On March 1, 2026, I ran a personal experiment: 30 days, log every transaction by voice, no preset budget, no rules, no savings goal. Just observation. I'm a 38-year-old Cairo mother working part-time with two primary-school kids. Combined household income is around 22,000 EGP/month. Here's what happened.
I picked voice specifically because manual tracking apps had failed for me before — the longest I lasted was nine days. The reason was always the same: I don't have the bandwidth to type a number and pick a category while walking out of the supermarket holding two bags. Voice logging changed that. 'Spent 380 on groceries at Carrefour' — done.
Week one — the first shock
I logged consistently. After day seven, the total was 4,200 EGP. Multiplied by four weeks: 16,800/month. I assumed our spending was 13,000–14,000. The 3,000 EGP gap was a problem.
Friday alone was 950 EGP. Nothing big — 60 coffee, 380 small grocery run, 120 lunch delivery, 90 Uber, 200 pharmacy, 100 kids' snacks. Each took 5 seconds to decide. Four Fridays a month would be 3,800 EGP. And it isn't only Fridays.
Week two — the surprise category
By week two, the pattern emerged. The leak wasn't rent (3,800 fixed), wasn't school fees (1,500 fixed), wasn't groceries (~1,200/week). It was 'small orders + food outside the home.' Over fourteen days: 1,820 EGP. Roughly 130/day.
What surprised me wasn't the number. It was that I had estimated this category at 800–1,000/month. Reality was 3,640. A 2,640 gap. The money didn't disappear — I just wasn't seeing it. Each transaction had its own justification in the moment.
Week three — the first conscious change
I tested a sub-experiment: one full week with no delivery and no coffee outside the home. Result: 2,800 spent, vs. an average week of 3,200. A 400 EGP weekly delta = 1,600/month. Not the full 2,640, because I didn't kill everything. But half.
What I learned: tracking alone makes you see. Seeing makes you choose. Choosing changes behaviour. It isn't about willpower — it's about clarity.
Week four — fixed costs that aren't fixed
Last month's electric bill: 380. This month: 510. A 130 EGP increase I never noticed. Internet was 200, became 240 three months ago. School fees went from 1,500 to 1,650 mid-year. Each small (10–15%), but together 320 EGP/month — 3,840/year — that vanishes silently into 'fixed' costs.
The total — what I uncovered in 30 days
What I changed after 30 days
- Cut delivery + outside food by 50% — target 1,800/month. Saves 1,840.
- Killed two forgotten subscriptions (Apple Music + a closed game). Saves 180.
- Switched kids from school bus to a shared neighbour ride. Saves 400.
Total potential monthly saving from three simple decisions: 2,420 EGP. Not theory — concrete actions actionable from next week.
5 lessons from a 30-day experiment
- Tracking alone is 80% of the fix. You don't need a complex budget — you need to see.
- Money doesn't leak through big decisions. It leaks through 200 small ones a month.
- 'Fixed' costs aren't fixed. Bills increase, subscriptions renew, fees adjust. Track them.
- Intuition vs. reality is a wide gap. I was sure of my numbers. I was wrong on every category.
- Voice tracking works because it removes friction. I quit every manual app at 9 days; I lasted 30 with voice.
What tool did I use?
Flosyfeen (Flosyfeen.com), an Egyptian web app. It worked because: it runs in the browser without install, it understands full Egyptian colloquial Arabic, it doesn't ask for bank credentials, and logging takes 5 seconds. Any voice tool with deep Egyptian Arabic would work the same way. The tool isn't the point — the decision to see is.