Yesterday, Roger Penske backed out of the deal he had struck with General Motors to buy the Saturn brand, ultimately signaling the end of GM’s two-decades old import-fighting brand. The deal was clever: Penske would buy all 350 dealerships and the Saturn brand, and GM would continue to build the current Saturn lineup for the next year or so while Penske found another supplier (widely rumored to be Renault). However, the third supplier backed out, giving Penske no reason to go through with it. GM has little choice but to wind Saturn down.

At only 19-years-old, Saturn was one of the shortest-lived brands GM ever had. Yet it was also pivotal to the company’s current fortunes – or lack thereof – in myriad ways, from the promises made during its development to the billions of dollars poured into the brand that could very well have saved other divisions in the process. GM established Saturn to learn a new way of doing business. Instead, the business it already had was almost undone.

There’s a great history of the Saturn brand over here, and I’m not going to repeat it. To me, Saturn had an inauspicious beginning. GM spent billions of dollars in development, forged an innovative new labor contract, built a brand new factory, implemented new manufacturing techniques and established a supposed laser focus on how to beat Japanese manufacturers at their own game. Yet the resulting cars were, at best, only pretty good. I remember one of the buff books at the time saying that the new SL sedan was at the “top of the middle of the pack,” easily the best domestic small car, but still behind the Honda Civic, Toyota Corolla and Nissan Sentra. Over time Saturns improved a little, but the cars were never best-in-class as GM had promised when it announced the brand. Instead, Saturn primarily survived because of its incredible customer service and resulting customer loyalty; anybody remember the two Saturn Homecomings? But great customer service can’t make up for two decades of inadequate cars. Today, at the end of its life, Saturn easily has the strongest product lineup it has ever had. Yet, in many ways, it’s still where it has been the whole time: Exactly not good enough.

The good news – sort of -  though is that some of the best examples of cars better than Saturns come from other GM divisions. For example, the Aura sedan is nice, but the Chevy Malibu, based on the same platform, is better. The Astra’s not bad, but it’s not better than the Cobalt, and the upcoming Cruz has the makings of a winner. The Saturn Outlook and Vue are fine, but both are outclassed by platform-mates from Chevy and GMC. The fact that GM’s product lineup is pretty strong right now is a silver lining on a very stormy cloud, and cutting this dead weight away will definitely help GM’s bottom line. But there’s no recovering the billions spent to keep the division afloat, and the General still faces an uncertain future as it attempts to lure skeptical customers back to showrooms.

The Saturn experiment was meant to teach GM how to build competitive cars. Instead, it learned to do that in spite of the Saturn albatross. What Saturn did teach is that each division needs a focus, that the cars have to be better than good enough, that great customer service is worthless if there aren’t great cars to back it up, and that at some point you have to stop throwing good money after bad. Today, as GM stands cut in half, humbled by bankruptcy, it remains to be seen if the real lessons of the Saturn fiasco have truly taken hold. If they haven’t, then Saturn surely won’t be the last GM division to face the axe.