We seem to be at the stage where we’re still figuring out what components of cinnamon are the mechanisms of its various actions. We’ve just recently finished establishing the the effects are genuine. Most published studies I find are clarifying the effects of cinnamon on different conditions, but not attempting to isolate the active compounds or exact mechanism. In fact, most studies don’t seem to specify the kind of cinnamon they used!
This PubMed article on the potential benefits of cinnamon in treatment diabetes mentions that Cassia is the type most often used in the research. However, the second page of the article notes the various suspected mechanisms; they mention a variety of suspects, but coumarin is not among them:
The mechanism by which cinnamon increases GLP-1 is unknown, but its effects may be similar to the GLP-1 analogues exenatide and liraglutide, or the dipetidyl-peptidase 4 (DPP4) inhibitors sitagliptin, saxagliptin, and linagliptin.
Cinnamon may also be involved in the activation of peroxisome proliferator–activated receptors (PPAR)-γ and PPAR-α.[3,13,14] Thiazolidinediones, including pioglitazone and rosiglitazone, target PPAR-γ and promote insulin sensitivity. The fibrate class of anti-hyperlipidemics target PPAR-α, which lowers plasma triglycerides and elevates plasma high-density lipoprotein cholesterol.
Finally, rodent models suggest that cinnamon may inhibit hepatic 3-hydroxy- 3-methyl-glutaryl CoA reductase activity.[15–17] This mechanism parallels that of the popular statin drug class.
I don’t even see mention of the family that includes coumarin. (For reference, coumarin, also called 1-benzopyran-2-one, is in the benzopyrone chemical class.)
So… the fact that Ceylon is low in coumarin is not a problem. However, it may also be low in whatever else is in Saigon and Cassia which have the beneficial effect. That remains to be seen. For now, the, most of the knowledge is about Cassia, specifically.