Serving size can certainly be an issue. One of the things that sprung to mind this morning is the question of whether you (any “you” reading this post!) have a linear or nonlinear glycemic response to a particular grain: that is to say, if you have exchanges of bread instead of one, will you run twice as high and come back down in the same amount of time, or the same upward jump and twice as long a fallback, or will the combination of degree of rise in blood glucose and time spent elevated be greater than twice that of one exchange? (In more technical terms, does the area under the elevation-time curve increase in proportions greater than 1:1 for one or more additional exchanges of a given reference food, eaten at the same time?)
The “reference food” on which to gauge a potential reaction to a particular breakfast cereal is likely to be the grain on which the cereal is based – a slice of bread or 1/3 cup cooked spaghetti equals 1 wheat exchange for a wheat-based cereal, a cup of unsalted, unbuttered popcorn as 1 corn exchange for a corn-based cereal, 1/3 cup of cooked, unsalted rice for a rice-based cereal, etc.
Note that the quantity of reference food one 15-carb-gram exchange. This is usually about 20 grams of uncooked grain or pasta, or ~40 g cooked. Note that the serving sizes listed on the packages are usually between 30 and 40 g carbohydrate – two to three starch units!
If you can normally tolerate a full package serving of pasta with tomato sauce, without meat, then you may be able to tolerate a full package serving of unsweetened wheat-based cereal (e.g., Shredded Wheat, Wheatena, or Cream of Wheat) with a half-cup of milk. If you can only tolerate that amount of grain product with meat, then you may need to eat the cereal alongside something like a serving of breakfast sausage or eggs – or have the protein as the main part of your meal, with a partial serving, like 1/4 cup of a puffed cereal, or 1/4 c cooked cereal, as a condiment or “dessert” to your breakfast.