Logic modelling can be helpful to design a programme evaluation. They are not the only way to support evaluation design, but they are a relatively easy way to assist you in structuring your thinking.

What logic models are used for

What logic modelling is good for

A good logic model can help you to:

  • get an overview of your intervention and how it relates to your programme objectives
  • understand what data you might be able to collect throughout your programme
  • identify gaps in your programme or data you might need to evaluate it
  • understand what you can and can’t learn from your evaluation

What logic modelling isn’t good for

Logic model is helpful for describing what you are doing and what you are expecting to happen; it’s can’t help you see why it is happening. On its own, logic modelling can’t help you to understand whether you are making the right interventions as part of your programme. The models are also intentionally simplistic. Many things could influence the outcomes experienced as a result of your programme.

Whilst they have limitations, logic modelling can help you to structure your evaluation, so that the data you gather throughout your programme delivery can demonstrate its effectiveness. Use them to structure your thinking, but validate the model with actual data throughout the delivery of your programme.

Create your model

There is no one right way to create a logic model. Logic models are typically structured as a table that details:

  1. inputs
  2. activities
  3. outputs
  4. outcomes
  5. impacts

Typically, you would start by filling in your desired impacts, and then fill in the remainder of the table with information about your programme.


Inputs are the resources and things you have available to you that will feed into your programme. Typical inputs could include:

  • money
  • staff or volunteers
  • time
  • tools


Activities are the interventions you will undertake as part of your programme. In the case of mentoring programme, they could include:

  • registration processes
  • induction events
  • mentor matching processes


Having undertaken your activities, outputs are the things you expect to have happened as the initial effects of your intervention. Outputs from your programme could include:

  • having a database of mentors and mentees
  • resources for mentees having been created
  • speed mentoring sessions taking place
  • emails being sent

You might be able to measure outputs as part of your evaluation. For example, a success measure could be to have 1,000 mentors and mentees registered on your programme.


Outcomes are the bigger picture effects caused by your intervention and its outputs. For example:

  • having undertaken induction events and provided guidance, mentors might feel more confident in supporting their mentees
  • having conducted a matching process, inducted mentees, and sent them notification emails, mentoring will have taken place, generating a certain number of hours mentoring per mentee

Again, you might be able to measure your outcomes as part of your evaluation. For example, a success measure could be that a minimum number of hours of mentoring is enabled by your programme.


Impacts are the desired effect of your programme overall. These are usually big picture, strategic effects. For example:

  • mentees feel more confident
  • senior leaders are more visible within your organisation
  • participants report that they have achieved a promotion after taking part in the programme

You might also be able to measure impacts, though some impacts might be difficult to link back directly to your interventions.