Concept
1. The basic formula
Yearly time saved = (manual time − automated time) × runs per year
Yearly cost = maintenance hours per year
First-year gain = yearly time saved − build time − yearly cost
Break-even = build time ÷ (time saved per month)
2. Worked example
A weekly sales report takes 45 minutes by hand. Automated, it takes 5 minutes (run + check).
| Item | Value |
|---|---|
| Saved per run | 40 min |
| Runs per year | 52 |
| Saved per year | 40 × 52 = 2,080 min ≈ 34.7 hours |
| Build time | 8 hours |
| Maintenance per year | 4 hours |
| First-year gain | 34.7 − 8 − 4 ≈ 22.7 hours |
| Saved per month | 34.7 ÷ 12 ≈ 2.9 hours |
| Break-even | 8 ÷ 2.9 ≈ 2.8 months |
Worth it. After year one, you save about 30 hours every year.
3. Build it in Excel
| A | B | |
|---|---|---|
| 1 | Manual time per run (min) | 45 |
| 2 | Automated time per run (min) | 5 |
| 3 | Runs per year | 52 |
| 4 | Build time (hours) | 8 |
| 5 | Maintenance per year (hours) | 4 |
| 6 | Hours saved per year | =(B1-B2)*B3/60 |
| 7 | First-year gain (hours) | =B6-B4-B5 |
| 8 | Break-even (months) | =B4/(B6/12) |
Multiply hours by an hourly cost (salary ÷ working hours) to show it in rupees.
4. Quick rule-of-thumb table
How much time you can spend building, if it pays back within one year:
| Time saved per run | Daily | Weekly | Monthly |
|---|---|---|---|
| 1 min | ~4 hours | ~1 hour | 12 min |
| 5 min | ~21 hours | ~4 hours | 1 hour |
| 30 min | ~125 hours | ~26 hours | 6 hours |
"Daily" assumes about 250 working days a year. Small daily tasks add up fastest.
5. Hidden costs people forget
- Debugging usually takes as long as the first build.
- Changes in input: a new column in the export or a renamed sheet breaks scripts.
- Learning time if it's your first macro or script.
- Handover: documentation so someone else can run it when you're on leave.
A safe habit: multiply your build estimate by 1.5–2.
6. Benefits that aren't time
- Fewer errors — a wrong figure in a GST return or salary sheet can cost far more than the hours.
- Consistency — the report looks and works the same every time.
- Speed — results at 9 AM instead of after lunch.
- Scale — the same script works for 5 branches or 50.
If a task is error-prone or high-stakes, automate even if the time ROI is borderline.
7. When NOT to automate
The task happens once or twice a year, the process is still changing every month, or nobody will be able to fix it when it breaks. Write a checklist instead.
Common mistakes
Counting only build time. Ignoring the cost of errors. Automating a process before it's stable.