Activate cartwheels. The North American comics industry has crawled out of the hole it's been in to raise estimated profits of $715 million, the best it's been since 1993 or 1994, according to cautiously optimistic numbers analyst John Jackson Miller.

But hold on. We haven't quite recovered from the mid-'90s crash quite yet.

While a number of sites are running with the two-decade comparison, it's not quite as clear cut, or as celebratory as it might suggest.

Miller himself notes the '93 and '94 figures aren't adjusted for inflation, and he added an update to his original post that went into this more. "The most frequently cited figure for sales in 1993, the market's all-time peak, is $850 million," he writes. "That amounts to an inflation-adjusted $1.35 million, nearly double the size of the current market." Once you add in increased cover prices and other factors, Miller notes, "we’re still quite a lot behind the early 1990s in adjusted dollars."

Miller also briefly touches on something I've long thought, that to truly measure the health of the industry, we should be making more comparisons based on units, not dollars. Sure, it's awesome to make money, and I realize it's pretty standard in business to focus on the dollars, but just looking at a stack of money doesn't really tell the whole story of how that money came to be. How many people are putting money into the industry? The most accurate way to do it would be to know how many eyeballs are reading each issue, but that would probably break some privacy laws or get into 1984 territory. So knowing how many copies are sold is the next best thing. This information is available for more recent sales records, but whenever we get to these year-end analyses or compare year-to-year figures, we usually focus on the dollar amount. But by just considering inflation alone, that's just not an accurate gauge.

1993 was the year the speculator boom peaked. From what we know of historical sales figures, it was our best year. Unit sales in the six-figure range was the norm then, some were hitting the millions. But the plummeting crash was just around the corner. Today, our comic book stores can barely sell more than 100,000 copies of 10 titles a month. Most sell copies in the five figures. When a comic sells more than 200,000 copies, it's time for a parade. Still, it's a mild improvement from the late '90s and early 2000s, when breaking 100,000 was a struggle.

Yes, things are getting better. But better doesn't mean all better. We're definitely headed in the right direction, with more opportunities and more channels for more units to be moved than ever in the history of the industry. We also have greater variety and higher quality than ever before. But even with comic stores, newsstands, subscriptions, book stores, libraries and digital platforms offering a number of genres and formats, looking at units sold tells us we're still reaching less people than we were 20 years ago. The good news is that because of these outlets and diversity of content, we're probably reaching a greater variety of readers, and probably with a greater ratio of readers to collectors (meaning less people buying up multiple copies to re-sell later). That's informed speculation but I don't think it's a risky guess.

DC's New 52 was striving to recapture the lapsed readers from 20 years ago, but based on Todd Allen's analysis, that just isn't happening in significant numbers. All publishers need to make greater efforts to reach not only those lapsed readers, but those who have never read comics or are just learning to read. Some are, some aren't.

What focusing on units tells us is that there aren't as many people buying comics despite higher cover prices bringing in more dollars than ever before. This is hardly revolutionary, but it puts hard numbers behind casual observations. It sets a more definitive bar and removes the fluctuating guess-work and sliding scale of inflation. 100,000 copies will always equal 100,000 copies no matter the year.

Maybe comics are doing enough things right that we'll see a real comparison to 1993 numbers in another five years or so. Maybe it's just a matter of time. I hope so. But we can always improve how we analyze our progress.