Logic Models Are the Worst

There, I said it. Not that this is a novel opinion – anyone laboring over a logic model chart during the last stretch of a proposal is probably chanting some variant of this thought. But I think it’s helpful to talk about what’s wrong with this common tool, in the hopes that we can all come up with something better. Logic models come from the same impulse as expansive program metrics, an effort to put concrete measurements in place so any outside observer can see if a program works. And the problems with logic models are very similar: logic models have become exercises for their own sake, rather than a helpful tool for program design.

I have worked with more than 20 organizations on more grants than I can count, and never once have I seen a program or organization that uses a logic model as a living document. We spend hours filling in fiddly boxes on complicated charts to submit with a grant or a report, and then never look at them again. How is that helpful for anyone? It clearly doesn’t help implementing organizations, and I’ve never seen a logic model that actually gave a good roadmap for funders to follow to understand how a program is supposed to achieve its goals. Instead logic models become something for nitpicky critique and mutual frustration.

So what should we use instead? I don’t have a magic tool that will fix everything. But I think the right starting point should be as basic as possible. What activities does your organization do? What will hopefully happen or change because of these activities? What does your organization need to do these activities? Yes, there are always external forces, and risks, and things that are outside the control of staff implementing activities. Logic models try to recognize complexity and account for it but end up making it nearly impossible to describe the basic aims of a program, because complexity is really hard to capture neatly in a one-page graphic. No program can anticipate and account for all layers of complexity.

Let’s start from the assumption that most nonprofits are well aware of the complexity of their work and are thoughtful in how they design and implement activities. Then, rather than trying to capture all of that thinking in a tool not suited to do so, let’s get back to basics again: if you don’t’ see any of the changes you anticipate from your activities, what would you investigate or change? If this can’t be well captured in a written proposal in paragraph, bullets, or a graphic of an organization’s choice, maybe this is a place for conversation between grantmakers and nonprofits. But any solution starts with acknowledging a common thought: logic models are not working. So let’s find something that does.