Leadership training programme template

Why leadership development fails as a single event

A two-day leadership offsite is easy to budget and easy to defend. It also asks people to change how they behave under pressure on the strength of one exposure, with no rehearsal afterwards and nothing owed once it ends. The content is rarely the problem; the delivery unit is. One event is the wrong shape for a change that has to survive contact with a real team on a real Tuesday.

A programme is not a longer event. It is a schedule of contact with obligations in between — something each participant attempts at work, reports back on, and is questioned about. Three short returns to delegation across four weeks, each following a real attempt, move behaviour further than a three-hour masterclass nobody revisits. Spaced practice does the work that session length cannot.

The analysis stage of ADDIE is what makes that schedule specific. "Our managers need leadership training" is a symptom, not an analysis. Name the missing behaviour instead — managers who never decline work, managers whose reports first hear a criticism at the annual review — and both the modules and the measurement follow from it.

Writing the 70-20-10 split into the plan

The 70-20-10 model — roughly seventy per cent of development from doing the job, twenty from other people, ten from formal instruction — describes what managers report about their own careers. It is a design heuristic, not a measured law, and it is most useful as a prompt: if the workshop is the ten, what exactly are the seventy and the twenty?

The seventy has to appear as assignments with owners and dates, or it is just learning on the job, which was already happening. Concrete looks like running the next retrospective without the senior engineer in the room, taking over the two hardest one-to-ones on the team, or owning a hiring loop end to end including the rejection call. Uncomfortable, reversible, small enough to fail safely.

The twenty is the part most programmes gesture at and never staff. It needs named people and a cadence: a fortnightly peer partner from another function, a skip-level with the manager once removed, a mentor who has agreed in advance to be asked awkward questions.

• Seventy: one dated stretch assignment per participant, with a named observer who will comment on it.
• Twenty: a peer partner, a skip-level, and a mentor who has actually agreed to the role.
• Ten: the sessions — the smallest part of the programme, and the only part most plans describe.

What actually breaks for a first-time manager

Most first-time-manager programmes teach leadership in the abstract, when the failure modes are narrow and predictable. The person was promoted for being excellent at the work, and the job now is to stop doing the work. Delegation is the first casualty: they keep what they are best at, because handing it over means watching it done worse and still carrying the result.

Feedback fails in two directions — praise too vague to repeat, and criticism saved up until it is too large to hear. The peer-to-boss transition is its own problem, and the first real test is the first refusal to someone who was a colleague last month. Neither is solved by being told to set boundaries; both are solved by rehearsing the actual sentences out loud, in front of people who will say when you hedged.

• Delegation: hand over the task you are proudest of, and pre-agree the standard you will accept from a first attempt.
• Feedback: rehearse one specific piece of corrective feedback the participant is currently avoiding.
• Authority without friendship: role-play the first refusal to a former peer, then do it for real that week.

Cohort design and peer learning

Leadership is one of the few subjects where the other participants are part of the content. A new manager needs evidence that what they are struggling with is normal, and that lands when a peer admits it rather than when a facilitator asserts it. Eight to twelve people works: below that a single dominant voice sets the tone, above it people stop speaking. Keep the cohort fixed so trust compounds, and keep participants' own line managers out of the room.

Action learning sets are the highest-yield format: each participant brings a live problem, gets a fixed slot, and the group is allowed only questions — no advice — for the first half of it. Put it on the standing agenda, or it becomes the first thing cut when a session overruns.

Objectives you can actually observe

Leadership objectives are usually written at the wrong level of Bloom's taxonomy. "Understand the principles of delegation" sits at comprehension, and there is no honest way to assess it: you can ask someone to describe delegation and learn nothing about whether they delegate. Move the verb up. "Delegate a task of your own to a named person and evaluate the outcome against the standard you set" is assessable, and it teaches the same thing.

SMART turns that into a plan rather than an aspiration, and the criteria quietly dropped are measurable and time-bound. Attach a number and a date even when the number feels crude: three documented feedback conversations by the end of week four is a blunt measure, and still far better than "improved feedback skills".

At session level, Gagné's nine events remain the most practical checklist for a ninety-minute block: open with a real incident, state the objective, connect it to what the group already does, present the model briefly, guide a first attempt, give feedback in the room, then name what each person will do at work before the next session. That closing step is the one cut for time, and the only one that reaches the workplace.

Measure behaviour, not satisfaction

Kirkpatrick's four levels — reaction, learning, behaviour, results — are quoted constantly and applied past the first rarely. Level one is a happy sheet: cheap, immediate, and close to unrelated to whether anything changed. It survives because it produces a number the same afternoon, and because nobody is embarrassed by an eight out of ten.

Level three is the one that matters for leadership, and it costs something. Decide before the programme starts on three to five behaviours you expect to see afterwards, then ask the people on the receiving end: a short pulse to direct reports beforehand, and the same questions 60 to 90 days after the final session. "My manager told me something I was doing badly within a week of it happening" has an answer. "My manager is a good leader" does not.

• Level 1: keep it, but stop presenting it as evidence of impact.
• Level 2: for leadership, test application in the room rather than recall of models.
• Level 3: three to five named behaviours, asked of direct reports, before and 60 to 90 days after.
• Level 4: tie it to one result you already track — attrition on those teams, internal promotion rate — and report it as correlation, not proof.

Common questions

How long should a first-time manager programme run?

Four to eight weeks of spaced contact suits most first cohorts: long enough for participants to attempt something at work between sessions, short enough to hold their attention and their sponsor's patience. The generator on this page defaults to four weeks in a course format, and you can change the duration and format before you generate.

What should a leadership training programme template include?

At minimum: objectives written at an observable level, three to five modules each carrying lessons, activities and exercises, the between-session assignments that hold the on-the-job learning, and the assessment methods you will use. The plans generated here cover the objectives, the modules with their activities and exercises, and the assessment methods. The named stretch assignments and mentors are the part only you can supply.

Does it write the on-the-job assignments for the 70-20-10 split?

It produces activities and exercises at module level, and several of those work directly as between-session assignments. It does not know your teams, so it cannot name which hiring loop or which one-to-one a particular participant should take over. Premium editing lets you write those in against the right module.

Can I edit the plan after it is generated?

Editing is a Premium feature at $19 a month. On the free tier you can generate and save one plan, copy the whole thing to your clipboard, and print it, so the text can always be moved into a document of your own.

Can I export the programme to share with stakeholders?

Premium includes PDF and Word (.docx) export. Copying to the clipboard and printing are available on the free tier. There is no SCORM or LMS package: this produces a plan document for people to read and work from, not packaged courseware.

Do I need an account to try it?

No. You can generate a real plan from this page without signing up, limited to three a day. The preview shows the full set of objectives and the complete first module, with the remaining modules listed as titles. A free account shows every module in full and saves one plan to your dashboard.

How do I show the programme worked?

Decide on three to five behaviours you expect to change before you run anything, then ask participants' direct reports about those specific behaviours before the programme and again 60 to 90 days after it ends. That is Kirkpatrick level three, the level most programmes skip. Pair it with one result you already track, such as attrition on those teams or internal promotion rate, and report it as correlation rather than proof.