MyFitnessPal recently sold to Under Armour for $475 million. In this exclusive excerpt from ReadWrite founder Richard MacManus’s new book, Trackers: How Technology Is Helping Us Monitor & Improve Our Health, he reveals founder Mike Lee’s secrets for building an app that has helped millions lose weight. Trackers is now available on Kindle and in other digital bookstores.
It was the summer of 2003 and Mike Lee, a 32-year-old Silicon Valley marketing executive, was looking forward to his wedding day later that year. The only trouble was he had put on weight over the summer and didn’t want to walk down the aisle with a pudgy tummy.
His fiancée also wanted to lose some weight before the big day, so the couple went to see a personal fitness trainer at their local gym, 24 Hour Fitness. At the end of the first session, after Mike Lee and his fiancée had sweated off maybe half a pound, the trainer handed them a small book. It contained calorie counts for about 3,000 foods. “You need to start counting your calories,” the trainer advised, tapping the book. “Write down everything that you eat in a notebook.”
Lee is a veteran of Silicon Valley. In the early 2000s, he’d worked for Handspring, a manufacturer of PDAs, or personal digital assistants, that went on to merge with Palm in 2003. PDAs were the precursor of the modern smartphone—some of you will remember the PalmPilot from the late 1990s—and so Lee had gotten an early entry into the world of mobile communications. His first reaction to being told to count calories on paper? There’s got to be a digital method!
So he immediately threw the calorie counting book away and looked for an online solution. Lee looked at over a dozen existing online calorie counters in the summer of 2003. There was no shortage of online solutions.
“But to my amazement,” Lee recalled a decade later, “none of them worked the way I thought they should work.”
The first thing Mike Lee wanted to change about 2003-era calorie counters was the user experience, which he found frustrating. Many foods were either not available in these products, and so had to be manually input, or if they were available then they were hard to find.
“When I was trying to use other products back then,” Lee told me, “the user interfaces were just really cringe-y.”
He concluded that it was actually easier to write down calorie counts on paper than to use these online products. The second thing he noticed was the lack of intelligence in the online calorie counters of 2003. A simple yet powerful notion, which would become a key feature of MyFitnessPal, was to remember the foods you eat the most.
“People tend to eat the same things fairly often,” Lee told me, adding that his own breakfast “doesn’t vary that much.” So he reasoned that an online calorie counter “should just remember what you eat most often and make that easy for you.”
Lee left Palm in October 2004. It was at that point that he began development work on MyFitnessPal.
“I started building the diet tracker I really wanted, because I finally had the time to sit down and build it,” he told me. It was also a good chance to brush up on his programing skills, being a long-time computing enthusiast and amateur programmer. “I had done a ton of programing when I was young,” he said. “I’d started at ten and programed all the way through high school and college. But it had been a long time.”
So it was that MyFitnessPal was born, as a side project in 2004 for a slightly overweight Silicon Valley executive. It was officially launched less than a year later, in September 2005.
MyFitnessPal Goes Mobile
In August 2009, after five years of part-time work building up MyFitnessPal, Lee finally decided to take the big step and quit his day job. His brother Al also joined him as a full-time employee of MyFitnessPal. It was to be a turning point for the young company.
It was also a key period for the consumer health technology market. As profiled in another chapter of Trackers, 2009 was the year that the Fitbit activity tracker began shipping. Other health-focused websites and apps began to appear too. The primary reason for the increased popularity of health apps like Fitbit and MyFitnessPal in 2009 was the smartphone.
The second half of 2008 saw two milestones on the Internet landscape, which ultimately propelled the smartphone into the limelight. Following the launch of the first-generation iPhone in 2007, in July 2008 Apple launched a second-generation model, the iPhone 3G, and—just as importantly—the App Store.
The first Android-powered smartphone to be released came just a few months later, in October 2008. So coming into the New Year, 2009, the market was primed for a new way to use the Internet: smartphone apps.
In an August 2009 blog post announcing that he and Al were devoting themselves to MyFitnessPal full-time, Mike Lee noted that they were already working on “an iPhone app which we modestly think will be the best calorie counting app in the app store.” The iPhone app was duly launched in December 2009. It led to a surge in popularity for MyFitnessPal. Like many health-related Internet products, MyFitnessPal was a perfect match for the smartphone.
“You want to be tracking when you’re actually eating,” explained Mike, “and that’s when you’re out and about. So, mobile was critical to us.”
How MyFitnessPal Tamed My Diet
I was a relative latecomer to MyFitnessPal. I started to use it in early 2013, when I began to follow a low-carb diet. I used the MyFitnessPal iPhone app to enter my food for a period of about three months. I focused mainly on counting carbohydrate intake each day, since that was what my new diet required. But I found myself interested in all of the different food data, for comparative reasons. For example, I could compare the calories that MyFitnessPal said I consumed with the calories that Fitbit said I expended. Every time I ate something, it took a couple of minutes at most to enter the data into MyFitnessPal.
I don’t think I would have used the product at all if it wasn’t for the mobile app. It would have been too much of a hassle for me to open my computer and enter the data multiple times a day. But I carry my smartphone around with me everywhere, so it was easy to track my food intake. That, in a nutshell, is why MyFitnessPal became so popular in 2010 and beyond: It was a killer smartphone app waiting to happen.
It became even easier when MyFitnessPal added barcode scanning to its Android app in November 2010 (the iPhone app got this feature in July 2011). The feature used the smartphone’s camera to take a photo of a product barcode, which the MyFitnessPal app would attempt to match to a product in its food database. If a match was found, which in my experience was more often than not, the food would be automatically added to your meal. In my discussions with Mike Lee, I was somewhat surprised to learn that the data doesn’t come from the food manufacturers. “We don’t go direct to the food manufacturers to get the food data, in most cases,” said Lee. Rather, it is crowdsourced. In other words, MyFitnessPal’s users upload the data and check it for accuracy.
One of the items I scanned into my food log every now and then was Kraft’s Philadelphia Regular Cream Cheese Spread. The calorie, carb, fat and other nutrition data for this product was member-submitted. It had “2 confirmations” when I last checked, meaning that two people had reviewed the data against the product’s label and deemed it correct. It’s easy enough to check yourself, since many countries legally require food manufacturers to have a nutrition facts label on the packaging. The one thing to be wary of is that some nutrition labels can be misleading or wrong. The US Food and Drug Administration states on its website: “Manufacturers are responsible for the accuracy of the nutrition labeling values on their products.” The FDA does random sampling for accuracy and it will prosecute violations, but it simply doesn’t have the resources to check everything.
The barcode-scanning feature led to more MyFitnessPal users adding data to an already impressive database of food information. MyFitnessPal’s food database has been critical to its success over the years. Mike Lee realized very early on that in a food tracking app, there’s nothing more frustrating for a user than not finding an item—because then they’d need to check the food packaging and manually input the data. What should be a quick one-minute update becomes a five- or even ten-minute time suck.
So top of mind for Mike Lee when he began developing MyFitnessPal in 2004 was building up a large food database. He began by manually adding the data himself, using product information from food manufacturers and retailers where available. In September 2007, he blogged: “Last night, I was able to add nutritional information for the entire Starbucks menu and all Amy’s Kitchen products to the food database. My goal is to more than triple the size of the current food database by the end of the year.”
But he also realized that MyFitnessPal would only truly scale if he called on his users to help—in other words, crowdsourced it by enabling MyFitnessPal’s users to input and check data. The alternatives were to get the data from open food nutrition databases or directly pipe it from food manufacturers. However, there wasn’t a comprehensive open database of barcodes that MyFitnessPal could tap into and getting the data from food manufacturers was far too much work.
Besides, if their own users entered the data then MyFitnessPal would own it. One of the defining characteristics of so-called “social” software—like Facebook and Twitter—is that the value of the business is almost all derived from the amount and quality of user data in its databases. MyFitnessPal is no exception, so Mike Lee was very smart to go the crowdsourcing route. Everything its users enter into MyFitnessPal’s food database belongs to MyFitnessPal.
Still, it was no easy task to build up the food database. If he wasn’t adding new foods himself, Mike Lee spent his time checking the accuracy on user submissions. It was a chicken-and-egg situation in 2005 when MyFitnessPal launched because there needed to be enough food data to grow its user base, but there was a danger that early users would be frustrated by lack of data and immediately quit. Lee got around this dilemma by letting users check data accuracy with just one click of a button. So gradually, from 2005 onwards, the database began to fill out.
MyFitnessPal has come a long way since 2005. Nowadays, with tens of millions of users, it has built-in quality control. “We have a million QA people now,” Lee said, chuckling … probably with relief that he no longer needs to check the data himself.
Different Strokes For Different Folks
When MyFitnessPal released its iPhone app in December 2009, it enabled people to track their food on the go. I asked Mike Lee what he’s learned since then about how to track food intake. The first thing a new MyFitnessPal user should do, he replied, is “track everything.” But it’s OK to estimate food data, he added. “It doesn’t have to be perfect. The most important thing is just keep tracking.”
He also advises not to worry about missing the odd meal.
“The more you do it,” he said about food tracking, “the easier it becomes and the more of a habit it becomes. Basically, the longer you stick with it, the more success that you’re going to have.”
Another tip Mike Lee offered is to enter your food before you eat, not after. That’s where the smartphone app really helps, because it takes a couple of minutes at most to enter this data. The benefit of doing it before you eat is that you can make adjustments if a food you’re about to consume is high in calories, or carbohydrates, or whatever your key metric is.
Some of the power users of MyFitnessPal use the app to plan their meals throughout the day, although many people will prefer less structure. Mike Lee himself prefers a more flexible approach.
“There isn’t some kind of rigid plan that we ask people to do,” he told me. “It’s not like you can never have a piece of chocolate cake again. It’s about making your own decisions about what you eat and understanding that there are big ramifications.”
In his own life, as a startup CEO, he often needs to go to industry events. Such events tend to serve nibbles along with alcohol and so are likely to push his calorie count up for that day. “So I’ll go lighter on lunch,” explained Lee, “or I’ll go for a run in the afternoon to try to burn off some extra calories.”
For Mike Lee, his app gives him an understanding of what tradeoffs he can make every day about his food and exercise mix. “It’s just really empowering,” he said. “You feel like you’re in control of what you want and goals that you’re trying to achieve and that’s kind of what I like about it.”
Ultimately the benefit of food tracking apps like MyFitnessPal is that they help you make better food choices, whatever flavor of diet you subscribe to.
“Everybody can benefit from eating better,” Mike Lee remarked. “Until, recently, we didn’t have the tools to make the [food] information we need easily accessible.” For example, Lee—who still uses calories as his primary measure—stopped using mayonnaise on his sandwiches after discovering how many calories it had. “One tablespoon of mayonnaise has 90 calories. Before I started calorie counting, I had no idea that was the case. Whereas mustard has five calories. So I just stopped using mayonnaise.”
This is the kind of food knowledge, says Mike Lee, that stays with you and helps you make better daily decisions.
It’s important to note that it’s not scientific knowledge that Mike Lee is referring to, but self-knowledge. The diet industry is one of the most confusing and contradictory around. When two diametrically opposed diets—low fat and low carb—both have science “facts” to back them up, it’s no surprise that most people have little clue which foods are truly healthy. In the final analysis, food facts may not matter that much. Mike Lee to this day emphasizes calorie counting in his own ongoing weight maintenance plan, even though counting carbs is more in fashion now. Whichever metric you use, monitoring your food intake will at the very least make you more mindful of what you’re eating and the impact it has on your weight.
There is no one diet to fit all. Each of us is different. What’s more, our circumstances change over time—from experience, I can tell you that getting type 1 diabetes will make you change your diet!
Ultimately, the only way you’ll find out which type of diet is effective for you is by testing them and tracking your progress. Whether you use a smartphone calorie (or carb, or salt, etc.) counter like MyFitnessPal, or whether you use good old paper and pen, it doesn’t matter, although the technology in MyFitnessPal does make things easier for you. The main point, though, is to test for yourself what works.
But how often do you need to use a tool like MyFitnessPal? From my own usage, I can attest that although it only takes five to ten minutes a day in total, it does take a conscious effort to do every update. Also, some updates are difficult to make, for example when you’re eating out and you don’t know exactly what’s in the foods you’re consuming.
Many people use MyFitnessPal intensely for a short period, then stop. I was one of those users. I began my low-carb diet in mid-March 2013 and started using MyFitnessPal at the same time. It helped me immensely in knowing how many carbs I was consuming every day, as well as avoiding eating foods that were high carb. After a few months of using MyFitnessPal regularly, I came to an understanding of what my average daily carb consumption was. At that point I stopped using the app. That’s because what I eat every day is fairly consistent. Even when I deviate from my eating routines, I know which foods to avoid now and so I don’t feel I need to enter the data into MyFitnessPal.
Mike Lee admitted that this kind of usage is fairly common for his app. Some users, he said, “use the app for a while, maybe they’ll hit their goal weight, for example, and they’ll feel like they don’t need to track anymore. But then, oftentimes they’ll come back a few months later for a little clean-up.”
MyFitnessPal sees a broad pattern of behavior, said Lee. “Some people calling it off, some people doing it every single day. It really depends on what the user is looking for.”
Even Lee himself doesn’t use MyFitnessPal for every single meal.
“I log most of the time,” he told me. “There are some meals that I’ll skip. But in those cases, I still use the general knowledge that I’ve gained from the app.”
On that last point, I differ from Lee. It’s not “general” knowledge he’s gained from tracking his foods on the app he created—it’s the specific knowledge of what is best for his own body.
Read more from Trackers, available on Kindle and in other digital bookstores.