Manual EFT files do not look expensive. Nobody writes a cheque for them. The cost is spread across a few hours each week, a few corrections each month and the occasional rejected file, and it never shows up as a line in the budget. That is exactly why it is worth writing down.
This article works through the cost of building EFT 1464 payment files by hand versus generating them from Zoho Books. Every number below is an assumption you can replace with your own. We are not reporting measured results from customers; we are showing you the arithmetic so you can run it for your team.
Where the three hours go
Ask anyone who prepares Canadian vendor payments in a spreadsheet or a bank template and the weekly routine looks like this:
- Export or copy the list of approved bills from Zoho Books.
- Look up each vendor's transit, institution and account numbers in a separate file.
- Type or paste amounts, dates and vendor details into the bank's template or a fixed-width file.
- Check the file: record lengths, leading zeros, totals in the trailer, the file creation number.
- Upload to the bank portal and wait for the acknowledgement.
- Record each payment in Zoho Books by hand so bills show as paid.
- Fix anything the bank rejected and repeat the relevant steps.
None of these tasks is hard. Together they take time, and steps 2, 3 and 6 are where typing errors happen. Three hours a week is a reasonable starting assumption for a team paying a few dozen vendors a week. Yours may be one hour or six; the method is the same.
A worked example, with the assumptions stated
Here is the baseline for a small finance team. Change any line to fit your own situation.
| Assumption | Value used |
|---|---|
| Time spent per week on manual EFT file preparation | 3 hours |
| Working weeks per year | 48 |
| Fully loaded hourly cost of the person doing it | $40 CAD |
| Zefty subscription used in the comparison | Starter, $29/month |
| Time still needed per week with automation (review, generate, upload) | 30 minutes |
Annual hours, manual: 3 hours × 48 weeks = 144 hours
Annual labour cost, manual: 144 hours × $40 = $5,760
Annual hours, automated: 0.5 hours × 48 weeks = 24 hours
Annual labour cost, automated: 24 hours × $40 = $960
Annual subscription: $29 × 12 months = $348 (Starter plan, CAD, taxes extra)
Total annual cost, automated: $960 + $348 = $1,308
Difference: $5,760 − $1,308 = $4,452 per year, or 120 hours returned to the team.
Under these assumptions the subscription pays for itself if it saves a little over 43 minutes a month ($29 ÷ $40 ≈ 0.73 hours). Everything beyond that is time you can spend on collections, cash forecasting or client work.
What if the assumptions are more conservative?
Suppose automation only removes half the manual time instead of five sixths, and the person costs $30 an hour:
- Manual: 144 hours × $30 = $4,320
- Automated: 72 hours × $30 = $2,160, plus $348 subscription = $2,508
- Difference: $1,812 per year
The gap narrows, but the direction does not change. The reason is simple: the subscription is a fixed cost and the manual work scales with the number of payments.
What if you are on the Team plan?
Replace $348 with $516 ($43 × 12) and add $14 per additional user per month if you need more than one seat. In the baseline example, a two-user Team subscription costs $684 a year and the difference is still above $4,000.
What actually changes with automation
The hours drop because the tasks that consume them are either removed or shortened.
| Task | Manual | With Zefty |
|---|---|---|
| Getting the bill list | Export or copy from Zoho Books | Synced nightly and on demand |
| Vendor banking details | Separate lookup, retyped each time | Read from the vendor record in Zoho Books, encrypted at rest |
| Building the file | Type into a template or fixed-width layout | Select bills, click Generate Payment File |
| Format checking | By eye, or by trial and error at the bank | Records built to CPA 005 / EFT 1464 layout; blocked rows flagged in red before you start |
| Partial payments | Manual arithmetic on the balance | Enter the amount in the Partial Payment column |
| Recording payments | Enter each Vendor Payment by hand | Posted back to Zoho Books automatically |
| Audit trail | Whatever you saved | PDF report of every A, C and Z record, plus History with Zoho Payment IDs |
The upload to the bank portal remains a human step. So does deciding which bills to pay. Automation does not remove judgement; it removes retyping.
Costs the arithmetic does not capture
The worked example counts hours. Some costs are harder to put a number on, so we list them without inventing one:
- Rejected files. A file the bank refuses means the payments did not go out. Vendors are paid late and someone spends part of the day finding the misaligned field.
- Key-person risk. If only one person knows how the template works, their vacation is a payment problem.
- Late payment consequences. Missed discounts, strained vendor relationships, and in some contracts, interest.
- Reconciliation time. When the file and Zoho Books disagree, the difference has to be found at month-end. That time is rarely counted as "EFT preparation" but it belongs there.
If any of these has happened to your team in the past year, add a line for it in your own version of the table.
Run the numbers for your own team
The formula is short enough to do on paper:
Annual saving = (manual hours per week − automated hours per week) × weeks per year × hourly cost − annual subscription
Three tips for an honest estimate:
- Time the manual process once. Most teams underestimate it because the work is spread across interruptions.
- Use fully loaded cost, not salary alone. Benefits, software and overhead are part of what an hour costs.
- Include the correction time. A file that needs fixing twice a quarter is part of the routine, not an exception.
If the result is small, you have a cheap process and should keep it. If it is large, the subscription is the smaller number.
Key takeaways
- Manual EFT file preparation costs labour every week; automation costs a fixed subscription.
- Under the stated baseline (3 hours a week, $40 an hour, 48 weeks, Starter plan), the difference is about $4,450 and 120 hours a year. Your numbers will differ; the method will not.
- The subscription breaks even at roughly 43 minutes saved per month at $40 an hour.
- The tasks that disappear are the ones where typing errors live: vendor lookups, fixed-width layout and manual payment entry.
- Rejections, key-person risk and reconciliation time are real costs the example does not quantify.
For a criterion-by-criterion view of the two approaches, including compliance, audit trail and who does what, see our comparison of Zefty vs manual EFT files.