Group Coaching Curriculum Design Playbook

A useful group coaching curriculum design playbook starts with what participants will be able to do. A sequence called confidence, communication and leadership might sound appealing while leaving the actual learning task unclear. People need to know what they will practise, how feedback reaches them and what counts as progress.

Curriculum design connects those decisions across the whole engagement. It also puts limits on them: a four-session group can support a specific capability without promising lasting workplace change, professional certification or an answer to every member's situation.

Coachful is our #1 recommendation for organizing that delivery workflow: Programs for the engagement, courses with modules and lessons for learning material, and worksheets for deliberately shared practice. This is our editorial recommendation for the workflow described here, not an independently measured market ranking or a guarantee that software improves learning.

Build a curriculum, rather than a collection of topics

A curriculum specifies a destination, a learning route and evidence that informs the next decision. A content library provides material that people can browse. Both can be useful, but a library becomes a curriculum only when you explain how its parts support a particular capability.

Consider a group learning to present a short decision brief. Knowing the names of its sections is introductory knowledge. Comparing options under a budget is application. Responding when the decision maker refuses the preferred option requires another level of practice. A quiz about section names cannot establish all three.

Grant Wiggins and Jay McTighe's Understanding by Design describes backward planning through desired results, evidence and a learning plan. The point is to choose learning experiences in relation to an intended accomplishment, then review what learner performance reveals. It is a curriculum design approach, not proof that a particular group coaching program produces business results. The original 2011 ASCD publication provides that conceptual account; the example below uses our own tasks and records, without reproducing its planning forms.

Carnegie Mellon's Eberly Center likewise explains that objectives, assessments and instruction should support one another. Its alignment guidance supports that design principle. The original curriculum decisions here do not copy the university's activities or assessment table.

Seven decisions that make the learning route usable

1. Define a capability and its limits

Describe an action in a situation: prepare a short brief that separates known facts from assumptions, compares feasible options and asks the authorized person for a decision. Add the conditions, such as a fictional case, a written response route and permission to consult the supplied facts.

Then state what the group will not establish. A successful classroom brief does not show that a participant can manage a contentious board meeting without support. An individual may understand the method while still needing subject expertise or authority. Keep the promised learning task smaller than the wider aspiration.

2. Choose evidence before producing lessons

Ask what you would need to see to make a reasonable judgment about that capability. For a brief, you might inspect factual accuracy, constraints, the decision request and treatment of the answer. These are observable qualities of the work, rather than judgments about confidence or personality.

Collect evidence from each participant. A polished group document can conceal one person doing all the reasoning. Let people contribute together during practice, then obtain a short individual response on a changed case. Explain the criteria in advance and distinguish an incomplete attempt from a refusal to share personal material.

3. Sequence prerequisites, then vary the problem

Order learning by what the next task depends on. Someone cannot compare feasible options reliably before identifying the constraints. Someone cannot report an agreement accurately before distinguishing a proposal from the other person's actual answer.

Early practice can isolate one difficulty. Later practice should combine the parts and introduce a meaningful change, such as a price increase or a refusal. Repeating the same answer with new names may demonstrate memory without showing adaptation. Conversely, changing everything at once makes it harder to understand the cause of a difficulty.

4. Make the teaching and coaching roles explicit

Teaching introduces a concept, models reasoning and checks understanding. Coaching invites people to choose an application, examine their interpretation and decide what they want to try. A group may need both. Tell participants when you are offering an instructional example and when you are asking an open coaching question.

For example, you can teach how to calculate a remaining budget, then ask which uncertainty a participant wants to explore. You should not treat an incorrect calculation as merely a different perspective. Equally, your preferred career decision is not an answer the client must discover to pass a coaching exercise.

5. Design participation and access into the task

Offer the important facts in a readable document before a session. Describe relevant visuals aloud and allow processing time. The W3C's accessible-events guidance addresses advance material, adaptable formats, access requirements and activities that may exclude people. Check the actual meeting environment and requested accommodations rather than assuming a platform choice settles access.

A written response can assess reasoning without requiring an impromptu speech. If speaking itself is the learning objective, agree a suitable supported speaking route. Do not quietly replace the objective or classify an access need as low motivation. Personal disclosure, public review and participation in a particular role-play also need clear choices.

6. Return feedback in time to use it

Eberly's assessment guidance distinguishes feedback that informs ongoing learning from evidence used to judge an endpoint. A useful design provides an opportunity to act on feedback before the final task. A comment delivered after the engagement ends cannot improve an earlier attempt.

Specify the return: who reads the work, what they comment on, how it reaches the participant and when a revision is possible. An internal coach note is not returned feedback. Also budget the participant's reading and revision time. Feedback that requires an unannounced hour of work changes the service people agreed to.

7. Revise the curriculum without rewriting the result

When several people make the same error, examine the instruction and task wording as well as individual understanding. Preserve the first attempt. A later corrected answer is new evidence; it does not mean the original teaching worked as intended.

Version the change, identify who needs it and tell current participants what has changed. Replacing a master plan is different from delivering a correction to an active group. Check whether the revision alters access, timing, promised feedback or workload before applying it.

A complete fictional four-session curriculum

Leila coaches four community project coordinators, Ren, Owen, Bea and Jules. Their fictional May 2042 group practises short decision briefs. They already understand basic project costs. They want to explain a recommendation without confusing their own preference with authorization.

Before joining, all four accept four Tuesday sessions on 6, 13, 20 and 27 May, 14:00–15:00 Europe/London. Each session includes a five-minute break. Each participant also accepts four ten-minute preparation blocks, one five-minute curriculum update read and a final fifteen-minute written feedback exchange. Those are the complete required activities. There is no compulsory workplace experiment or unpaid peer review outside the sessions.

The agreement permits fictional cases throughout, oral or written presentation, and a private final submission to Leila. No session recording is offered. Each receives one final criterion-based comment in their private client conversation by 28 May at 16:00. They can use their fifteen-minute exchange allowance to read it and send one clarification by 29 May at noon; Leila will answer by 30 May at noon. Learning material remains available through the Program's ongoing Keep access content setting after completion. That does not promise future live coaching or replies.

The individual endpoint is an accurate brief on changed fictional facts, followed by an accurate account of the decision maker's response. The four criteria are: distinguish fact from assumption; compare options against every stated constraint; identify the actual decision owner and request; and reflect the answer without inventing acceptance. This is Leila's original learning rubric, not a validated psychometric assessment.

Teach one complete unit, then inspect what happened

The first case concerns a fictional repair café. Its organizer controls the booking. The venue must be step-free, accommodate twenty-four participants and keep the total within £180. Materials cost £50. Hall A costs £120, is step-free and holds thirty. Hall B costs £80 and holds thirty, but has stairs and no accessible alternative entrance.

In session 1, Ren correctly identifies Hall A's £170 total and £10 remainder. Hall B's £130 total is lower, but it fails the access condition. Leila asks whether a cheaper option that excludes an intended participant meets the brief. Ren says it does not. Arithmetic and feasibility are separate checks.

Session 2 uses this sixty-minute plan: five minutes reconnecting to the purpose, ten minutes demonstrating an options comparison, ten minutes individual drafting, five minutes break, twenty minutes role-play and feedback, and ten minutes reviewing what to carry forward. The twenty-minute role-play segment gives each participant a five-minute turn, including their explanation, Leila's response and immediate feedback. Everyone can read the same facts directly.

Ren recommends Hall A and asks the organizer to approve it. Leila, in role, says, “I need to check the date with the building manager before I can approve.” Ren then writes “Hall A approved, book Friday.” Two other participants also treat a conditional response as approval. Jules correctly leaves approval pending.

The result is three inaccurate response records out of four, despite all four producing a feasible cost comparison. Leila's example had emphasized making a clear recommendation but had not sufficiently demonstrated conditional answers. She records both findings. A satisfaction question would not have revealed this particular gap.

Make a specific curriculum correction

Leila uses thirty minutes between sessions to rewrite the teaching example and the next practice, then ten minutes to compose and send a single update to all four through their existing client conversations. It reads:

Our venue comparison was workable, but my demonstration did not give enough attention to conditional answers. On 20 May we will practise recording what the decision owner actually says. Approval will stay pending when a check is outstanding. There is no added homework or session time. Please read the revised example in your next five-minute update block.

All four read that update before session 3 in their agreed five-minute blocks. No separate acknowledgment is requested or counted. Leila changes the relevant lesson's authored example and the session plan; she does not claim that editing a Program template automatically updates an existing cohort. The current cohort's actual resource assignment and opening are checked separately.

Session 3 gives ten minutes to the corrected explanation, ten to individual rewriting, five to a break, twenty to another four-person role-play rotation and fifteen to reflection. That totals sixty minutes. The participants use the new distinction during the session rather than being sent away to obtain free labor from colleagues.

The changed fictional facts make Hall A £140, so its total is now £190. Hall B still fails access. A third room is step-free and costs £110, but holds only twenty people. Its £160 total meets the budget while its capacity fails the original twenty-four-person requirement. The correct next step cannot be “choose the affordable room” without the owner's decision about the smaller event.

Ren's first revision asks for another £10. Leila, as organizer, refuses extra funding but says a twenty-person event could be acceptable if the proposed invitation limit is explicit. Ren must now revise the scope, not rename a failed budget check as success. In session 4, the final exercise uses these changed facts with the organizer's authority and refusal stated in writing.

Show the final output and the actual answer

Ren's complete final brief is:

Hall A now totals £190 including materials, exceeding our £180 limit. Hall B costs £130 but fails step-free access. Room C costs £160 and is step-free, with capacity for twenty rather than twenty-four. You have declined extra funding. I recommend reducing invitations to twenty and using Room C, leaving £20. Will you authorize that reduced capacity and the booking? Until you answer, both remain proposals.

Leila, still acting as the fictional organizer, answers: “I authorize twenty invitations and Room C within the £160 total. Please record the reduced capacity. This is approval of the exercise proposal, not a real venue booking.” Ren records that answer, identifies Leila's assigned role and retains the £20 remainder. There is no assertion that a real organizer, venue or funder agreed.

Session 4 allocates five minutes to instructions, fifteen to individual writing, five to a break, twenty to four five-minute response turns, ten to each person's next learning choice and five to explaining the closing feedback route. During the individual writing period Leila supervises; her later private review is separately counted.

All four attend all sessions. On the final exercise, Ren, Owen and Jules meet all four criteria. Bea's options comparison and request are accurate, but her response record still describes a conditional answer as approval. Her criterion remains unmet. Leila does not average it away because the group enjoyed the session.

On 28 May Leila spends ten minutes per participant reviewing their submitted record and five composing and sending their individual comment. Ren's actual comment is: “Your £190, £130 and £160 totals are accurate. You kept Room C's reduced capacity conditional until the organizer authorized it. Your final record matches that answer. This demonstrates the classroom task; we have not observed you doing it in your own project.”

Each participant spends five minutes reading their comment. Ren, Owen and Jules use another five to write one clarification and five to read Leila's answer. Ren asks whether the same brief would establish authority in his own association. Leila answers that the simulation does not establish his association's approval route; he would need to identify its actual decision owner before proposing any booking. Bea uses her remaining ten minutes to revise her response record privately and chooses not to send it. Her final assessed result remains unchanged.

Leila uses five minutes for each of the three clarification threads, including reading the question, composing and sending the answer. Ren's answer is sent at 10:00 on 30 May and read by 10:05. The other two are also sent and read before noon. No final acknowledgment is requested. Bea has received the promised comment without being required to disclose her private revision.

Account for the whole learning workload

A curriculum's workload includes design, administration and receiving feedback. The following is the completed fictional account through the final reads on 30 May. Earlier unrelated business work and future workplace application are outside this finite curriculum project.

Leila uses 685 of 720 minutes, leaving 35. Participants together use 1,240 of 1,300, leaving 60. The whole account is 1,925 of 2,020 person-minutes, with 95 unallocated. The live sessions occupy four hours on the calendar but consume 1,200 person-minutes across five people. Neither number substitutes for the other.

In her final twenty-minute review, Leila reads the completed records and checks the promised returns. She records a useful curriculum correction and a three-of-four final classroom result. That small, coached group supplies no causal comparison, durable follow-up or evidence of workplace transfer. Bea's remaining difficulty becomes a reason to improve instruction and offer a separate choice, not proof she needs to purchase more coaching.

Organize the curriculum in Coachful

Start with the agreed service and audience. Program structure uses Modules, which can span weeks; a module is not automatically a month. A course provides its own modules and lessons. In Advanced, the course editor is under Create > Courses; Simple opens Courses through Today > Worksheets & resources. Build the teaching route from the written plan rather than letting a template determine your learning promise.

Check the named course's inclusion and the actual cohort's resource access. Template changes and delivery sync are separate. For an existing cohort, Content > select the cohort > Tools > Sync from Template can apply template changes, so review the recipient and replacement scope before using it. A newly written lesson or copied link alone does not establish access.

Course start, module release and lesson release are distinct controls. Modules and lessons can open from the start, after configured days or on a date; completion-dependent module release has its own conditions. Program timing does not replace course drip. In Leila's fictional setup, the relevant teaching material is open from the start, and there is no completion-dependent lock preventing preparation. A future dated version would require a separate check against every session date.

Drip release does not itself send a lesson-unlock email. Communicate required preparation and revisions through an agreed channel, and check any separately scheduled messages when changing dates. A coach preview can bypass lesson locks, so use an authorized client access check to establish what the intended participant can open.

For deliberately shared exercises, use a worksheet or form and make the audience clear. Coach review is private to the authorized coach team; saving it does not deliver a comment to the client. Leila returns the promised comments through each participant's private client conversation. Private course takeaways are personal reflection, not an assessed submission the coach can inspect. General journal pages have their own explicit client-sharing possibilities; do not equate all journals with shared homework.

The Coachful feature overview explains the broader workspace. Keep curriculum decisions distinct from group coaching pricing, and use the group participation guide when a particular interaction needs a separate boundary or repair.

Put the design into practice

1. Write the end task. Specify the participant, situation, capability and limits.
2. Define visible evidence. Choose a few criteria and explain them before practice.
3. Map dependencies. Place necessary knowledge before integrated decisions.
4. Allocate the experience. Include teaching, individual thinking, practice, breaks and coaching choices.
5. Agree the receiving work. State preparation, feedback, revision and communication allowances.
6. Test and revise deliberately. Preserve the original result, change the instruction and deliver the update to the actual group.
7. Close with bounded evidence. Return promised feedback and distinguish classroom performance from unobserved application.

Troubleshoot the curriculum before adding content

People finish lessons but struggle with practice: inspect whether the lessons teach the reasoning the task requires. Another video may add work without addressing the gap.

One person carries the group: separate collaborative practice from an individual demonstration. Do not infer competence from a group document alone.

Feedback arrives too late: shorten the task, narrow the criteria or reduce the roster before promising a faster return. Count sending and participant reading as work.

A revision creates an access problem: check the actual resource, audience, release date and instance. Do not make private work public to repair a missing entrance.

The final result is mixed: report it plainly. Offer choices about further learning without converting an unmet criterion into a sales diagnosis.

Frequently asked questions

How many modules should a group curriculum have?

Use the fewest that support the learning dependencies and available practice. Four modules suit the fictional task because each adds a distinct demand. A different capability may need another structure; module count is not a quality measure.

Should every session include coaching?

Include coaching where participants need space to examine their own choices. Explain instructional segments honestly. Do not disguise a factual correction as an open question or force a personal disclosure to satisfy a lesson plan.

Does a completed course prove a participant learned?

No. Completion records activity under the course's rules. Inspect relevant work and responses, then state the limits of that evidence. Even a successful final simulation does not establish lasting workplace performance.

Can I change the curriculum during a cohort?

Yes, when the change is appropriate and properly communicated. Review current assignments, access and workload, preserve earlier results and deliver the changed instruction. Master template edits are not proof that an active cohort received them.

What should I do when someone declines public feedback?

Respect the agreed sharing choice and use a private route when it meets the learning purpose. If public presentation is itself essential, discuss a suitable alternative or service-fit decision explicitly. Do not publish private work to manufacture participation.