A training manager once asked us to run a two-day programme on emotional intelligence because two team leaders had resigned without telling anyone why. Her diagnosis was probably right. The label she used to describe it was doing a lot of quiet work that nobody had examined, and that is usually where these programmes go wrong.
The term arrived in Indian organisations through the leadership literature, which used it generously. It now appears in job descriptions, appraisal forms and engagement survey commentary, often meaning something like “pleasant to work with”. That is not a skill and it cannot be trained. The narrower definition can.
What the term covers, and what it does not
The most useful version is the ability model, associated with John Mayer and Peter Salovey. Emotional intelligence, in that account, is four related capacities: perceiving emotional information in yourself and others, using that information to think more clearly, understanding what produced a feeling and where it is heading, and managing your own response and the shape of an interaction.
That is a set of skills. Each one can be described, practised and observed. Each one also has limits, which is what makes it honest.
The wider definitions caused the trouble. Once emotional intelligence is defined to include optimism, initiative, adaptability and achievement drive, it stops being a distinct capability and becomes a compliment. Nothing can falsify it and nothing can be trained in it, because it is no longer one thing. Practitioners who roll their eyes at the phrase are usually reacting to the wide definition, and they have a point.
Where it stops being personality
Personality describes reasonably stable tendencies: how much company you seek, how quickly you feel threatened, how much structure you prefer. Emotional intelligence describes what you do with the emotional information available in a specific situation.
The distinction has practical consequences. An introverted analyst who dislikes social contact can still run a fair performance conversation, notice that the other person has gone quiet, and ask a question rather than filling the silence. Calling that person “low EI” because they do not enjoy the room is a category error, and it is a common one in promotion decisions.
It also sets the honest limit. Training can change what someone does with their temperament. It does not change the temperament, and a programme that implies otherwise is selling a personality transplant with a skills vocabulary. That promise is why so many participants leave an emotional intelligence workshop enthusiastic and return to the same behaviour within a month: they were asked to become someone else, not to do something differently.
What is trainable
| Element | Improvable with practice | What the work actually looks like |
|---|---|---|
| Noticing your own early signals of irritation or anxiety | Yes | Reviewing two recent incidents, naming the physical cue, writing a stop rule for the next one |
| Naming a feeling precisely | Yes, often quickly | Vocabulary work; “frustrated” and “embarrassed” call for very different responses |
| Pausing before replying in a tense exchange | Yes | Rehearsal against a scripted provocation, then practice in low-stakes meetings |
| Reading other people’s cues | Partly | Feedback from someone who was in the room; self-assessment is unreliable here |
| Staying regulated under real pressure | Partly | Repeated exposure, a debrief after each attempt, coaching in between |
| Warmth or sociability as a disposition | Barely | Not a training objective |
| Baseline threat sensitivity | No | Design the environment around it rather than promising to change the person |
The rows marked “partly” are where most of the money should go, and where most programmes spend the least time. A session that explains the four capacities and asks people to reflect on their childhood produces insight and no behaviour. A session that runs the same difficult conversation four times, with feedback between attempts, produces behaviour and less insight. The second is less satisfying to attend and more likely to change what happens on a Monday.
Three conditions separate the two.
- Real incidents, brought by the participant, not case studies written by the facilitator.
- Observed practice with feedback from someone who watched, not peer affirmation.
- A follow-up conversation in the participant’s own workplace within a month, where the new behaviour either happened or did not.
Leave out the self-report questionnaire used as a gate, the “your EI score” handout, and the type labels. They create a category for a person and then a defence of the category, which is the opposite of what a skills programme needs.
Measuring it without fooling yourself
Self-report measures capture how someone sees themselves. Ability tests capture something closer to performance on a puzzle, which is not the same as behaviour in a room. Neither tells you whether the person will pause before answering a hostile client.
What does tell you is observation. An assessment centre can put someone in a structured exercise and watch whether they notice the cue, name it, and adjust. A 360 collected carefully can tell you whether colleagues have seen a change, provided the questions ask about specific behaviour and the responses are not attributed. Both approaches are slower and less tidy than a score, and both are more likely to be true.
Two cautions on the way the label gets used. First, “not emotionally intelligent” is often a polite way of saying “disagrees with me”, and it is worth noticing who gets described that way in appraisal discussions. Second, the burden of managing a room’s feelings tends to land on whoever is most junior or most willing, usually regardless of role. A programme that teaches people to read a room without teaching them to say what they actually think has taught them to absorb other people’s discomfort, which is not the skill anyone asked for.
Emotional intelligence is not a virtue some people have and others lack; it is a small set of skills, and skills respond to practice and feedback.
There is one more thing worth saying to sponsors. Emotional intelligence training will not fix a team whose manager shouts, whose workload is impossible, or whose escalation route punishes bad news. In those conditions, reading the room accurately is a survival skill, and using it well may mean leaving. No amount of training compensates for a system that rewards the behaviour you are trying to remove.
What to do on Monday
Pick one behaviour, not a capability. “Asks one question before responding when a client complains” is a behaviour. Write the anchor, decide what it looks like when done well, and observe it in three real meetings over the next fortnight.
Then give the person feedback the same week, in plain words, about what you saw. If they want help between observations, that is what coaching is for, and it works better on a specific behaviour than on a general aspiration to be more self-aware.
Finally, read how to make difficult conversations easier alongside this. Most of what people call emotional intelligence at work is the ordinary craft of staying in a hard conversation without either going silent or going sharp, and that craft is teachable in an afternoon if the practice is real.