{"id":46349,"date":"2025-08-15T10:33:17","date_gmt":"2025-08-15T10:33:17","guid":{"rendered":"https:\/\/www.mindtools.com\/?p=46349"},"modified":"2025-08-15T10:33:18","modified_gmt":"2025-08-15T10:33:18","slug":"how-to-prove-the-impact-of-ld-in-your-business","status":"publish","type":"post","link":"https:\/\/www.mindtools.com\/blog\/how-to-prove-the-impact-of-ld-in-your-business\/","title":{"rendered":"How to prove the impact of L&D in your business"},"content":{"rendered":"\n

Open LinkedIn on any given day and you\u2019ll see someone talking about how L&D needs to get better at proving impact. Peruse any awards criteria and \u201cimpact\u201d will be in there.<\/p>\n\n\n\n

We all know it\u2019s something we should do. But it\u2019s easy to get bogged down with the production side of things and fall into a \u201cbuild it and ship it\u201d frame of mind.<\/p>\n\n\n\n

But in not proving impact, are you running the risk of your L&D function being viewed as a drain on costs? How can you change that and start proving the impact of L&D in business?<\/p>\n\n\n\n

<\/p>\n\n\n\n

What does \u201cimpact\u201d actually mean?<\/h3>\n\n\n\n

To paraphrase my fellow learning designer, Ross Dickie, it\u2019s answering this question: \u201cHas this thing we\u2019ve invested so much time, money and effort into actually made a difference?\u201d<\/p>\n\n\n\n

Let\u2019s break that down. To establish if \u201cthe thing\u201d made a difference, we need to know what the starting point, or benchmark was. L&D programs usually emerge from a need of some kind. Whether that\u2019s one that the business recognizes or the L&D function discovers doesn\u2019t matter \u2013 the need is there. And it\u2019s usually tied to a business priority or goal.<\/p>\n\n\n\n

So, there\u2019s your starting point. You know what you\u2019re trying to achieve, now you can measure it.<\/p>\n\n\n\n

<\/p>\n\n\n\n

What should you measure?<\/h3>\n\n\n\n

The actual metrics you measure will depend on the program. If it\u2019s a career development program, you could measure internal promotions, job moves and attrition. If it\u2019s onboarding, you could measure early attrition and speed to competence or capability. Or if it\u2019s leadership development, you could measure leadership capabilities.<\/p>\n\n\n\n

<\/p>\n\n\n\n

Measure the before and after<\/h3>\n\n\n\n

Like our client South Western Rail<\/a> did on their award-winning LEAP leadership development program. Our Insights and Analytics team worked with them on audience research and evaluation to provide some direct comparisons before and after the program. The result? An average of 12% improvement in leadership capability among participants at a 95% confidence level that the gains resulted from the program. The program also had some ripple effects which are more difficult to measure, like better collaboration and increased confidence.<\/p>\n\n\n\n

<\/p>\n\n\n\n

Look for correlation if you can\u2019t prove causation<\/h3>\n\n\n\n

Correlation and causation are not the same thing, and you\u2019ll often see \u201ccorrelation does not imply causation\u201d quoted by the data-savvy.<\/p>\n\n\n\n

Correlation<\/strong> is a statistical association between two things without a direct cause. When thing A changes, so does thing B.<\/p>\n\n\n\n

Causation<\/strong> is where there\u2019s a direct cause-and-effect relationship between thing A and thing B.<\/p>\n\n\n\n

It\u2019s easy to assume that to prove impact, you have to prove causation. But that\u2019s not always practical. For example, it might be too risky performance-wise to run a control group study, or the business wants a fast intervention. In these cases, correlation is your friend.<\/p>\n\n\n\n

Proving direct causation can be tricky anyway. Learning doesn\u2019t happen in a vacuum. It\u2019s not always easy to pinpoint the cause of an improvement to the L&D intervention, operational improvements, coaching, or a combination of all of the above. Dr Ina Weinbauer-Heidel\u2019s 12 levers for learning transfer<\/a> shows just how dependent transfer is on factors outside of the L&D function \u2013 5 out of 12 levers are organizational and 3 out of 12 are on the individual, leaving 4 out of 12 for the L&D function.<\/p>\n\n\n\n

In our South Western Rail<\/a> example, we used a confidence level to indicate how much of the success was attributable to the L&D program. But what if you don\u2019t have data scientists in your L&D function? Or the budget to partner with a behavioral insights team<\/a>?<\/p>\n\n\n\n

<\/p>\n\n\n\n

Look for evidence of learning transfer<\/h3>\n\n\n\n

In the Custom Learning team, we\u2019re all huge fans of Will Thalheimer\u2019s Learning-Transfer Evaluation Model (LTEM)<\/a>. Tiers 2 (learner activity), 5 (decision making), 6 (task performance) and 7 (transfer to work performance) are good indicators of whether your program contributed to an improvement in the business.<\/p>\n\n\n\n

<\/p>\n\n\n\n

Don\u2019t forget the impact on learners<\/h3>\n\n\n\n

Engagement metrics might not be perfect, but they have their use in the overall data story of your program. Completion rates on their own get a bad rep, but pair completion rates with a comparison of before and after performance and you\u2019re on the right track.<\/p>\n\n\n\n

Learner perception scores also get a bad rep. But measure their perception on the right things and that bad rep seems a bit undeserved. Instead of asking learners if they \u201cliked\u201d the facilitator, slide deck, animations, ask if they found the activities useful. Ask them to identify what they\u2019ll be able to do differently or better as a result of the program. Likes are cheap, look for the value.<\/p>\n\n\n\n

<\/p>\n\n\n\n

How Mindtools can help<\/h3>\n\n\n\n

You don\u2019t have to prove impact on your own. This is the kind of thing we love doing at Mindtools! Here are just some of the ways we can help:<\/p>\n\n\n\n