
As I've been reflecting on the piece, considering what perspective to apply to understand Dell's debacle, I think it is one of growth into ineptitude. Another story of a business model that successfully grew for quite some time, inadvertently sowing the seeds of its own destruction.

In Dell's case, there seem to be three classic sources of that destruction of their successful business model.

Let me see if I have this picture right. You take a group of people and tell them they aren't going to be "real" Dell employees, with Dell status and benefits level. Then you assign them the task of fielding calls from interested potential retail buyers. When they do a good job, you tell them they get nothing but "paid." And then Dell wondered why they lost ground in the retail segment.
That's a classic example of someone in charge being totally inept, and not caring a whit about the results of their ineptitude. Nor, apparently, for at least a few years, did their immediate and two-levels-up superiors.
Next, we have Dell's success in placing PCs into American homes become a factor in its own demise. By raising the bar with market penetration, the next buying decision for many families was approached differently than that of their first PC. Perhaps their needs changed, and grew. Or, perhaps they taught themselves that now, having successfully bought and used one PC, online or by phone, they knew enough to go to a locally sited big box retailer and buy "any" brand which satisfied their now-savvy technical/application needs and price points.
Lesson for marketing executives. When a market is immature, a brand can be worth quite a bit in terms of quelling customer post-sale cognitive dissonance. When the customer gets confident in his knowledge of the product, the rebuy decision won't be the same. And the customer may well value "brand" much less than when he was ignorant of the product.
So that seems to be another factor in Dell's lagging performance. Its marketing executives didn't keep their finger on the pulse of retail consumers, and, as a result, they missed important changes in that segment's buying behaviors. Score one for the bricks and mortar retailers, and the companies that elected to collaborate with them, like HP.
Finally, Dell has, I think, simply run out of road for its once-vaunted business model. This might be mostly a business development issue, rather than a marketing one. However, good marketing intelligence would have seen this coming, with the aforementioned changes in consumer buying behavior, as the changes were occurring, rather than two years later.
Would opening a chain of Dell stores have mattered? Maybe not. I've been looking around their website lately, and the cupboard's pretty bare for consumer products. In this case, I think one could fairly cast the issue as one of poor segment choice. The consumer segment seems to still be buying PCs, but Dell didn't really keep pace with features and price points, as its competitors did. They bet on the business segment, and it didn't provide sufficient growth and profitability to sustain Dell's formerly consistently superior total return performance.
It's not about exploding batteries, so much as just poor marketing in the swollen, mediocre ranks of Dell. And, thus, perhaps an unfixable problem, given the mature nature of the sector, and Dell's important and longstanding mistakes which have hurt it now for several years. It may well be the GM of the computer sector, going from dominant producer to an also-ran with product nobody really wants to buy anymore, and little room for a comeback.
Time will tell. And I don't think it will take all that much time, either.