How to Know Which Leadership Skill to Build Next
Most leaders trying to develop themselves work on leadership in general, which produces nothing measurable, or on whatever the last book or piece of feedback suggested, which produces thrashing. The alternative is narrow: one skill per quarter, chosen deliberately, made observable enough that you can tell at the end of the quarter whether you did it. This page is the selection method — four tests in priority order — and the practice loop that turns a choice into a change.
What’s inside
One thing per quarter, not eight
The four selection tests
Making it observable
The practice loop
How long to hold it
The thrashing failure
Where Blomma fits
One thing per quarter, not eight
The standard state for a conscientious leader is feeling behind on many things at once: delegation, feedback, strategic thinking, running meetings, developing the executive team, communicating better. All of them are real gaps and the list is paralysing, so the usual outcome is either reading about all of them or working on none.
One at a time works for a specific reason. Leadership skills are behavioural, which means they change through repetition under real conditions rather than through understanding. Repetition requires attention, and attention divided across six behaviours produces no repetition in any of them. A single behaviour, attended to for twelve weeks, becomes normal. Six behaviours attended to for twelve weeks stay novel.
There is also a second-order benefit. One deliberate change, explained to the people affected, reads as intentional. Six simultaneous changes read as instability, and unpredictability from a leader costs more than any single gap.
So the discipline is subtraction: from your list of eight, pick one, and consciously accept that the other seven remain unaddressed this quarter.
The four selection tests
Run these in order. The first that gives a clear answer is your skill.
One. What is the company’s constraint right now?Work backwards from the business. Execution keeps slipping — the skill is probably clarity of expectations or accountability. Good people keep leaving — feedback or development. Decisions are slow — delegation and decision rights. Strategy is vague — direction-setting. Let the company tell you, because a skill that addresses a live constraint has both urgency and a visible payoff.
Two. What are you avoiding?Your avoidance list is an unusually accurate skills gap analysis. The conversation you have postponed three times names the skill you most need. This test is uncomfortable and it is the most reliable of the four.
Three. What did the last three pieces of feedback have in common?Not the content — the theme. Feedback repeats itself for a long time before anyone escalates, so a theme you have heard three times is well-established and you have been discounting it.
Four. What does the next stage require?If you are about to cross a threshold — first managers, first executive team, first time past a hundred people — build the skill for the stage you are entering rather than perfecting the one you are leaving.
If several tests point at the same thing, that is your answer with high confidence. If they point at different things, take the second test’s answer; the avoidance signal is the strongest.
Making it observable
A skill named as a quality cannot be practised or assessed. “Get better at delegation” is not a plan. Convert it into behaviour with a count.
The format that works: a specific action, a number, and a deadline.“Hand over the three decisions currently queuing on my calendar, each with a written brief and a review point at two weeks.”“Give same-week feedback to each of my five reports, at least twice each, this quarter.”“Ask for their read before giving mine, in every one-to-one, and note the exceptions.”
Three properties make these work. They are countable, so at the end of the quarter you know whether you did it rather than whether you feel improved. They are behavioural rather than attitudinal, so they can be practised deliberately. And they are small enough to start this week, which matters because a development plan that requires a clear month never begins.
Write it down somewhere you will see it, and tell one person — a coach, a peer, your cofounder. A skill you have named to someone else gets practised considerably more often than one you have named to yourself.
The practice loop
Four components, and the third is the one people skip.
Reps.The behaviour, repeatedly, under real conditions. Badly at first, which is the point — you are not demonstrating the skill, you are building it.
Immediate notice.After each rep, thirty seconds: what happened, what surprised me. Not a journal entry — a mental note that keeps the attention on the behaviour rather than letting it become automatic in its old form.
External read.Someone telling you how it landed, because your own assessment of a behavioural skill is unreliable by definition. This is the component that most development plans lack entirely, and its absence is why leaders can practise something for a quarter and get no better.
Weekly review.Ten minutes: did I do the reps, what is getting in the way, what do I adjust. Enough structure to notice a stall in week three rather than week eleven.
The loop is unglamorous and it is the entire mechanism. Leaders who read about a skill and leaders who run this loop end the quarter in genuinely different places.
How long to hold it
A quarter is the right unit for most skills. Long enough for a behaviour to become normal, short enough to feel finite.
At the end, assess honestly against the count rather than the feeling. Did you do the reps? Did the external read change? Then one of three things: it is now normal and you move on; it is partly there and you hold it another quarter, which is a legitimate and underused option; or you did not do the reps, in which case the interesting question is what got in the way — usually the skill was chosen because it sounded important rather than because a constraint demanded it.
Then pick the next one, re-running the four tests. The constraint will have moved, and the second test in particular will have a new answer.
Four skills a year, genuinely built, is a substantial rate of change — considerably faster than most leaders manage, and it looks slow from inside because each individual quarter feels like working on one small thing.
The thrashing failure
The opposite of stalling, and the more damaging of the two.
Thrashing is attempting to become a different leader every quarter based on the most recent input — a book, an investor comment, a piece of feedback, a podcast. Each change is well-motivated and none is held long enough to take.
What makes it worse than stalling: it makes you unpredictable. Your team can work with a leader who has known, stable gaps — they adapt. They cannot work with a leader whose operating model changes every quarter, because they cannot tell which version of you they are dealing with, and every interaction becomes a risk assessment.
Two defences. Choose from the four tests rather than from the most recent input — the tests are anchored in your company’s actual constraint, which changes slowly. And when you do change something, say what you are changing and why to the people affected. That single act converts what looks like inconsistency into something your team can follow, and it models exactly what you want from everyone below you.
One marker of thrashing: if you cannot name what you were working on last quarter, you were not working on anything.
Where Blomma fits
Two parts of this method are genuinely hard alone. The second test — what are you avoiding — requires a candour about yourself that is easier to reach with someone who has no stake in the answer. And the external read is structurally unavailable to most senior leaders, because the people who can see how your behaviour lands are the people who report to you.
Blomma is an always-on AI career coach with no stake in your development. Use it to run the four tests and pick one skill rather than carrying a list of eight. Use it to convert that skill into a countable behaviour with a number and a deadline, which is where most plans fail. Use it for the weekly ten-minute review, which nobody sustains alone. And use it as the external read on the reps themselves — rehearse the delegation conversation, the feedback, the question you are trying to ask before giving your own view, and find out where the behaviour is not yet what you think it is. Where it warrants a human who has led at your stage, bring one in.
The second application: everything above applies to every manager in your company, and most of them have no development plan at all beyond an annual goal nobody revisits..
There is no way to get better at leadership in general. There is a way to get better at one specific behaviour in twelve weeks, four times a year — and that compounds faster than any amount of reading.