I love my video doorbell.  I won't mention the brand, but you have probably seen it advertised on TV.  What I like most about it is its' simplicity and ease of use out of the box.  I quickly understood exactly what it's features/capabilities were, and it delivered upon them with ease.  It took me about ten minutes to set up, and it has been working flawlessly ever since.  It has seamless firmware updates, easy app based controls, quality robust hardware, and works every time.  This is tech as it should be; and it was likely designed with those that hate complicated tech in mind.  

I have been utilizing and now selling cycling power meters for over 20 years, and I am just now seeing this small industry come around in the last few years to more consumers based thinking like my doorbell.  As a coach I worked with athletes a lot longer than I have owned a retail store, and tend towards a more customer-centric perspective.  I have witnessed former industry leaders failure to adapt, and the subsequent drubbing they took as a result.  Other brands have completely changed their product lines just in time to save themselves.   On the consumer side, power meters are still a relatively confusing product to the consumer, and I find many people base their purchase decision on features they may never use. 

Crank spider, crank arm, pedal, wheel, bottom bracket and frame are all places you can put a power meter, and they all have their pros and cons.  Some promise better accuracy, some offer more flexibility, more complex analysis, and still others an attractive price point.  It is important to note that I have listened to reps from various manufacturers tell me that it was "impossible" to put a power meter in a lot of these locations, and those that were attempting to were fools.  Some of these fools are now selling a lot of power meters.

If I had to stratify the power trained athletes I have worked with over the years, I would roughly put them into the following categories...

5%  Data geeks/junkies- Athletes that love to analyze their data to the nth degree and believe their athletic progress is hidden in the data.  They usually drive their coaches and themselves nuts.

30% Tech Savvy- These athletes will upload and analyze their data as they want to see the efficacy of their training.  They will use the data to effect their training but are not overwhelmed by it.  For these athletes a power meter is a solid investment.

45%  Tech Tolerant athletes will upload their data inconsistently.  They will use their power meters on the bike, but don't have the time or patience for complex analysis, or use a coach to give them feedback.  A power meter has value, but they will not spend a lot of time with it.

20%  Tech Adverse athletes had no business buying a power meter and probably did so because a friend or coach told them they needed it and it would make them faster.  It is basically an expensive bike computer that gives them a lot of confusing and arbitrary information they have no patience to analyze.

To the data geeks/junkies these percentages have no basis in fact and are simply my own rough estimates based on my own subjective experiences (please don't report me to DC Rainmaker).  My main point in displaying these metrics is that in my opinion manufacturers have marketed and designed their products for the top 5% for far too long.  There was a time when power meters were very expensive tools used only by the people willing to make the big investment in this new technology.  But as prices came down and new markets opened up that paradigm shifted, and some brands failed to adapt as I noted.    

Garmin invented the first multi-sport wearable computer.  It was not very good but it was the first.  Subsequent models got better and better, but they also added more and more features- seldom subtracting them.  What eventually evolved was a good device that was enormously over-complicated and frustrating to the user.  I would surmise that 90% of these features were/are unused by the wearer that is endlessly scrolling through screens trying to find the 10% they do use.  The thinking seemed to be that the more features you added the more people would buy it.  What it did was frustrate the user.  Sunrise alarm?  I have two of them called eyes.  And I have yet to find ONE person that is still using the swim metrics a few months after purchase.  But they sold and still sell a lot of them because people believe they need these features, and that they will somehow make them faster.    

Power meter users' suffered a lot of the same trials.  One of the most popular brands of days past used to require removal and shipping of the entire crank for a battery change.  But we have reached a point now where like my doorbell they are becoming simple, easy, and robust which is great.   

As a consumer you should not base your purchase decision on which power meter is the most popular or "best," it is how much tech you need, or need to pay for.  Do you really need 12 points of data analysis?  Is this better than 4?  If your power meter is 1% less accurate than another brand will this 1% have any affect on your training efficacy?  Take for instance left/right leg analysis, a feature found on more expensive power meters that have essentially two power meters- one on the left and one on the right.  This allows you to see power differentiation between your left and right leg.  The first thing to understand is that everyone, naturally, has a dominant or stronger leg.  It would be a very rare athlete that has 50/50 power distribution.  Beyond leg dominance there are a wide variety of factors that affect power distribution such as muscle imbalance, functional or actual leg length discrepancies, a variety of other biomechanical variations such as sacroilliac joint displacements or forefoot varus/valgus, as well as pedaling style.  Left/right analysis data only tells you that there is a variation, not necessarily what is causing it.  This information can be useful in the fit process and may indicate a flaw in pedaling mechanics- but that is just a starting point.  Addressing economy in any sport is a process that requires careful identification, then application of drills or other form improvement techniques.  This is a great place to for a coach to insert themselves and identify the things that can be worked on.  I have found that one of the best tools is a very expensive and complicated piece of technology called a mirror.  Once areas of opportunity are identified, simply setting your trainer up in front of a mirror gives you constant feedback of your pedaling form.  A left leg external deviation can be instantly corrected.  Adding a vertical laser line (available at your local home improvement store) is an excellent visual cue for your alignment and tracking.  Just like those swim metrics I mentioned above, left/right analysis rarely gives the athlete actionable data in my opinion; so why pay for it?

A better way to select a power meter is by choosing the features you need.  Flexibility is a key feature to consider, not just portability from bike to bike, but also what types of devices you can use for analysis.  I do like power meters that send both the Ant+ and Bluetooth signals.  All a power meter does is essentially send a series of torque readings, and then the device or software interprets and utilizes this data in various ways or with various algorithms.  There are a wide variety of cycle computers that pick up the Ant+ signal.  However, you can only effectively (and safely) monitor a few pieces of data while on the bike, and the athlete should choose the ones that are most actionable real time.  More complex analysis is of course appropriate after the workout or post competition.  Many Ant+ device manufacturers also produce a piece of software for this.  But the Bluetooth signal allows you to pick up power data on a smart phone or tablet.  This means your phone can double as a cycle computer, or allow you to adjust the features of your power meter via an app, or update the firmware, or use it in a variety of analysis or workout apps.  And an app is a lot easier to change/modify/improve than a piece of hardware such as the manufacturers propriety head unit.  Dual signal units offer a lot more flexibility over Ant+ only units. 

Because there is such variance in power meter prices I like to start with the customers rough price range first.  This may eliminate a lot of models right off the bat.  Compatibility is the next step as some power meters simply won't work with some bicycles or cranksets.  Cranksets in particular can be onerous as there as so many variations, and then you have a wide variety of bottom bracket incompatibilities to further confound the consumer.  It is important to research these things carefully before choosing a power meter, or it is often back to the drawing board.  Then we move on to features with questions such as do you want to use one device on multiple bikes?  Do you utilize different wheels for training and racing?  Do you want one device for training indoors and outdoors?  Do you like to train indoors with a particular app?  And don't forget your tech tolerance; some devices are more difficult to work with than others.  Lastly I steer them towards manufacturers with a proven track record of product availability and warranty support.  If there is an issue it should be rectified quickly, easily, and to the customers satisfaction.  Unfortunately I have not found that to be the case with some brands, others we have had almost no problems with and when they did occur they were handled promptly and effectively.  

As power meter prices have fallen the market has expanded; which has attracted a lot of new players.  This is generally good for the consumer as there are more options and downward price pressure.  But I don't recommend being the early adopter of new technology either.  Let a new brand get some seasons under their belt before purchasing.  Read the reviews and visit their website, this will tell you a lot about what kind of company they are.  Unless you are a data geek don't pay for features you will likely not use.  Simplicity, robustness, consistency and flexibility are more important than a .5% improvement in accuracy in my opinion.  Think doorbell.