People & HR
Training needs assessment
Ask before you book the trainer: what people cannot do yet, and what stopped them coming last time.
Why this survey gets sent
“I booked the course everyone said they wanted, twenty seats, and eleven people came. The feedback said it was excellent. A year later nobody is doing anything differently.”
Most training budgets are spent on whatever was easiest to buy, then measured by a satisfaction sheet filled in while the coffee is still on the table. The survey that is worth running happens six weeks earlier and asks the two things a feedback sheet cannot: what people cannot do yet, and what made them miss the last session.
What is in it
- How confident are you doing each of these today? (rating grid)
- Which topics would help most in the next six months? (multiple choice)
- A skill we have not listed (short answer)
- How would you rather learn? (single response)
- How much time could you realistically protect? (single response)
- What stopped you attending last time? (multiple choice)
- What do you want to be able to do in a year? (long answer)
About 4 min to answer
Training needs assessment
About 4 min
How to write this survey
Run it before the budget, not after the course
Almost every organisation measures training at the wrong end. The evaluation sheet goes round while the room is still warm and asks whether the trainer was good — a question about the last two days, not about the next two years. The survey that moves a budget runs six to eight weeks before anything is booked, while there is still a decision left to make.
It also changes who owns the plan. A list of courses chosen upstairs arrives as an instruction; a list assembled out of what people said they could not do yet arrives as a reply. The second one fills its seats, and attendance is the first number anyone holding the budget will ask you about.
Ask what people can do, not what they want
Wants are cheap, and everybody has the same three. The useful question is a self-rating of confidence, task by task, because it produces a gap you can rank. Twenty people who are only slightly confident about running a meeting that ends in decisions is a course; two people who would like leadership training is a conversation.
Write every row as a task and never as the name of a skill. “Communication” means nothing and everyone rates themselves highly on it; “writing a report someone can act on” is something a person either did last month or did not. Add a “not part of my job” column so nobody has to rate work they do not do — and keep the rows word for word next year, because the same grid asked again is the only evidence you will ever have that the training worked.
Let people choose, but cap the list
A multiple-choice list of eight to ten topics, each written as an outcome rather than a course title, gives you a ranking in one question. Longer than ten and people tick everything that looks free; shorter than eight and the list is your plan rather than their answer. Leave exactly one short text box under it for the skill you did not think of — that box is where the specific, unexpected request lives, and it is often the cheapest thing on the whole plan to deliver.
The priority is not the most-wanted topic; it is the one where a strong want and a weak confidence overlap. Something many people ask for and are already good at is an away-day. Something few people asked for and almost nobody feels confident about is the risk sitting inside your operation — and it usually turns out to be the systems everybody touches every day.
Format and timing decide who actually turns up
Two questions about shape — how would you rather learn, and how much time could you realistically protect — save more money than the topic question does. A full day is unbookable for anyone on a shift or behind a counter, while two half-days or a short weekly session for a month fill the same hours without emptying a department. Ask it rather than assuming it: the answer differs completely between the office and the floor.
Then ask the question most needs assessments leave out: what stopped you attending last time? The answers are rarely about interest. They are about notice that came too late, a clash with a shift, a manager who could not spare anybody that week, or an invitation that never arrived at all. Every one of those is fixed by scheduling, not by buying a better course.
Say what you chose, and what you dropped
Publish the shortlist within a month, with the dates, and name what did not make it and why — the budget, the numbers, or a plan to run it next year instead. A needs assessment that disappears teaches people that answering it was optional, and next year's response rate is the receipt for that lesson.
Route the supervision question separately while you are there. Whether somebody manages other people, or is about to, splits the whole plan in two — the same topic needs a different session for each group — and it costs you one dropdown at the end rather than a second survey.