The mysteries of weight loss are on the way to be being solved. Fitbit is measuring heart rates, daily steps, and minutes active for millions.
Customers can set weight-loss goals and diet plans. Unlike its steps and activity metrics, which are captured automatically, a Fitbit user has to manully enter their daily calories eaten. Not perfect, but still useful for those interested in understanding how fitness and diet impact their weight-loss goals.
Millions of customer records provide Fitbit the data necessary to create precise models for customers looking to hit weight-loss goals with a combination of better dieting and increased activity.
I have mocked up a possible work-flow of how Fitbit could enhance the weight-loss goal experience. Currently Fitbit makes it easy to setup a weight-loss goal, but does not ask the user to set a target time frame in which to hit the goal. It is a matter of time before they add a time dimension to the goals.
Mockup for weight-loss goals
Note: I have no affiliation with Fitbit. These mock-ups are my own take on possible future implementations of their technology and have no bearing or insight in their actual product plans.
In Figure 1, the user would not only be able to set a weight-loss target, but also a target time frame in which to make the goal.
In Figure 2, Fitbit would model a plan for the user to hit their weight-loss goal. Here’s where the health records Fitbit has captured can be used to create highly effective recommendations. In Figure 2, the customer would be able to use the sliders to personalize the diet and activity targets. For example, one customer might not wish to reduce their daily calories eaten targets from their current baseline, and could slide the calorie reduction to 0%. Fitbit could then increase the activity and steps increases needed for the user to hit their weight-loss goals within the set time period.
In Figure 3, the customer could see how they are doing against on their weight-loss goal. The weight chart would show the model of the expected weight loss over time if the user adhered to the diet, step and activity recommendations set by Fitbit. The chart would also show the actuals of the user’s current weight, and if they were above or below the Fitbit recommendations.
Many users may not maintain the recommendations that Fitbit has made. At the end date of the goal period, Fitbit could calculate how long an extention to the campaign would be needed to achieve the weight-loss target.
The message might say:
“Congratulations, you have hit 58% of your targeted weight loss goal of 18 pounds for your 30 day campaign ending 12/3/2016. You lost a total of 10 pounds! Great job. If you continue at the diet, activity and step baselines you maintained during the 30 day campaign, our projections show that you will reach your targeted weight goal within 17 days. If you wish to extend the program for 17 more days, we can help you measure your progress.”
“Yes, let’s extend” or “No thanks, I’m good for now”
Other health targets could be modeled
Weight loss is a great starting point, since this it is easy to measure and a goal that many are interested in. There may be others who are interested in other health metrics such as resting heart rate. In a similar fashion to the weight-loss goal and model, Fitbit could target these other health goals.
Calories eaten is a weak link
As mentioned in the intro, calories eaten is a laborious manually-entered data set. Without an automated way to capture this data set, many will not know or bother with taking the time to enter this data. A work around is that Fitbit could ask the customer to provide a daily estimate for calories eaten and hold this as a daily constant. Once someone comes up with a more automated way to capture this data, the modeling for weight-loss can be even more precise. However, even with a close estimate of this variable, Fitbit should be able to produce reliable models and recommendations.
Running a race with a known finish line
Weight-loss has been a mysterious and frustrating battle for many. Crash diets and exercise programs have been tried and abandoned. The problem is that none of these practitioners could reliable guarantee lasting results to their customers. Accurate data modeling will demystify the weight-loss answers. Far more people will be motivated to take the necessary steps to lose and keep the weight off, provided they know that the finish line is in site, and not a mirage. Data capture and data insight continue to improve at break-neck speeds. The formula for successful weight-loss will soon be known.
