Walking the Walk: Using Data Visualisation to Achieve Personal Goals

I’ve over-gone something of a transformation in the last year or so.

?

Back in March-23, I had an uncomfortable realisation that I was badly overweight and out of shape. At 28, I shouldn’t have been sat up at night with heartburn or gasping and gulping after walking up a flight of stairs. It goes without saying(!), but I wasn’t the svelte, lean machine I had been in my earlier career either... I was in an odd position, working a bit too hard, albeit in a sequence of jobs I loved, and taking any number of shortcuts at the end of the day. Food was fuel and I was a long way away from my days in Michelin starred kitchens.

?

I decided to do something about it, but in typical ‘Tom Croxford’ fashion, if I was going to do something, I was going to hold myself to account and make sure exactly how I was treating myself was visible and something I reflected on daily. For better or worse - I look at myself as a machine.

?

Now don’t panic, from here on in I will recap the exercise I undertook to set a target and manage my performance against it, exactly as a good project controls engineer would approach a new problem, KPI or conduct some analysis(!). So…

?

Setting a Target

The objective was to lose weight, but how much weight? Well, there seemed to be a reasonable number of information sources and assumptions I could make:

?

• At what weight had I previously felt my best?
• What did the NHS suggest I should be?
• Or just what arbitrary number would make me feel best?

?

Fortunately, I did have some historic data for what I used to weigh (which was useful), but largely I leant on the NHS’s (somewhat flawed) view of Body Mass Index (BMI) and backwards engineered a weight which would put me in the bottom 25% of the optimum range - 20 kg reduction from my then current weight of approximately 95 kg. Was it a SMART objective? Truly, no. I didn’t have a time measure, mainly as I had no concept of how quickly I would be able to lose weight and keep it off. Out of intimidation, I dodged that commitment (there is probably another interesting topic in assumptions and unknowns affecting target setting to investigate one of these days!).

?

Identifying my Data Sources

?

With my target now set, I could begin to map out what data I thought I needed:

?

• My current weight, both starting and at intervals over time
• My target weight

?

Fundamentally, those two data points tell me where I am in relation to the target. But… So what? How do I know if I’m heading in the right direction until I get to a point of taking the measurement? And how do I know what has caused it?

?

To that end I, for the second time in this series, revisited my GCSE biology and how our bodies convert food into energy (measured in kilo-joules or calories), that energy then either being expended by our activity, or stored for use later as fats. If I was going to have an ongoing measure to indicate if I was on the right path - ideally not as depressing as measuring my body mass each day - that might be a good place to start.

?

Therefore, I identified a need to capture:

?

• My energy intake (as calories)
• My energy expended (as calories)
• How great my daily calorie deficit would need to be to meet my overall target weight in a reasonable timeframe

?

With that in mind, I could compare these lists and start to map out what sources I would gather them from:

?

• Starting weight and current weights could be gathered from the trusty Salter scales gathering dust in the bathroom
• My target weight had already been gathered based on the investigation into BMI and other healthy weights
• My energy expended could be extracted from the data already collected through my Apple Watch and iPhone (also offering me the opportunity to trend this historically)
• Energy intake would be a little harder, and more effort than everything I have been through so far... There are a number of calorie counting apps available, I went with ‘Lose It!’, but much to the irritation of my long-suffering partner, I still needed to get in the habit of recording everything I consumed in the day. However, I am nothing if not committed and analytical, and once I was locked into the habit, it took a lot longer to stop than it did to get started!

?

So, I had a target, I had my data… What was I going to do with it?

?

Configuring a Meaningful Report

?

Reflecting on my article last week about reporting requirements - what question was I hoping to answer? Fundamentally I wanted to know:

?

1. Am I above or below my target weight?
2. Does my level of activity, and the food I consume, support my goal to be in a calorie deficit?
3. Given #2, what is my weight forecast to be?

?

For ease, I chose to co-locate my data entry and the data visualisations, with a visualisation specifically to answer the above questions. Those visualisations were:

?

Calories Expended and Consumed Area Graph

?

Answering Question #2, this was an area chart for my calories consumed (red) and calories expended (green), with the expended data series brought to the forefront. The logic with this was that at a glance I could not only see the trend of how many calories I was burning daily, and whether this was trending up or down, but any flashes or peeks of red above that green area would indicate I over consumed and was in a calorie surplus for the day. Beyond simply looking at the surplus or deficit (see following heading), this also allowed me to see if an ‘underperformance’ was down to a lack of exercise, increased food intake, or a bit of both.

?

Daily Calorie Surplus or Deficit Column Chart

?

A slightly different, simplified way of answering Question #2, this chart plotted my daily calorie surplus of deficit, allowing me quickly to single out the dates in which I overconsumed (far more naturally with a column chart). Principally it made it easier to highlight clusters of dates where I consecutively was in a calorie surplus (inevitably when I was travelling for work, we were on holiday or visiting family). As I had the data available to me, I also included a cumulative calorie surplus / deficit curve to keep how far I had come visible to me, and ensure even if 'performance' flattened, it would not begin to be undone.

?

Actual Weight vs Target Weight vs Anticipated Weight Line Graph

?

The most ‘back-end involved’ of the visualisations, this one required a bit of working out based on daily calorie surplus or deficit which had been calculated.

?

According to Healthline, if you have a calorie deficit of 3,500 over a week, a typical person will lose 0.5 kg in body mass. A little backwards engineering of this led me to identifying that a single calorie in deficit will contribute to a reduction of 0.142857142857143 g in body mass (let’s say 0.143 g for ease). By multiplying my daily calorie surplus or deficit against these, I was able to calculate my daily anticipated body weight change and summarise this into a cumulative position daily. This allowed me to see where my body mass was expected to be, and by the time I took a true weight reading, I could also see how accurate this had been.

?

Interestingly - this identified that my body mass reduction was always more than what is expected for a ‘typical’ person, indicating I hadn’t quite lost of all the fast metabolism that had made me a skinny boy at school!

?

Conclusion: Did Using that Report make a Difference?

?

Well… I really did use that report every day. And I really did lose 22 kg (3.5 stone in old money) across a 12 month period. And I do believe that seeing the trends, performance against rolling averages and whenever performance slowed or halted really did make a difference in maintaining my focus - 'you manage what you measure' after all. I ultimately chose to stop capturing data and measuring myself in that way - although 6-months later I feel like the rigour might be something I need to pick up again in the future...

?

All of which is irrelevant to the point that my own, personal data, 80% of which was already being collected enabled me to set targets, hold myself to account and visualise my progress - as I started this article off, walking the walk.

?