"We lose 4 hours a week copying stuff from one Excel file to another." That's what I hear most when a small business owner contacts me about automation. Not "we want AI" or "we need digital transformation." Just: we waste time on repetitive work and we'd like it to stop.

The problem is nobody knows what it costs. Automation agencies sell "process audits" at EUR 3,000 before writing a single line of code. No-code platforms promise "everything without a developer" — until the day it isn't, and the Zapier scenario loops at 2 AM generating 47,000 emails.

Here are five automations I delivered over the past two years. Real prices, real timelines, and what went wrong.

1. Automated competitor price monitoring — EUR 480

The need: an auto parts e-commerce owner manually checks prices from 3 competitors across 200 references every Monday. Three hours of copy-paste into a spreadsheet.

What I delivered: a Node.js script that scrapes 3 competitor sites in parallel, extracts prices, and emails a comparison table every Monday at 7 AM. Hosted on his own server via cron.

Time spent: 1 day. The longest part: understanding each site's HTML structure and handling out-of-stock items (missing price vs zero price).

What went wrong: after 3 months, one competitor redesigned their site. The scraping broke silently — the script ran without errors but returned zero prices. I added an alert when more than 10% of prices come back as zero.

Estimated ROI: 3h × 4 weeks × 12 months × EUR 20/h = EUR 2,880/year. The script paid for itself in 2 months.

2. Automated quote follow-ups — EUR 750

The need: a carpenter sends 15-20 quotes per month, follows up on maybe 3. Conversion rate: 18%.

What I delivered: an integration between his quoting tool and a PHP script that checks daily for quotes sent more than 7 days ago without a response. The prospect gets a personalised follow-up email. No response at day 14: second follow-up. Day 30: marked as lost automatically.

Result: conversion rate went from 18% to 27% in 4 months. On an average order of EUR 3,200, that's roughly 2 more converted quotes per month — about EUR 6,400/month in additional revenue. For EUR 750.

3. Multi-platform stock sync — EUR 2,200

The need: a home decor shop sells on WooCommerce, Etsy, and in-store (Lightspeed POS). Stocks aren't synchronised. Result: regular oversells, unhappy customers, and an employee spending 1 hour daily correcting quantities by hand.

What I delivered: a Go service that listens to webhooks from all 3 platforms and updates stock in real-time. When a sale happens on Etsy, stock drops on WooCommerce and Lightspeed within 30 seconds.

Time spent: 4 days. Etsy API is decent, WooCommerce REST API too. Lightspeed was the nightmare: approximate documentation, aggressive rate limiting, and a bug on their side with in-store sale webhooks that required a 5-minute polling backup.

4. Monthly PDF reporting — EUR 1,400

The need: an accounting firm generates monthly activity reports for 35 clients. A partner spends 2 days per month copy-pasting data from their accounting software into Word, exporting to PDF, and emailing. 35 times.

What I delivered: a PHP script that connects to the accounting software API, extracts key metrics per client, generates a clean PDF with the firm's logo and trend charts, and emails everything. Runs automatically on the 5th of each month.

5. Lead qualification bot — EUR 4,800

The need: a real estate agency receives 80-100 enquiries per month. A salesperson reviews each one, asks the same 5 questions by phone (budget, location, timeline, property type, financing), and qualifies the lead. 60% aren't serious.

What I delivered: a multi-step form that asks the 5 qualification questions before human contact. Responses are scored automatically. Score > 7: salesperson notified by SMS immediately. Score 4-7: automated follow-up email with matching listings. Score < 4: generic email with search link.

Result: the salesperson went from 15h/week of qualification to 4h. Hot prospects are contacted on average 12 minutes after enquiry instead of 48 hours. Conversion rate rose from 8% to 14%.

What all 5 projects have in common

None required AI, machine learning, or "digital transformation." It's classical code — APIs, scraping, crons, emails. The value isn't in technical sophistication, it's in understanding the business process and automating it without distorting it.

A custom script costs more upfront. But it belongs to you, runs on your server, and does exactly what you need — not what the template allows.

Conclusion

Automation isn't a big-company luxury. At EUR 480 for competitor monitoring or EUR 750 for quote follow-ups, it's the best ROI most small businesses can get from their digital budget. The condition: a developer who understands your business, not a generic tool that promises to do everything without code.