#14. Experiment 3: What insights I can get from my data

After I chose the thesis topic — Life documentation, I wanted to explore three main aspects under this topic.

• How to record our personal lives well?
• How to present our records well?
• How to use our records well?

So far, I explored first and second aspects with an AI voice journal and artistic record visualization. And now I wanted to explore third era- How to use our records well. And before starting it, I made a new hypothesis below.

Hypothesis

If my personal data is collected and analyzed, I can get insights from them to improve my daily life

This hypothesis seems true, because we have heard of this a lot. Especially, for health data, it’s clear that we can get our information easily with tracking app otherwise we don’t know. However, other than health data, I personally felt this concept is not tangible. People said that we can get insights from our data. But what actual insights I can get? I don’t know. So I decided to test myself with my data to see what benefits there are in analysis my data.

As a data type, I chose ‘photo’ that I took in 2016 and analyzed them since I thought photos could reflect my daily life and by analyzing them, unique patterns could be found.

Process

First of all, I collected all photos taken this year (Jan — Oct) and divided them by month. All photos are grouped around similar character. Based on this, I tried to find my patterns and see if there are any insights I can get from them.

Analysis and Self-testing

From superficial discovery of my photos data, I expanded my thinking. What further patterns can be discovered and where can this data be used potentially.

Result

After self-test, I came back to my hypothesis.

If my personal data is collected and analyzed, I can get insights from them to improve my daily life.

Well. I felt this hypothesis couldn’t be validated because it’s too broad. The types of personal data are varied. So I changed it as follows.

If my photos are collected and analyzed, I can get insights from them to improve my daily life.

I acknowledge this experiment and result is very subjective. However, the intention of this experiment was also to know if I FEEL I could get any INSIGHTS personally from my data. In that sense, I would say this hypothesis was wrong to me. Because, I could see my patterns and imagine potential usage but I don’t think these findings can improve my daily life.

Even though this hypothesis seems to be wrong to me, still I felt it couldn’t be validated because I missed an important factor in my hypothesis. That is, purpose. I realized when we talk about data and insight, we should narrow down which data type we want to use and what purposes is intended. For example, if I change my hypothesis to

If my photos are collected and analyzed, I can get more relevant personal recommendations based on my pattern indicated in my photos.

Then this hypothesis will be (possibly) right, or at least it can be verified. My original hypothesis was still vague and broad, so it cannot be verified.

Even though this hypothesis test ended ambiguously, this experiment revealed how to approach the concept of “Personal data” and “Insights” we commonly used as a designer. What we want to do with our data must be clear and a data type has to be chosen for the purpose. When this happens, we can get close to INSIGHTS which we want to get through our data.