Measuring ROI in Executive Coaching: Costs and Evidence

Measuring ROI in executive coaching requires a defensible monetary benefit attributable to the engagement, compared with its full cost. The usual net formula is: benefits minus costs, divided by costs, multiplied by one hundred.

The difficult work is deciding what belongs in those numbers. This guide examines five financial questions and completes an original fictional claim reconciliation: a sponsor proposes fifty percent coaching ROI, but the invoice comparison, changed supplier price and full initiative cost support a narrower conclusion.

It is a guide to reviewing financial claims, not an expected-return benchmark or investment recommendation. The research review examines published coaching-return reports; the broader measurement guide develops evaluation design. Here the focus is accepting, revising or excluding proposed monetary items.

Define the calculation before collecting benefits

Let B represent attributable monetary benefits and C the full relevant cost, on an explicitly stated basis. Net ROI is (B − C) ÷ C × 100. The benefit-cost ratio is B ÷ C. With benefits of $27,000 and costs of $18,000, the ratio is 1.5:1 and net ROI is fifty percent, not 150 percent.

Those formulas do not establish the inputs. An invoice total, a retrospective opinion and an estimated avoided loss have different evidential meanings. Even a precisely calculated percentage can rest on an unsupported benefit.

ROI Institute’s own methodology distinguishes business impact, monetary conversion, isolation of program effects, full costs and benefits left unpriced. Its process is more than applying a formula to a testimonial. Referencing it does not make this article or its original case a certified implementation of that methodology.

HM Treasury’s 2026 Magenta Book distinguishes implementation, impact and value-for-money questions. It also addresses what would have happened without the intervention. This is government evaluation guidance, not evidence of coaching returns or a private company’s accounting policy.

Specify the object. Is the question about coaching alone, a combined leadership initiative or a particular purchasing decision? A combined result cannot automatically be divided by the coaching fee to produce a coaching return.

1. Test retention against a credible alternative

A person staying in a role is observable. A departure avoided because of coaching is a further claim. Ask what contemporaneous evidence supports the departure risk, what changed and which other influences affected the decision.

Use relevant organizational records where an avoided replacement estimate is warranted. Recruitment fees, temporary cover and onboarding may have different timing and evidence. Do not apply a universal salary multiplier or price the person’s whole contribution as a replacement cost.

A planned departure can also be a reasonable outcome. A participant’s interest in leaving an unsuitable role should not be erased because retention looks better in the sponsor’s spreadsheet.

In the case below, a participant remains employed, but there was no documented plan to leave and no evaluation establishing an avoided departure. A proposed $75,000 retention benefit is excluded. That does not prove coaching had no personal value; it means this financial item lacks the required support.

2. Separate usable capacity from spending reduced

Define whose time changed, doing what, over which comparable opportunities. A leader’s estimate of less meeting time is not the same as a timed process record, and a quieter quarter may contain different work.

Then ask what happened to the time. A canceled external purchase or an actual reduction in paid overtime may change cash spending. Reallocating salaried time can create useful capacity while salary expenditure remains unchanged.

An internal hourly rate can express a resource valuation, if its meaning is explicit. It does not show that each released hour became additional profit. Demand, complementary resources and the work actually performed matter.

Check overlap. If the time enabled a commercial contribution valued elsewhere, adding its full hourly value may count the same benefit twice. If outsourced work moved to internal staff, include the relevant internal effort rather than calling the whole invoice reduction a free gain.

The demo form builder illustrates structured review prompts. A questionnaire response is not independently verified financial evidence or a counterfactual.

3. Examine incremental contribution rather than gross sales

Revenue after coaching can reflect an existing relationship, market demand, pricing, product changes or additional staff. Establish the event and period before assigning a cause.

Where a commercial benefit is supported, use the organization’s accepted contribution basis and relevant incremental costs. Do not equate a $40,000 contract with $40,000 profit. A $40,000 recognized sale with $25,000 of relevant incremental costs leaves $15,000 on that defined contribution basis, before any attribution question.

Keep revenue recognition and collection distinct. A contract, an invoice and a payment are different events. Use finance’s appropriate basis consistently rather than selecting whichever produces the most attractive number.

A sale also needs an incremental counterfactual. If the relationship and commitment predate coaching, the new invoice alone does not establish that coaching generated additional business. A useful commercial conversation can remain described without assigning it the entire contract value.

4. Do not price every conflict as a lawsuit avoided

Coaching may help someone prepare a clearer conversation, understand a disagreement or refer a conduct concern appropriately. Those outcomes can matter without a monetary legal claim.

A proposed avoided settlement needs evidence that the cost was a credible alternative, its relevant amount and coaching’s contribution. A worst-case number supplied after the event is not automatically a benefit.

Preserve the difference between coaching, mediation, investigation and legal advice. A private conversation cannot replace a formal process merely because someone assigns a large hypothetical cost to it.

In the fictional review, no claim, expected settlement or legal assessment supports the sponsor’s $100,000 “conflict avoided” item. The evaluator records a reported improvement in a working conversation without pricing a lawsuit.

5. Use the full cost of the same evaluation object

The coaching invoice may include provider preparation and delivery, but that does not necessarily cover participant time, coordination or evaluation. Define each resource, rate, period and source. Count costs once, with no hidden addition of provider effort already covered by the fee.

For a combined initiative, include its other components. A new process, paid cover or training is not free because it was funded by another department. Conversely, avoid adding an internal time estimate when that same work is already included in a paid contractor amount.

Distinguish an expenditure total from a resource-cost total. Salary time can be a real use of resources without being an extra payment made that quarter. Label a mixed cost basis clearly.

For longer periods, appropriate treatment of timing, currency, inflation and discounting may need finance or evaluation expertise. The short illustrative case uses one currency and agreed nominal internal valuations; it does not set a universal tax, accounting or discount-rate policy.

Complete fictional case: the fifty-percent headline

An invented company runs a combined leadership initiative from 1 October to 31 December 2026. It includes coaching, a redesigned approval form and temporary paid capacity. Sponsor Ed asks whether to repeat the package. Finance owner Rosa and evaluator Tomas reconcile the financial claims before Ed makes a new commitment.

The monetary-benefit window is October through December. The full cost boundary includes preparation and the January evaluation close, through 29 January 2027. No later benefits are assumed. This explicitly includes the work needed to review, correct and accept the final report.

The coaching provider’s fixed fee is $15,000. It covers that provider’s preparation, delivery and agreed finalization; no extra provider-hour amount is added. The company’s recorded participant, coordination and evaluation time uses separately agreed internal resource rates.

Ed initially drafts a private slide: “Coaching generated $27,000 savings on $18,000 cost: fifty percent ROI.” The arithmetic matches that claim’s inputs. Whether the inputs support the claim is still open.

Reconcile the external-spend comparison first

Rosa checks the same external-rework invoice category in two three-month periods. July through September contains 300 billed hours at $130, totaling $39,000. October through December contains 120 billed hours at $100, totaling $12,000. The recorded spending difference is $27,000.

She confirms a supplier rate change took effect on 1 October under a procurement arrangement independent of the coaching. She also confirms that the initiative changed internal handling of work. The data do not establish which part of the hour reduction came from coaching, the new form, temporary capacity, demand or other conditions.

Tomas makes an explicit rate-normalized comparison:

• At the new $100 rate, the earlier 300 hours would cost $30,000.
• The difference between the actual earlier $39,000 and that hypothetical $30,000 is $9,000.
• The difference between the normalized $30,000 and actual later $12,000 is $18,000, representing 180 fewer billed hours valued at the new rate.
• The two parts sum to the recorded $27,000 difference.

This is an accounting decomposition with a stated rate convention, not an identified causal split. The $9,000 is a price-only scenario at the earlier volume, not an additional cash receipt. The $18,000 still lacks an established no-initiative volume and a coaching-specific attribution.

Neither the original $27,000 nor the normalized $18,000 can simply be relabeled a proven coaching benefit. Comparable categories and exact arithmetic improve the description; they do not create a control group.

Check a second decomposition before assigning a cause

The same invoice difference can be decomposed another way. Value the 180 fewer hours at the earlier $130 rate: $23,400. Then value the $30 rate reduction on the later 120 hours: $3,600. Together these also equal $27,000.

Under the first convention, the price component was $9,000 and the volume component $18,000. Under the second, they are $3,600 and $23,400. The $5,400 interaction, 180 hours multiplied by the $30 rate difference, has been allocated differently.

Neither calculation is an additional benefit. Both are arithmetic descriptions of the same total. Neither tells Rosa how many rework hours would have occurred without coaching, or whether internal staff absorbed work that the supplier previously did.

State the convention instead of choosing whichever component makes the preferred story larger. This is why a finance check should include quantities, prices and changed work allocation, not only the invoice totals. It also shows why a neat division of the observed change into named components is not automatically a causal analysis.

Close the other proposed benefit items

The meeting estimate is a second, unrelated $9,000 calculation: sixty hours multiplied by $150. It is not the price-normalization amount above. Separate labels prevent matching numbers from being mistaken for the same evidence.

The sale concerns a different project from the rework process. Its $25,000 incremental delivery costs belong in that project’s contribution calculation; they are not silently added again to the leadership initiative’s cost table. No contribution benefit is admitted to this coaching numerator.

The excluded items are unsupported valuations, not measured zeros. The ledger cannot be summed into an attributed benefit simply by selecting a smaller percentage of each number.

Complete the full cost reconciliation

Rosa confirms that these are distinct allocations. Participant time includes the participants’ coaching, preparation and final review. Coordination includes Ed’s administration and report-acceptance work. Evaluation includes Rosa and Tomas’s combined analysis, correction and closing-review effort within the six person-hours. The separate analyst’s twenty hours are not participant or evaluator hours.

All these actual allocations are complete through the 29 January close, not forecasts with unpaid final work omitted. The $27,000 provider-plus-temporary-capacity charges are cash spending in this fictional record. The remaining $6,000 values internal time; the $33,000 is therefore a resource-cost total, not an additional-cash total.

The case assumes finance has confirmed no other relevant initiative cost within that boundary. Real evaluations need to check their own omissions and valuation basis. A table is complete because its scope was reconciled, not because it has a row called “total.”

Correct the claim without substituting another unsupported ROI

At the 28 January review, Tomas shows why Ed’s slide is misleading. It puts the whole external-spend difference over the coaching component’s cost even though the result concerns a combined change and includes an unrelated rate change.

Ed: “Then should we publish minus eighteen percent instead?”

Tomas: “No. That would still treat the whole $27,000 difference as an attributed benefit. We have corrected the scope, not established the counterfactual.”

Rosa: “We can confirm the invoices and the $33,000 resource cost. We cannot sign off coaching-attributed monetary benefits.”

Ed: “I will remove the coaching-return headline and use the qualified comparison internally.”

If someone mechanically puts $27,000 over $33,000, net arithmetic is ($27,000 − $33,000) ÷ $33,000 = −18.18 percent. That calculation demonstrates the effect of the cost correction only. It is not an accepted estimate of the initiative’s actual ROI, and it is not proof that coaching lost that percentage.

Tomas records the unresolved financial return as not established. He does not substitute minus one hundred percent: an unknown attributed benefit is not the same as an observed benefit of zero.

Do not use a small attribution percentage as a substitute for evidence

Suppose Ed suggests crediting only one quarter of the $27,000 comparison to coaching. That would put $6,750 in a proposed numerator. The smaller amount may look cautious, but the fraction still needs a reason grounded in the evaluation.

Ask what the estimate refers to. Is it a participant’s opinion about coaching’s usefulness, a finance owner’s estimate of the combined initiative’s effect, or evidence about additional monetary benefit? Those are different questions. An opinion about usefulness does not automatically divide an invoice reduction among its causes.

An estimate can be part of an explicitly limited evaluation when its elicitation, assumptions and uncertainty are appropriate to the decision. Record who supplied it, what information they had and how alternative explanations were considered. Do not present it as an experimentally identified share.

In this case, Rosa does not accept the suggested quarter-share. The underlying volume counterfactual and component attribution remain unresolved. Tomas leaves the financial return unestablished rather than selecting a fraction that produces an apparently defensible headline.

Use scenarios to expose a requirement, not invent a forecast

For illustration only, hold the $33,000 cost constant and vary a hypothetical attributable monetary benefit:

Values are rounded where shown. The zero-benefit row is an assumption, not the treatment of missing information. These scenarios are not a confidence interval, probability distribution or prediction that benefits lie between the endpoints.

The table answers a useful arithmetic question: on this basis, fifty percent net return requires $49,500 of attributable monetary benefit. It does not show that the initiative can produce it. A future decision still needs prospective evidence and the actual cost of the proposed next arrangement.

Complete the report and the present decision

On 29 January, Rosa and Tomas complete the corrected internal report. Its finished summary reads:

External rework spending was $39,000 in July–September and $12,000 in October–December. Supplier price, internal processing and available capacity changed. The $27,000 difference is a spending comparison, not an established coaching saving.

The combined initiative’s reconciled resource cost through evaluation close is $33,000, including $27,000 of provider and staffing charges and $6,000 of internal time. Coaching-attributed monetary benefits and net ROI are not established.

Retention, meeting capacity, sales and conflict items do not provide additional supported coaching benefits in this review. Reported learning and a clearer working conversation remain described without a financial price.

Ed reads and accepts that final version on 29 January. He withdraws the private fifty-percent slide and confirms it will not be used externally. He defers repeating the full package until a new proposal addresses the supplier-rate basis, relevant comparisons, capacity cost and evaluation design. No renewal, new budget, payment or future outcome is reported.

This is a completed review and a completed decision to defer. It is not a claim that a future evaluation has been funded or that the evidence can now establish causation retrospectively.

Prepare the next financial brief in order

1. Name the decision and object. Coaching alone, combined initiative or proposed new purchase.
2. Define the period and basis. Benefit window, full lifecycle costs, currency and cash versus resource valuation.
3. List each proposed benefit once. Event, source, counterfactual, attribution and monetary conversion.
4. Test alternatives and overlap. Pricing, demand, staffing, other projects and duplicated time or contribution.
5. Complete the cost record. Include preparation, delivery, participation, coordination and evaluation without double counting.
6. Choose the permitted conclusion. An attributed return, a scenario, a descriptive comparison or an unpriced outcome.
7. Approve the exact report and next decision. Keep internal acceptance separate from permission for external marketing.

Choose proportionate evaluation expertise before collecting convenient numbers. The executive-program metrics guide helps retain nonfinancial outcomes rather than forcing every useful change into money.

Organize the coaching evidence in Coachful

Coachful is our #1 recommendation for organizing the coaching contribution to this workflow. We make Coachful, so this is our product recommendation, not an independent ranking or a promise of financial return.

Assign an original review resource through Clients > select client > Worksheets & resources > Assign. A copied private link alone does not grant access. Use goals and tasks for selected actions while keeping the organization’s finance evidence in its authorized system.

Native invoice and payment records can support a relevant activity or cost check, but identify what each record represents. A payable invoice is not proof of cash collected. A CRM payment log records an actual receipt; it is not a full initiative cost statement or proof of a benefit. Avoid duplicating an externally recorded amount to make the evaluation appear complete.

Prepare any sponsor summary under the actual information-sharing agreement. An action’s completion does not establish an employer’s acceptance or a monetary return. Unshared client journal or takeaway pages remain unshared; an enabled Share with coach control is a deliberate client choice to expose that page. Do not treat all personal reflections as financial evidence or sponsor-report material.

Explore Coachful’s features around these records. The platform does not independently construct the counterfactual or turn a before-and-after comparison into attributable ROI.

The demo Payments screen shows invoice amounts and status. It does not supply participant time, other departments’ costs or attributable benefits.

Keep public claims within the evidence

The FTC’s advertising guidance says advertisers need support for express and implied claims before an ad runs. That is a US marketing reference, not a determination about every jurisdiction or this fictional company.

A testimonial, disclaimer or conservative-looking attribution percentage does not repair a claim whose meaning exceeds its evidence. Keep the exact period, basis and limits beside any permitted number. An internal report’s acceptance does not authorize use of a participant’s story or private information in marketing.

Frequently asked questions

Is a 1.5:1 ratio the same as 150 percent net ROI?

No. On the same cost and benefit basis, a 1.5:1 benefit-cost ratio means fifty percent net ROI. The ratio includes recovering the cost; net return measures the amount above it.

Can I multiply time released by salary?

You can state an appropriate resource-value estimate with its basis and limits. It does not prove cash spending fell or profit increased. Check what happened to the time and whether another benefit already includes it.

Does a lower supplier invoice prove coaching savings?

No. Check volume, rates, scope, timing, internal work and other changes. The fictional case’s invoice difference remained descriptive because causal attribution was unresolved.

Does unknown ROI mean minus one hundred percent?

No. That result requires assuming or establishing zero monetary benefit against positive cost. Missing attribution is not an observed zero; report the unresolved return accurately.

Must every executive engagement calculate financial ROI?

No. Choose evaluation that serves the decision. Describe useful learning, behavior or other outcomes with evidence and limits when monetization would require unsupported assumptions.