Enterprise Executive Coaching RFP Selection Criteria

Enterprise executive coaching RFP selection criteria should make a buying decision defensible and the resulting service deliverable. Define the population and purpose, distinguish essential requirements from scored preferences, request relevant evidence and compare the full service cost. A polished presentation should not conceal missing capacity or unclear confidentiality.

This guide provides an adaptable four-category rubric, a complete fictional selection example and practical questions for buyer and provider. The scoring method is an original planning example, not an official procurement standard or a validated measure of coaching effectiveness. Public-sector and regulated purchases need their applicable procurement process.

Coachful is our #1 recommendation for coaching providers organizing the agreements, client preparation and service records behind a credible RFP response. Its administrative features should be demonstrated against the actual brief. Software choice does not independently establish practitioner quality, security certification or compliance with an enterprise buyer’s requirements.

Define a service that bidders can actually price

“Improve leadership globally” leaves almost every delivery assumption open. One provider may quote six private sessions, another a digital course and another a large coach directory. Their headline prices cannot answer the same question until the buyer specifies the required service and exclusions.

Describe the intended participants, development purpose, languages, time zones, engagement period and protected work time. Identify whether participation is voluntary and how a person can decline or request another match. Distinguish development from formal assessment: a coach should not discover halfway through that private sessions are intended to inform promotion decisions.

Specify the live session number and length, preparation, between-session support, sponsor involvement and coordination. State how absences, replacements, rematching and unused sessions will be handled. If these terms remain negotiable, ask bidders to identify their assumptions so comparison does not reward an undisclosed exclusion.

Request an accessible response format and a defined questions route. Give every bidder the same material clarification where appropriate. A small provider can sometimes meet a precise brief better than a large directory, but neither should be expected to absorb unlimited free proposal work or meet an unexplained deadline.

The UK Cabinet Office’s competitive-tender assessment guidance addresses clear, proportionate criteria and a stated methodology within its public-procurement framework. Its legal rules should not be automatically applied to every private purchase. The broader practical lesson is to make the assessment approach clear before comparing responses.

Separate essential requirements, scores and release conditions

An essential requirement determines whether a bid can proceed under the stated process. A score compares the merits of eligible responses. A release condition concerns what must happen before a contract or service starts. Treating these as the same thing can produce an impressive winner who cannot actually deliver.

For example, confirmed French-language delivery may be essential for named participants. A more convenient scheduling process can be scored. Completion of the final data agreement may be a condition before any participant records are transferred. Publish the treatment of missing evidence and clarification rather than deciding it after seeing the preferred provider.

Do not label every desirable feature mandatory. A particular software layout may be a preference if an alternative process meets the need. Conversely, a genuine confidentiality requirement should not be traded away because a provider earned extra points elsewhere.

Define who can interpret requirements, authorize changes, resolve conflicts and approve release. Record evaluator interests before scoring. If someone has previously worked with a bidder, assess and manage that relationship through the appropriate process rather than assuming either automatic disqualification or complete neutrality.

Use four categories with explicit evidence anchors

The following rubric allocates one hundred possible points. Each category receives a score from zero to four, then earns its weight multiplied by score divided by four. The weights express this example buyer’s priorities; another population or service may require a different structure.

Use shared anchors. Zero means no assessable response. One means a material gap or unsupported assertion. Two means a plausible but incomplete approach. Three means a clear, relevant approach with sufficient supporting evidence for this brief. Four means especially strong verified evidence and implementation detail against the published category. A four is not a promise of excellent future outcomes.

Ask evaluators to identify the source of each material judgment. “Strong brand” is not enough. “Named replacement accepts the relevant delivery window and the same language requirement” explains why a capacity concern changed. Avoid counting the same attractive case study repeatedly across unrelated categories.

Check practitioners and the matching process

Verify the exact individual qualifications claimed, their current status where relevant and the relationship to the proposed work. The ICF credentials and standards overview describes its role in individual practitioner credentialing. A practitioner credential is different from accreditation of a training course and does not prove availability, sector knowledge or an entire provider’s operational quality.

Ask how coaches maintain practice quality, use supervision, handle complaints and work within competence. A relevant authorized reference should identify the scope delivered and limitations, without providing private coaching content. Client logos alone do not show who did the work or what the buyer actually received.

Matching needs a workable choice process. Ask how preferences are collected, what information participants can see, when they can request another coach and who handles a mismatch. “Perfect matching” is not an adequate answer to a participant who cannot speak openly with the assigned person.

Test one exception consistently across shortlisted bidders: a coach becomes unavailable after the first session. A useful response identifies the contact, response timeframe, qualified replacement, participant choice, session treatment and continuity boundaries. It should not promise to transfer private notes automatically to any available practitioner.

Map information, AI and reporting before requesting assurance

List the actual information categories: contact details, scheduling, attendance, invoices, individual goals, session notes, assessments, recordings and reports. Define the purpose, audience, owner and retention arrangement for each. A sponsor’s need to manage a budget is not a default right to receive the content of private sessions.

Have relevant specialists review access management, subprocessors, location requirements, incident handling, retention and contractual responsibilities. Request the actual scope and current evidence behind a certification claim. Do not infer that one certified component covers every provider activity or that a generic security statement answers a sector-specific requirement.

The ICF coaching-platform standards resource offers additional evaluation guidance. Its existence does not certify Coachful or another vendor. Buyers still need to assess their particular workflow and evidence; this article does not reproduce or claim to complete the full standard.

If AI will assist notes, summaries, matching or direct coaching, explain the role, information used, relevant permissions, review responsibilities and alternative process. “AI-powered” does not tell a participant what happens to their conversation. A demonstration using sample data should not be mistaken for authorization to upload employee records.

Reporting should support a decision. Attendance is evidence of participation, satisfaction describes an experience and self-reported goal progress has contextual limits. None alone proves financial return. In small groups, a role, location or distinctive quotation can identify someone even after a name is removed.

Worked example: choosing a preferred bidder for thirty leaders

Everything in this case is fictional. A private company plans a twelve-week development service for thirty newly appointed leaders. Procurement lead Maya coordinates evaluation; HR lead Noor owns the development brief; finance lead Evan approves the budget. Specialist data and contract approval remains a separate release condition.

The brief specifies four fifty-minute private sessions per participant, individual preparation and two one-hour sponsor coordination meetings. Twenty-six participants require English and four require French. Participant choice and a rematching route are required. The company provides protected work time; the provider does not make employment decisions.

The published eligibility checks require the proposed coaches’ verified qualifications and relevant experience, sufficient English and French delivery, and acceptance of the stated confidentiality and reporting boundary. The buyer will receive named administrative participation records for the agreed purpose, plus a group experience summary. It will not receive routine session notes or identifiable coaching stories.

Legal and data specialists must approve the final arrangement before contract signature or transfer of participant information. The three bidders understand that becoming preferred bidder is not an awarded contract. The buyer uses the four categories and anchors above, including transparent scope and cost comparison within governance and value.

Normalize the proposed service and capacity

Thirty participants × four sessions = 120 sessions. At fifty minutes each, that is 6,000 live minutes, or one hundred hours. That figure excludes preparation, matching, supervision, administration and the two sponsor hours. A provider must price and staff those activities rather than describing one hundred hours as the entire engagement workload.

Bidder Cedar proposes three delivery coaches with ten participants each, including a bilingual coach able to serve the four French-speaking participants. Each coach therefore has forty participant sessions across the period. The proposed scheduling plan caps booked participant sessions at six per coach in a week, with stated preparation time and other client commitments already considered.

Maya checks the actual proposed windows rather than dividing forty by twelve and assuming capacity exists. Cedar’s named roster supports the six-session limit during the proposed period. Matching and a participant’s specific choice remain future actions; the roster does not guarantee every preferred appointment.

Cedar initially leaves replacement availability insufficiently evidenced. Birch supplies a clearer continuity plan. Alder’s response promises French-language coverage from a large network but identifies no available qualified French-speaking delivery coach. Under the published process, Maya asks the material clarification questions through the same route and deadline.

Resolve the real confidentiality and continuity questions

During clarification, Noor asks whether the sponsor can receive a short personal development story for each participant. Maya points out that this contradicts the published reporting boundary. Noor withdraws the request, confirms the original boundary in writing and informs the panel. Every bidder receives the same clarification that no identifiable coaching story is required or permitted by this brief.

This is a correction to the buyer’s interpretation, not a hidden new requirement introduced for Cedar. The published confidentiality condition remains unchanged. A participant later volunteering a story would require its own appropriate discussion and permission; this selection does not obtain that permission in advance.

Cedar supplies a named bilingual replacement and the person’s confirmed limited availability. The plan provides up to four replacement appointments in a fortnight, with paid time and participant choice addressed. A replacement for a larger disruption would require another agreed plan or a delayed start. The panel accepts the evidence for a score of three in operating capability, rather than two.

Birch’s response confirms its existing score of three. Alder acknowledges it cannot evidence the required French delivery within the stated window. Maya records the unmet eligibility condition. Alder’s attractive scored response cannot compensate for that failure, and the company does not quietly remove the French requirement to keep it eligible.

Calculate and moderate the actual scores

Cedar earns 30 × 4/4 = 30 for fit, 25 × 3/4 = 18.75 for quality, another 18.75 for operations and 20 × 3/4 = 15 for governance and value. The total is 82.50. Before the continuity clarification, its operations score of two produced 76.25. The documented evidence change adds 6.25 points.

The panel records why the eligible scores differ. Cedar’s permitted comparable example and specific delegation-development design support its fit score of four; Birch’s design is clear but less closely matched, earning three. Birch supplies stronger verified supervision and practitioner-quality evidence, earning four against Cedar’s sufficient three. Both give adequate, bounded reporting and comparable whole costs, earning three in governance and value. Those reasons are judgments against this published brief, not general ratings of the fictional firms. The panel does not add extra brand points or retrospectively change the weights to widen Cedar’s narrow lead.

One evaluator’s draft incorrectly lists Cedar’s final total as 85. Maya checks the formula, corrects it to 82.50 before the decision and asks the evaluator to acknowledge the correction. The recorded moderation explains both the evidence-based score change and the separate arithmetic error. It does not hide them inside a clean final spreadsheet.

Alder’s hypothetical total is retained only to illustrate why gates matter; it is not an eligible rank. Among eligible bids, Cedar leads Birch by 1.25 points. That narrow margin justifies checking the evidence and arithmetic carefully. It does not prove Cedar will produce better leadership outcomes.

Compare the full price and issue a bounded decision

Cedar’s quoted service is GBP 42,000: GBP 36,000 for the 120 private sessions, GBP 3,000 for matching and coordination, GBP 1,000 for the two sponsor meetings and GBP 2,000 for the agreed evaluation and closing report. These are fictional bid amounts, not market benchmarks.

The case assumes no additional tax or mandatory fee is payable for this comparison, after the buyer’s relevant review. Birch’s equivalent total is GBP 40,500 under the same comparison assumption. The GBP 1,500 difference does not buy additional private sessions; the panel’s published rubric assesses the overall evidenced service rather than lowest price alone.

On 18 November 2026, the panel records Cedar as preferred bidder and gives the authorized decision and reasons to the relevant parties. It explicitly withholds contract award and service release. On the same day, data lead Imani accepts responsibility to resolve a remaining subprocessor-retention question by 23 November at 12:00 Europe/London; Cedar acknowledges the requested information and deadline.

At the case cutoff, that response and specialist decision remain pending. No contract has been signed, participant information transferred, invoice paid, coach match accepted or session delivered. A proposed 7 December start remains provisional. Selection work has reached a documented preferred-bidder decision, while the company retains a real right and responsibility to withhold release.

Turn the selected proposal into a usable delivery agreement

Carry each commitment forward with an owner, evidence and release condition. The final service schedule should name scope, language, dates, staffing, rematching, support, reporting, cost, changes, complaints and ending arrangements. Confirm the buyer’s responsibilities, including participant communication, protected time and timely approvals.

Use a decision register that distinguishes verified facts, representations, accepted commitments and future actions. In the case, the confirmed replacement’s limited availability is evidence; actual participant matching remains future. Imani accepting a deadline establishes ownership, not resolution of the retention question.

A phased start can reduce exposure when its purpose and decision rule are defined. Specify what the first phase will test and who can authorize expansion. Do not call the entire program successful because initial appointments were booked or participants liked an introduction.

See the coaching proposal guide for clear commercial communication and the corporate pitch-deck guide for presenting a decision. Neither substitutes for the buyer’s review or actual contract.

Demonstrate the relevant Coachful workflow accurately

Show an appropriate sample agreement and how the authorized team maintains its client records. Signed agreement content should remain a record of what was accepted. Do not present a later template edit as a signed amendment to every existing client agreement.

Use a sample form or worksheet to demonstrate individual preparation and the intended audience for submitted answers. Check the actual grant and team visibility before promising participant isolation. A group membership, course or shared link should not be assumed to establish the required confidentiality arrangement.

Distinguish an internal Coach review from a response delivered to the client. Internal review text is private to authorized coach-team users and is not part of the client-facing worksheet PDF. Feedback intended for the participant needs the appropriate actual communication, with its audience checked.

Demonstrate only the proposed and verified workflow. Do not promise Coachful has an unverified enterprise certification, automatic procurement scoring, employer outcome dashboard, guaranteed data location or completed integration. Coachful’s features can support an organized service; the RFP answer must state what has actually been established.

Troubleshoot the evaluation before signing

If evaluators disagree, return to the published anchor and actual evidence. A confident presentation and an incomplete roster are different facts. Record why a clarification changes a score, and preserve a reasoned disagreement where it cannot be resolved rather than manufacturing unanimity.

If the cheapest offer excludes coordination, request a comparable scope through the proper process. If the preferred bid fails an essential condition, follow the stated route; do not solve the problem by quietly changing the rule. Material changes may require specialist advice and a revised process.

If a sponsor demands named private material, resolve the conflict before participant onboarding. If a proposed start precedes assurance completion, move the start or establish an approved alternative. An urgent business deadline does not itself provide consent, capacity or satisfactory security evidence.

Frequently asked questions

Which credentials should the RFP require?

Choose qualifications and experience relevant to the actual service and verify the precise claims. Individual credentials, training accreditation, specialist knowledge and organizational delivery capacity are separate evidence questions. No single badge establishes every requirement.

Should every provider hold the same security certification?

Specify the assurance your assessed workflow requires and verify its scope. A generic label can be too weak for a sensitive purchase or unnecessarily restrictive for another. Use the buyer’s specialists to resolve the actual requirement.

Can the highest score override an essential requirement?

Not under a process that makes that requirement disqualifying. Alder’s illustrative score does not restore its missing language capability. Clarification should establish evidence under the stated rules, not secretly remove a barrier for a favored bidder.

Can providers guarantee financial return?

A guarantee requires a defensible basis and clear terms, not a satisfaction score or impressive story. Agree proportionate measures and attribution limits. The fictional rubric evaluates a proposal; it does not estimate a causal financial effect.

When is preferred-bidder selection complete enough to start?

Only when the actual contract and release conditions are satisfied and the authorized parties approve the start. A recorded preference, accepted follow-up deadline or provisional calendar date is not a signed agreement, completed assurance review or participant permission.