I Turned My Kobo Reading Data into a Dashboard (and Discovered I’m a Night Owl Reader) - chap 1/2 : the workflow.

From raw SQLite files to Streamlit dashboards, with a bit of Python.

We carry our books everywhere, but rarely do we see the data behind our reading habits.
I was wondering: How much do I really read? Which authors do I binge? Am I more of a sprinter or a marathon reader?

So I decided to hack my Kobo e-reader by installing KO-reader.
Spoiler alert: every page turn is quietly logged in a humble SQLite database, and that’s where my reading story begins.

All screenshots in this article come directly from my dashboard.

---

Step 1 – Getting the data

Every time I plug my Kobo into my laptop, I copy its statistics.sqlite file.
This little file knows everything:

• What books I’ve loaded

• How many hours I’ve read

• Where I stopped last night at 1:43am

• And the exact moment in the year I decide it’s time to reread Harry Potter

Streamlit + Plotly combo 🔥

Step 2 – Building the pipeline

I wrote Python scripts that:

1. Cleans the raw data

2. Builds Parquet tables:

   • fact (reading sessions)

   • dim (books metadata)

   • agg (summaries per dataviz)

3. Updates incrementally (so I can rerun it whenever I sync new data).

Turns out I read a lot more than I sleep, productivity goals, redefined.

Step 3 – Filling the gaps with AI

Not every book comes with nice metadata. Some don’t even have a proper genre.
So I used a LLM to enrich the data:

• Detecting book categories / genre

• Guessing missing publication years

• Completing author information

Since this can be the slowest and most costly part (LLM calls!), I designed it to run incrementally (only new books and missing metadata are processed).

This saves tokens, API costs, and execution time

Suddenly, my dashboard could tell me: “You read more sci-fi in winter, and fantasy in spring”.

Social life goes into hibernation in December

Step 4 – Visualizing with Streamlit

I built a Streamlit dashboard where I can:

• See a timeline of all my reading

• Check how long it takes me to finish a book

• Compare authors and genres

• Track my average reading speed

It’s like Goodreads, but powered by my own data pipeline : no ads, no recommendations, just my reading habits, raw and visualized.

After 10 PM, you’ll probably find me lost in the latest Cosmere book

Lessons learned

• Even a tiny SQLite file can grow into a full data engineering workflow

• Incremental pipelines make a huge difference when you update frequently

• LLM is a surprisingly good ally for enriching messy real-world data

• Visualization isn’t just for boring work dashboards, it can reveal patterns everywhere

What’s next?

• Deploy the dashboard online

• Write the chap 2 of this article (next week 😱)

• Make it plug and play for the e-reader community ❤️

Until then, I’ll keep reading, and my Kobo will keep logging every page.