are needed to cover it. So the number that is hard to argue with comes from the conservative end, half a year’s salary. This matters most for senior roles, where Gallup’s own 200% figure sits exactly at the top of its one-half-to-two-times range; quoting only that would put every estimate on the most flattering value available. The calculator shows both and leads with the cautious one.
Because it measures something narrower. SHRM’s 2025 average cost per hire — $5,475 non-executive, $35,879 executive — covers recruiting: advertising, agency fees, HR time, at a median 44 to 45 days to fill. It does not include the six to twelve months before a replacement reaches full productivity, or what the team does in the meantime. The Gallup multiple does. Both are on the page precisely so the gap is visible rather than quietly exploited.
Then retention is not your reason to do this, and the calculator will say so in exactly those words. There are good reasons to engage a coach that have nothing to do with turnover: a specific leader in a specific transition, succession readiness, a decision that keeps getting deferred. We would rather sell you the right thing for the right reason than build a spreadsheet that manufactures a case.
Written down before the first session. Session content is confidential without exception. The sponsor receives progress against the goals agreed at the outset — whether they are being met — and nothing about how the conversations went or what was said. Stakeholder interview input is attributed only in aggregate. The single fastest way to make coaching useless is for the person being coached to be managing what gets reported upward.
No, and this is the point in the page where it matters most to be straight with you. Halden & Ross is a demonstration practice built by Orabeam to show what a real coach’s site should look like. It has no clients, no credentials, no case studies and no references. Rather than fill this section with the invented CEO testimonials that are standard in this market — “I was promoted to CEO of a $200M company…” — there is no testimonial section at all. Real client references go here at build
Ninety minutes, no charge, and it is genuinely allowed to end with us saying this is not a coaching problem. That happens often enough that it is worth putting on the website.
Under a minute. We come back with two or three times that work.