Here is a scene I have watched play out for close to twenty years.
An organisation books a two-day leadership workshop. The room is good. The catering is fine. People laugh, they do the exercises, and they fill in the workbook. On day two, someone says the thing everyone says: “This has been really valuable.”
Then they go back to work.
By Friday, the workbook is in a drawer. By the end of the month, the language has evaporated. Six months later, when one of those leaders is sitting across from a team member they need to have a hard conversation with, not one thing from that room is available to them.
That is not a failure of the content. It is a failure of design.
And it is expensive. Most organisations are not underspending on leadership. They are spending in the wrong way.
Why leadership development stops working the moment people leave the room
Three reasons, in my experience, and they compound.
One: the timing is wrong. Leadership is not learned in a room. It is learned at 4:45pm on a Tuesday when someone has just walked into your office in tears, or when your best performer has gone quiet in every meeting for a fortnight, or when you have to tell someone their role is changing. That is the teachable moment. A workshop three months earlier or two months later is not in the room with them.
Two: It is bought by one person at a time. Send a single leader on a program, and they come back with a new language that nobody around them speaks. They try it once, it lands oddly, and they quietly drop it. Development given to an individual inside a team that has not moved is almost always absorbed back into the culture rather than changing it.
Three: There is no shared standard. Ask ten leaders in the same organisation what good feedback looks like, and you will get ten answers. Without a common model, every conversation gets renegotiated from scratch, and consistency — the thing your people actually experience as fairness — never arrives.
Seven things that do work
None of these require a big budget. Most of them require a decision.
1. Develop the team, not the individual
If you can only afford to move one thing, move the group. Ten leaders holding the same model will reinforce each other in corridors, in one-on-ones, in the language they use about their people. One leader holding it alone will lose it. The unit of change is the team, not the person.
2. Choose one behaviour per quarter, not twelve
The most common mistake I see in leadership programs is scope. Twelve competencies, four days, a beautiful framework, zero behaviour change. Pick one. “This quarter, every leader in this team runs a real one-on-one every fortnight.” That is it. Master it, then add the next one.
3. Replace the two-day event with a fortnightly rhythm
Fifteen focused minutes every two weeks will beat two days once a year, every single time. Learning sticks through repetition and retrieval, not immersion. If your development calendar has one big rock on it and nothing else, you are buying a memory, not a capability.
4. Ground your hard conversations in what a camera would have recorded
This one is practical enough to use today. Most feedback fails because it leads with the story rather than the evidence: “You’ve been disengaged lately.” That is a conclusion, and the other person will fight it, because it is arguable.
Instead, lead with Direct Observational Data — only what a camera or a microphone would have captured. No adjectives. No motive. No pattern language.
“In the last three project meetings, you didn’t speak unless I asked you a direct question. That’s different from how you were in April, and I wanted to check in.”
That is not arguable, so it does not trigger defence. It opens a conversation instead of starting a negotiation about whose version of reality is correct. It is the core of what I teach as the Courageous Conversation DOD™ approach, and it is the single highest-leverage habit I know of for a leadership team to adopt together.
5. Measure behaviour, not satisfaction
A happy sheet at the end of a session tells you whether people enjoyed the day. It tells you nothing about whether anything changed. Pick two or three observable indicators before you start — frequency of one-on-ones, time to raise a concern, internal mobility, regretted turnover in a specific team — and look at them at ninety days. If you cannot name what you expect to move, you have bought entertainment.
6. Make it safe to be bad at it first
Every new leadership behaviour is clumsy the first six times. If your culture punishes clumsy, your leaders will only ever practise in private, which means never. Say it out loud: we are all learning this, we will all be awkward at it, and awkward is the price of competent.
7. Go first, visibly
If you lead the leaders, they are not watching the program. They are watching you. If you are asking your team to give direct feedback and you have not asked for any yourself in a year, the program is decoration. Nothing in a curriculum outranks what people see the person at the top actually do.
What I built, and why
I have spent close to two decades in rooms with leaders, and the same frustration kept surfacing. The work was good. The retention was terrible. Not because people did not care, but because the support arrived on a calendar rather than in a moment.
So I built Worksparks — AI-powered leadership coaching that puts my frameworks in a leader’s hands at the point of need. Before the hard conversation, not three months after it. It is not a replacement for human coaching, and I will never pretend it is. It is the thing that lives in the gap between the sessions, which is where most of the leadership actually happens.
For teams, right now
Worksparks is currently on Indiegogo, and there is a tier built specifically for teams.
Team Spark: ten seats for twelve months at A$1,950, down from A$4,200.
That is ten leaders, a full year, group reporting so you can see what your team is actually working on, shared features, a kickoff session to get everyone started, and a signed set of all three of my books presented at that session.
Ten leaders for a year, for less than the cost of sending one person to a two-day workshop, they will half-remember by Friday.
Founder pricing exists for one reason: the people who back something early carry more risk, and they should get more value. It closes when the campaign does. Delivery is in October 2026.
If you lead a team, run an HR function, or you have been staring at a leadership budget wondering how to make it stretch further — take the tips above and use them regardless. They cost nothing.
And if you want the rhythm built in rather than relying on willpower, the link is below.
Back Worksparks on Indiegogo →
Sonia McDonald is the founder and CEO of LeadershipHQ, a keynote speaker, executive coach, and the author of Leadership Attitude, Just Rock It!, and First Comes Courage.