Many of us bought fitness wearables in 2014. We counted our steps, our distance walked, our quality sleep time. But almost as many of us probably ended up abandoning our new devices before we’d worn them even a month.

Because my job involves testing many of these devices, I’ve been through the adopt-and-abandon cycle several times over. I had hoped that one of these devices would stick — that I would continue wearing it and benefit from the so-called “actionable insights” gained from the data. But they all ended up in the junk drawer.

Here are some of the reasons that happens, and what needs to change to keep the wearable revolution headed in the right direction.

They need better functionality

Many fitness wearables on the market now are just accelerometers wrapped in a piece of plastic. The very limited set of features and functions that this enables is perfectly built for people who are OK with using a wearable for a couple of weeks then giving up. Many of these devices are so inexpensive that this loser of a use case can be a money maker for the manufacturer. But as the wearables market becomes saturated in 2015 this will be a tougher sell.

Many users will already have purchased (and thrown away) their first wearable, and will expect far more from their second one. In the last couple of months, we’ve seen wearables companies like FitBit and Jawbone begin to release devices that have more sensors and more refined algorithms for making sense of biometrics data.

The inclusion of a heartbeat monitor will become table stakes for wearables in the next year. They will not all be accurate, but consumers will begin to expect them. Manufacturers are also beginning to build smartwatch-like features (like notifications from a paired smartphone) into fitness wearables.

They need to be comfortable

This may sound fussy, but it’s a big deal. You’re asking me to wear your amazing device around day in and day out, even while I’m at the gym or in bed. So the thing has to be light, and it has to be comfortable. Many wearables simply are not comfortable on the wrist.

For instance, I was initially impressed with the Microsoft Band, but I eventually stopped wearing it because it felt to hard and blocky around my wrist. Plus, the touchscreen on the Band becomes very unresponsive when covered with perspiration.

Smartwatch makers need to take a deeper look into the materials available for watch bands, and get more creative about selecting them.

They should look good

Many wearables that entered the market this year were focused on hitting a low price point. For instance Xiaomi’s Mi Band sells at $13. Aesthetic appeal hasn’t been a priority, and cheaper wearables often look plasticky and toy-like.

Meanwhile, top-end fitness watches and smartwatches are often big and blocky. Part of that is an aesthetic problem, but part of it is a technical one. As more and more features and functions are built in to these devices, larger component parts, such as bigger batteries, are needed to run them. That’s why they’re bigger.

The lithium ion battery won’t be able to deliver the capacity improvements needed to get us through the smartwatch revolution. Manufacturers need to double down on power supply R&D and come up with some viable alternatives. This problem alone could kill the smartwatch category.

The numbers need to add up to something

At any one time I’m using several health devices, like a connected scale, a nutrition tracker, a sleep tracker, and a fitness band. I can open all the apps that accompany these devices and peruse all the numbers they’ve collected about my steps, deep sleep minutes, my calendars, etc. But what do I do with this data?

Too often wearables generate a lot of data, but they don’t give me contextual information I need to put the data to use. A device might give me my resting heart rate, but chances are it doesn’t know enough about me to tell me when my heart range has entered the danger zone. A dangerous rate for me, after all, might be perfectly fine for an athlete in training, or a life-threatening rate for someone who is morbidly obese.

I hope that smartwatch makers in 2015 will think of better ways to wrap enough contextual information around the numbers to make it all meaningful.

The apps should be easier to use

Some of the apps that accompany wearable devices are so poorly designed that it’s difficult to even see and make sense of the data they’ve collected. If the app is hard to read and understand, the device is basically useless. Since biometrics data is often best presented in charts and graphs, the app must make careful use of screen space to display it properly. Often, this doesn’t happen.

The numbers show that the market for health and fitness wearables is set to expand rapidly in 2015. As one wearable brand CEO told me, the devices are so easy and inexpensive to make, and the market is so ripe, he’d be crazy not to sell health wearables.

Hopefully, pure competition will weed the weak players out of the market, and new, more feature-rich and comfortable wearables will emerge.