How I, a Programming Novice, Built a Coffee Tracking App with ChatGPT (Part 4)

~The Completed Personal App~

　Through this process of trial and error, the Dolce Gusto management app gradually took shape.
　The app is named “Capsule Log”

　I use it to keep records of my Dolce Gusto.
　In case I ever have the opportunity for others to use it in the future, I decided on this name so it could serve as a “management app for capsule coffee,” making it compatible with other capsule coffee machines like Nespresso, Keurig, or UCC.

　First, the central feature is the calendar screen.

　I set it up so that the number of cups of coffee drunk that day is displayed under the date.
　And when you tap a date,
“Record coffee for this day”
　a button appears, allowing you to open the recording screen from there.
　The entered information is then displayed as a list below the calendar.
“When and what did I drink?”
　I made it so I can check this immediately.
　I open the app and record it while the capsule is set in the machine and brewing.
　By making this a habit, I can manage:
● Types of capsules consumed
● Date and time
● Inventory

　Furthermore, I added a monthly summary display to the recording screen.

● Number of cups drunk this month
● Amount spent
● Average cost per cup
　...and so on, which can be seen at a glance.
　The cost is set up so that you enter the purchase price when registering a capsule in the management screen.
　It automatically calculates and records the cost per cup from the purchase price.
　This was also a specification suggested by ChatGPT.
　Since I started my Dolce Gusto life for the purpose of saving money,
“How much am I actually spending?”
　was a point I was quite concerned about.

　Additionally, I made it possible to display graphs for:
● Number of cups drunk by capsule type
● Consumption and purchase amounts
● Popularity ranking by capsule

　The graphs, in particular, were a part that was quite fun to make.
　Even though I think I am drinking various capsules evenly, it is interesting to see that it is actually quite biased when looking at the data.
　This kind of “visualization” is one of the parts that made me glad I built the app.
　ChatGPT also fulfilled my request to create such graphs.
　It accepted my thoughts—how to display the graphs, what information to include, etc.—and proposed ideas and wrote the code.
　While I was struggling with my thoughts, ChatGPT built it quite easily.

　And personally, I was quite particular about the management screen.

　Here, you can register capsules and manage inventory.
　I was especially particular about the “capsule color.”
　Since Dolce Gusto capsules have different colors for each type, I wanted to be able to tell at a glance when looking at the list.
　Since I can adjust even subtle color differences myself, it is fun to adjust the colors while looking at the capsules, thinking, “Maybe it’s a little brighter...”
　Furthermore, for Starbucks-brand capsules, I used a special design with half the icon in green.
　It fulfilled this request of mine as well.
　It would have been difficult and I likely would have given up if I tried to make it from scratch, but ChatGPT easily created the half-green circle icon.

　In the list of registered capsules, I made it possible to check the remaining number of cups.
　When you purchase more, you can enter it directly. Furthermore, capsules with zero inventory are moved to the archive. I also made it possible to enter data when re-purchasing from there.

　At first, I thought,
“Should I just delete the capsules I’ve finished?”
　but in reality, there are times when I re-purchase them.
　It took shape while repeating, “Maybe this way is more convenient” while using it.

　And at the very bottom, I added a backup function.
　It would be quite a shock if I changed my smartphone model or if the recorded data disappeared for some reason.
　I added a function where you can save data to a separate file and load it by pressing a “Restore from backup” button.
　I gradually added “features I would want if I were actually using it myself.”

　Thus, my personal capsule coffee recording app was completed.
　What started as just rough rectangles in PowerPoint gradually became a “useful tool.”
　Of course, building it was not easy.
Errors
Incomprehensible English
Endless settings
　Honestly, I thought “maybe this is impossible” many times. But because of that, the sense of accomplishment when it was completed was quite significant.

　And at the same time, I realized the greatness of the work involved in app development.
　The apps we use casually every day are built on top of massive amounts of code and error handling.
　Thinking about that, my view of system engineers and developers has changed quite a bit.

　However, there are things I noticed after actually starting to use it.
　No matter how convenient an app you build, it is ultimately a human who records the data.
“I’ll enter it later.”
　I think that and forget. I have probably already failed to record it several times.
　In fact, even though the capsules were gone, the app’s record showed an inventory of 2 (lol).
　To eliminate such human error, ideally, I would want the Dolce Gusto machine and the app to link up and automatically record the moment I brew coffee.
　For example, a mechanism where the capsule has a QR code, the machine reads it, and it automatically records,
“I drank this today.”
　...Technically, I don’t think it’s impossible. However, to realize that, my own efforts alone are not enough.
　I hope Nestle will release such a machine (lol).

　After all, the purpose of starting to build this app was not to release a perfect product to the world from the beginning.
“Can I, someone with no programming experience, really build an app?”
　It was a challenge that started with a feeling like a small experiment.

　Of course, I still don’t understand technical terms well. There are many times when I look at an error screen and don’t understand what is happening.
　Even so, ChatGPT acted as my teacher when I was in trouble.
　When I ask questions about things I don’t understand, it tells me what to do next.
　When I put into words the vague image in my head of “it would be convenient if there was a feature like this,” it thinks together with me about how to give it shape.

Looking back, perhaps what I needed wasn't advanced knowledge, but the desire to create something and the ability to communicate that idea to ChatGPT.
My 'Capsule Log,' which started as a rough rectangle drawn in PowerPoint, has now become a tool I use every single day.

ChatGPT is not a magic tool that gives you all the answers.
It is about thinking through what you don't understand together and giving it shape.
For me, it is my creative partner.

'Capsule Log' is the first project I completed with that partner.

(End)

I am submitting this work to the [Creative Awards 2026, Business Category].
I have compiled the serialized entries into a magazine.
Please read the rest, and I would appreciate your support through follows and comments.

#CreativeAwards2026 #BusinessCategory
#GenerativeAI #ChatGPT #AppDevelopment
#DolceGusto #Coffee
#Nescafe #Nestle