Almost one year ago in mid-March 2011, the New York Times put up a paywall on its site and started charging frequent visitors. So how is it doing? Traffic to the NYT predictably took an initial hit and is still slightly below its pre-paywall levels. But the number of paid digital subscribers is growing steadily.

In its most recent quarter, the New York Times announced today that there are now 454,000 paid online subscribers to the NYT and the International Herald Tribune, up 16 percent from the previous quarter. (If you add in the Boston Globe’s 18,000 digital subs, the total rises to 472,000, but I’ll stick with just the NYT numbers for this analysis and the chart above).

The number of paid subscribers to the NYT is up 62 percent since the second quarter of 2011, when it had 281,000. Digital subscribers pay between $15 and $35 a month (although you can get a 4-week trial right now for 99 cents, and print subscribers get the online versions for no additional charge). Total circulation revenues at the New York Times Media Group is up 12.8 percent to $190 million, while total advertising revenues are down 4.5 percent to $173 million. Circulation revenues are now more than advertising revenues.

And it’s not just weakness in print advertising that is bringing down those numbers. Digital advertising revenues in the news group alone are down 2.3 percent (and down an even greater 10 percent if you count the struggling About.com unit). Without the paywall, digital ad revenues would be higher.

But the tradeoff seems to be working somewhat. Circulation revenues are growing faster (up 12.8 percent) than advertising revenues are declining (down 4.5 percent). All of these numbers are just for the New York Times Media Group, but the same trends are showing up across the entire company, which includes numbers from the Boston Globe and About.com (Total advertising revenues are down 8.1 percent, while total circulation revenues are up 9.7 percent).

The NYT is now at about half a million paid digital subscribers. That number is partly inflated by special discounted trial offers, but if it can keep those trial members paying at full price or convince print subscribers to stick around, the paywall will be doing its job. It really needs to get to one million paid digital subscribers to prove enduring. By that point, digital advertising revenues might also rebound because paid subscribers are considered more loyal, and thus more valuable to advertisers. That’s how it works in print media. Magazines and newspapers routinely spend marketing dollars to increase paid circulation so that they can charge more for the ads. The ad rates are typically based on the number of “paying” subscribers (even if they are not all paying full price).

Online it’s a little bit different because the same ads are shown to paying subscribers and casual visitors ( who can read 10 stories a month for free). Paying subscribers, though, will make up an increasing percentage of the total pageviews since they are not limited in how many pages they can view. The NYT salesforce should be trying to charge more for its paying audience, which overtime should result in a higher blended CPM (cost per thousand impressions) for its online ads—or at least keep upward pressure on those rates.

It doesn’t take that many more paying subscribers to impact pageviews significantly, which is what the advertising revenues are based on (impressions). If you are paying for the NYT, chances are you go there every day and click to your heart’s content. I’d love to know the number of clicks per paid subscriber. But I’d be willing to bet that at about one million paid subs, the pageviews more than make up for what they lost to the paywall.