WeLearn2Profit

Opinion Learning and development · Team building

Old-school team building vs modern L&D: which actually changes behaviour?

[email protected]

Published October 22, 2026

Publisher's interest: our publisher, WeLearn2Profit, runs team-building and training programmes.

In short

A fun day out can build goodwill, but there is little rigorous evidence that it changes how people work. The evidence for structured training is larger and more consistent. So design team building like training, or call it a reward.

Many HR teams face the same budget question each year. Book a team-building day, or invest in learning and development (L&D)? Both promise better teamwork, but the evidence behind them is not the same.

What I mean by each

By old-school team building, I mean a day of games, outings or challenges, run mainly for fun and bonding. By modern L&D, I mean structured training with clear goals, practice, feedback and follow-up at work.

The line is blurry. Good team-building days often include structured work, and many training days include games.

Team building: the evidence is mixed

A 1999 review of team-building studies by Salas and colleagues found “no significant effect of team building on performance” overall. Programmes that focused on clarifying roles were more likely to help.

A 2009 review by Klein and colleagues was more positive. It found “a positive moderate effect across all team outcomes”. The effect was strongest for feelings and attitudes, and for how the team works together.

These reviews looked at structured team building, such as work on goals, relationships, problem solving and roles. Evidence on fun outings on their own is thin.

A 2008 meta-analysis of ropes and challenge courses found medium-sized gains overall. But most of its studies were dissertations, most outcomes were self-reported, and only a few appear to involve workplace teams.

We found no controlled Malaysian study of employees. The Malaysian studies we found were surveys of employees, or studies of students and trainees on outdoor camps.

Structured training: stronger evidence, with a catch

A 2003 review of organisational training by Arthur and colleagues reported “a medium to large effect size”. The effects were similar for reactions, learning, behaviour and results.

Two 2017 reviews point the same way. McEwan and colleagues found medium-sized effects of teamwork training on teamwork and team performance, in controlled studies. Lacerenza and colleagues found that leadership training improved reactions, learning, transfer to the job, and results.

The catch is transfer: whether people use the skills back at work. A 2010 review by Blume and colleagues linked transfer to the learner’s cognitive ability, conscientiousness and motivation, and to a supportive work environment.

In other words, support back at work appears to matter for whether training gets used.

My view

Fun outings are not bad. But on their own, I think they are a weak tool for changing how people work. There is little rigorous evidence that they do.

If you want behaviour to change, design the day like training. Use a real work problem, make roles clear, practise, give feedback, and agree what happens next week.

If you want to thank people and build goodwill, call it a reward. Do not expect it to fix a performance problem.

The other side

Supporters of traditional team building make fair points.

  • Goodwill has value. Klein’s review found that structured team building had its strongest effects on feelings, attitudes and how teams work together. Supporters argue that a fun day can build similar trust, though this has not been rigorously tested.
  • Not everything needs a business case. A celebration or a day out can simply be a thank-you. Judging it on performance misses its purpose.
  • Training fails too. A course cannot fix poor management, unclear roles or broken processes. Without support back at work, much of it may never be used.
  • The research has limits. Salas’s review found a small but significant gain on rated (subjective) measures of performance, though not on objective ones. Blume’s review found that links with transfer looked stronger when everything was measured by the same source, in the same setting.

Questions to ask before you book either

  1. What should people do differently afterwards?
  2. How will we know, and when will we check?
  3. Who follows up at work: the manager, HR or the provider?
  4. Does the activity use a real work problem, or a generic game?
  5. Is this a reward or a fix? Be honest about which.

Sources

More opinion

All opinion