From what I've seen of PR, it's not very effective, especially not for the $/results ratio. The point of Andrew Chen's article is indeed good -- viral growth, if you can achieve it, is awesome. But press is far from useless.

I have no idea what the product did at the last startup you worked at. A lot of times companies think pounding the press will bring them success, which is, again, not true (a shitty product that gets press won't go anywhere). However, a good product that receives press should settle down at more signups/day than before the coverage. Enough exposure, and a good product should eventually take off: people keep telling people about products they recommend, slowly, over time. You eventually build up a base of enough people that are recommending you that you really start to grow.

If it doesn't, I'd take a hard look at the product.

As another data point, I've seen the exact same thing. About 8 years back, I worked at a funded startup that got tons of press, because we were all teenagers and it was the height of the dot-com boom. We got a spike everytime we were mentioned, but it didn't stick, and we were actually losing regular users (they thought we'd gone all corporate, which I guess we had).

Build something worth talking about before you get people talking. In our case, we got press because of who we were and not what we'd built, and that's not sustainable.