I think you have to be careful not to go too far oversimplifying the metrics by which you measure local foods (we did with recycling and I think it has been a real problem for the industry - if interested you can check out my series of blogs on the issue at the Max-R blog).

I think when it comes to local foods, you have to account separately for grown locally, processed locally, and at least grown regionally (say 500 miles as opposed to 50-150 for local). I think you also may want a category for "contains some local ingredients," say for stuff with more than 10% local but less than 75 or 80% local.

Some examples:

Locally roasted coffee may be great from a flavor standpoint and a local economic development standpoint, both of which are very important. Not as sure though of the advantage from a sustainability standpoint (there may be one but I have not yet seen the sustainable supply chain study on this).

Local grains in bread - I think this is something we all ought to encourage, but to be blunt, unless you include local water in the calculation or are living in heavy-grain producing areas of the Midwest, I doubt you are going to get more than 50% local and in most cases will be hard pressed to get more than 20% local ingredients. I would hate to see us undermine this important and developing market for locally-adapted grains, by choosing the wrong definition of "local" in our benchmarking metrics.

Seasonably variable processors - there may be some processors that like restaurants, use as much fresh local produce as possible in season, but have to augment at other times of the year (especially in bioregions with short growing seasons). Some may not differentiate between the two (i.e. you may not be able to get only the "local" batches), so on aggregate, they may not qualify for some definitions of "local" but I would hate to see their seasonal efforts discredited by an overly restrictive metric.

One thought is that you might want to consider using "stacking bars" in your graphs to show local. This might allow you to show all of these variations in a single bar chart if you are concerned that people won't pay attention to multiple charts.

Hope this helps.

Thanks Roger. Yes, as we are examining our food supply chain more, we are finding simply defining local by miles as overly simple. For example, we have some well known beverages in our dining halls that use local water as it's main ingredient (over 50% of it's ingredients), but is it a local product? And is it food?
We might have to also look at what we define as food. That is one of the questions we are grappling with.
In another case, we have a local producer that canned tomatoes so the production is local but the tomatoes are not. So we are trying to figure out how to incorporate that as local.
We're finding it's not as easy as an all or nothing. We've begun to pro-rate purchases based on the amount of ingredients or percent of the production that takes place locally but that relied on guess work to a degree.
I like your stacking bar suggestion and your labelling idea. I'll bring them back to our food service division.