Skip to main content
Announcements
Introducing Qlik Answers: A plug-and-play, Generative AI powered RAG solution. READ ALL ABOUT IT!
Michael_Tarallo
Employee
Employee

jason.pngHello Qlik Community! In this post, I’d like to introduce you to today’s guest blogger and Qlik enthusiast, Jason Yeung. Jason is our North East Solution Architect Manager who has over 15 years in the BI industry…flexing his muscles in Pre-sales, product management, consulting, and support. Jason wanted to get healthy, but traditional means of health data tracking weren’t enough. Continue reading to learn more about why Jason applied Qlik Sense to his fitness plan and how it was a better alternative to traditional fitness tracking methods.

This past year, I decided to get in shape.  Over the years, I went from being an avid runner to a car service, driving my children to hockey rinks on the weekends.  This year, things were going to change.  I had a strong plan, motivation, and allocated the right amount of time to be successful.  However, I needed an analytical tool to track my progress and keep me honest.

The Analytics Problem with Fitness Apps

Today, we have access to data around our personal well-being and there are data collection tools everywhere.  We wear them on our wrists, have them in our pockets, and have them built into our treadmills.  While these tools serve as great motivators, they all seem to fall short in 3 key analytical areas:

  • Data in, not data out.  These tools do an excellent job collecting large amounts of data, but tend to fall short when it comes to analytics. They often provide summary level dashboards that just show fitness activity over time, but with limited abilities to explore the data further.
  • Lack of self-service.  My fitness “dashboards” are never quite the way I want them.  Everybody uses slightly different metrics and different ways to visualize their fitness activity and it seems we are limited to static dashboards that can’t be modified easily.
  • I want “other” data.  No one fitness tracker really tells the whole story because they’re all used for specific purposes.  Additionally, these “apps” are all isolated from one another with no unified way to bring the data together, especially data from other sources. This is important because when it comes to physical fitness, we tend to only see the data relationships and associations that we want to see.

Enter Qlik Sense

At Qlik, I’m constantly working with customers and prospects to build and develop innovative analytical solutions to solve their real-life complex information needs.  Qlik Sense provides these organizations with an analytical tool that enables them to build dynamic and interactive dashboards from many different sources of data.  As you can see, this is the same problem that I’m trying to solve, so I put Qlik Sense to the test.

What I did
Here’s a two view dashboard that I built to track my fitness activity.

1.png


At first glance this may not seem to be any different than other fitness dashboards, but the fundamental differences are as follows:

  • Data from four different data sources. RunKeeper, Fitbit, Weatherbug, and my personal data entry spreadsheet.  Therefore, no more isolated views of data.
  • No modifications or “massaging” of data needed. I was able to use the applications’ “out-of-the-box” functionality to export the data as-is and combine them together.
  • Personalization. I was then able to build my own custom visualizations tailored to how I wanted to see them.
  • Speedy. Based on my knowledge of Qlik Sense, I was able to easily load the data and build the visuals very quickly, while gaining additional insight from the data of the various tracking devices.

The Biggest Insights Can be Simple to Discover

The previous dashboard view was extremely effective in tracking my fitness activity. But the dashboard itself really didn’t shed any new insights.  However, the additional data from Weatherbug did!  For runners, weather can be the single biggest motivational driver. In the visuals below, there are two main discoveries.

The weather varies from week-to-week, but my weekly running activity remained relatively constant:


6.png

But the “type” of day played the biggest role. I noticed that while my pace and distance doesn’t significantly change based on temperature, the outliers on the right suggests that I typically only run on “Clear” days.

7.png

Telling the Story

After analyzing all this data, I needed to build a story to communicate the meaning behind the numbers. Using Data Storytelling, Qlik Sense allowed me to build a dynamic presentation with a narrative to highlight my overall activity. I wanted to share this information with my support structure, my family.


8.png

And the main point to communicate was:

9.png

The Result

What was the outcome of all this?  My wife bought me a windbreaker!

In summary, as you can see, analytics is all around us.  In addition to being a top priority across all organizations, it should be a top priority for individuals to better manage their health.  Using data to gather and build new insights allows us to be more productive, make better and more informed decisions, and hopefully live healthier lives. Being able to communicate your findings effectively increases your chances of getting the message heard and understood. I used Qlik Sense to unleash my intuition and so should you. Download Qlik Sense today for free.

Jason Yeung

Qlik

17 Comments
anguila
Partner - Creator
Partner - Creator

Hi Adam,


As this site it's not an sport community, basically, imho, this post is a "Look! It's soooo easy to make a stunning dashboards. I can even use my running data to wow you!".

So my comment is: "Yes, it's, but be careful.. It's also easy to screw it up".

If you want to argue about the "analytics" the running apps offers, I'm afraid I'm completely agree with you, they all completely sucks or are very poor..  This fact is not relevant here, though.

Regards,



P.D: I'm using Endomondo too, did you share your dashbord? I don't want to reinvent the wheel..

0 Likes
1,269 Views
Anonymous
Not applicable

The Endomondo based app was built in QlikView with the activity and lap data manually translated into a pair of spreadsheets. Not very high tech but it worked for what I needed. I tried to incorporate heart rate data as well but as I couldn't find consistent definitions and the data was intermittent I gave up on it. As it was always a work in progress you'll excuse me if I don't share the whole app but here is the main dashboard to give you an idea.

Endomondo Dashboard Crop.png

I am following a similar approach now with cycling data in Strava, and using a mixture of QlikSense and QlikView. With what I said above about beautiful design vs functional design here are the two types of output.

Qlik Sense:

Strava Dashboard Sense.png

QlikView:

20150322 Summary.png

1,269 Views
anguila
Partner - Creator
Partner - Creator

Cool - You're in shape

Thanks for sharing some screenshots. It's a pity not being able to download and play with it.

Regarding Endomondo data, you might find useful this extractor to gpx files: Endomondo Export - Home

Regards,

0 Likes
1,175 Views
Anonymous
Not applicable

If you can send me some contact details I'll happily share the app with you directly, just didn't want to publish it. abh@qlik.com

0 Likes
1,175 Views
Not applicable

I got a few questions around weather data.  While I'm sure that there are better and more scalable ways to do this, I chose the path of least resistance.

You can pull historical weather data from wunderground.com using a web file.  To do this

1.  Create a web file connection and add the following connection string.  You can change the time periods and the location in the URL.  In this example, I used 2014 data in Boston.  http://www.wunderground.com/history/airport/KBOS/2014/1/1/CustomHistory.html?dayend=31&monthend=12&y...

2.  You can then just load the data in as such.  Also, if you want to accumulate history, you can keep changing the years (e.g. yearend=2013) and do a concat load.

LOAD

EST as Date,

month(EST) as WeatherMonth,

year(EST) as WeatherYear,

[Max TemperatureF],

[Mean TemperatureF],

[Min TemperatureF],

[Max Dew PointF],

[MeanDew PointF],

[Min DewpointF],

[Max Humidity],

[Mean Humidity],

[Min Humidity],

[Max Sea Level PressureIn],

[Mean Sea Level PressureIn],

[Min Sea Level PressureIn],

[Max VisibilityMiles],

[Mean VisibilityMiles],

[Min VisibilityMiles],

[Max Wind SpeedMPH],

[Mean Wind SpeedMPH],

[Max Gust SpeedMPH],

PrecipitationIn,

CloudCover,

if(Events='','Clear',Events) as Events,

[WindDirDegrees<br />]

FROM [http://www.wunderground.com/history/airport/KBOS/2014/1/1/CustomHistory.html?dayend=31&monthend=12&y...]

(txt, codepage is 1252, embedded labels, delimiter is ',', msq, header is 1 lines);

1,175 Views
Not applicable

To answer David question and to echo Adam's response, this was never meant to be a visually stunning application.  In fact, it was meant to be the opposite - throw a bunch of data into this thing for personal consumption.  As a result, the look-and-feel is really not very important at all - it's the rapid time that I was able to go from data to insight was the #1 driver - which is what self-service is all about.

1,175 Views
DeepaliChakre
Partner - Contributor
Partner - Contributor

Hi @Michael_Tarallo  and @Anonymous ,

I am trying to connect to Fitbit from Qliksense to visualize my Fitbit data.

I have the access_token and the API calls are working fine from POSTMAN but Not with Qlik.

I get the below error when connecting:

HTTP protocol error 401 (Unauthorized): Requested resource requires authentication.

How did you pull the Fitbit data? Using rest connector?

Can you please help me out with my error? How can I resolve this?

 

0 Likes
244 Views