Fitbit, maker of elastomer step-tracking bands and Wi-Fi-equipped scales, sold 4.5 million devices over the past few months, with revenues up 235 percent from a year ago. That’s according to the company's second quarter earnings report, the first report since Fitbit went public in June.

One key metric absent from yesterday’s data? Active users.

Fitbit’s co-founder and CEO James Park said that Fitbit won't share this info on a quarterly basis, but annually — an avoidance tactic that could raise more questions about the long-term use of these types of activity trackers.

Until recently, the evidence was anecdotal: you probably know someone who loved their Fitbit or Jawbone or Samsung Gear Something for awhile, then tossed it in a drawer. One research firm reported that about a third of owners of "smart wearables" ditch their devices within six months of purchasing them.

Once market-leading Fitbit became a publicly traded company, wearable watchers finally got a glimpse at real numbers. Fitbit’s revelation that it makes money — a rarity among the current list of buzzy San Francisco startups — almost made it easy to miss the fact that nearly half of registered Fitbit users weren’t active. Out of 19 million registered users at the time, only 9.5 million were active. Prior to that, Fitbit had shared that out of 10.9 million users, around 6.7 million were considered active.

Fitbit has even coined its own term: Paid Active Users. This refers to anyone who is paying for Fitbit Premium or Fitbit-owned FitStar, but also includes anyone who has logged just 100 steps, or has even connected one of Fitbit’s trackers or scales to the app — a pretty loose standard. (If you’re straphanging on the subway and the person standing next to you is wearing a Fitbit, you may be considered an active user by osmosis.)

When asked in an interview about the gap between registered and active users, Park, an intently focused guy who is a quiet contrast to startup founders battling large municipalities or raising money at billion-dollar valuations, said "engagement levels aren’t what drives the business," but that he’s hoping to see improvement over time.

"In context, you have to compare Fitbit devices to gym memberships or home gym equipment," Park said, implying that Fitbit has a better retention rate.

But the abandonment rate for gym memberships has remained fairly consistent at 80 percent for a long time — so what does that say about wrist wearables?

In fairness to Fitbit, this is not a Fitbit thing; it's a categorical problem

In fairness to Fitbit, this is not a Fitbit thing; it’s a categorical problem. Other makers of activity trackers have their own strange ways of defining "active users" or "engagement," or may not release any information at all.

Jawbone, for example, which has a small slice of the wearables market, says that it has an 80 percent retention rate for users of its Up activity-tracking bands, provided those users are connected to a couple friends within the app and have also linked the Up app to another third-party app. It also claims three app opens a day, or an average of 20 app opens per week, among active users. How many active users? Shrug.

But the magic solution for making these products truly stick is still a giant TBD. One major complaint has been that they may not be accurate. Another is that the category is fragmented, so a wearable or even a wireless scale may share data with one app, but not another app, and may not work with this back-end software, and so on. And some believe that smartwatches, like Apple Watch, could subsume all of these products.

It’s not a stretch to suggest that if active user numbers were great on a long-term basis, more companies would share them. Maybe eventually, they will. Until then, the anecdotal pile-in-the-drawer continues to build.