In part 1, we covered what can run without human intervention in an online store – from invoices to emails. Now let's look at the tools, costs and how the transition actually works. Plugin, Make/n8n or custom solution? A model ROI calculation. And most importantly: when automation doesn't make sense.
Key Takeaways:
- Three automation paths: plugin (cheap, limited), Make/n8n (fast, for smaller stores), custom (flexible, for specific workflows)
- Custom e-shop to accounting integration costs 20–60k CZK and pays back in 4–6 months
- Transition happens gradually with parallel operation – no downtime, no "big bang"
- Automation doesn't pay off with fewer than 5 orders per day or unstable processes
- For what exactly to automate, see part 1
Plugin, Make/n8n or Custom Solution – When to Use What
Not every store needs a custom solution. And not every store can get by with a plugin. There are three paths and each has its place.
Plugin / connector
- Cost: 0–5,000 CZK/year
- Deployment: hours
- Flexibility: limited (what the plugin supports)
- Maintenance: depends on the plugin vendor
- Best for: standard processes where a ready-made solution exists
Make / n8n
- Cost: 5–15k CZK for setup + Make license ~500–1,500 CZK/mo
- Deployment: days
- Flexibility: medium (visual builder, no code)
- Maintenance: self or consultant
- Best for: smaller store (up to ~20–30 orders/day), simple integration
Custom solution
- Cost: 20–70k CZK one-time
- Deployment: weeks
- Flexibility: unlimited, full control over infrastructure
- Maintenance: service agreement with developer
- Best for: larger store, specific workflow, B2B
Make / n8n – When It's Enough
Make (formerly Integromat) and n8n are visual automation platforms – you connect systems without writing code, by clicking. For a smaller WooCommerce store with simpler requirements, this can be the fastest and cheapest path.
Example: WooCommerce → Make → FlexiBee. A new WooCommerce order triggers a Make scenario that calls FlexiBee's REST API via the HTTP module and creates an invoice. Setup in a few hours, consultant cost 5–15 thousand CZK + Make license. Make doesn't have a native Pohoda module – for Pohoda you'll need a custom solution or workaround.
When it's no longer enough:
- Higher order volume (Make has operation limits – at 100+ orders/day you hit paid plans costing thousands monthly)
- More complex logic (conditional discounts, B2B pricing, multiple warehouses, multi-channel)
- Need for 99.9%+ reliability – Make occasionally goes down, you don't have full control over infrastructure
- Custom PHP platform without a ready-made Make connector
If you have WooCommerce, up to 20–30 orders daily and standard B2C processes – try Make first. When it's no longer enough, come to me.
Indicative Pricing for Custom Solutions
- E-shop ↔ Pohoda (orders, invoices, stock): 20–60k CZK – simple B2C vs. B2B with variants, multiple VAT rates, multiple warehouses
- E-shop ↔ FlexiBee: 15–40k CZK – FlexiBee has a REST API, significantly simpler than Pohoda
- E-shop ↔ Money S3: 25–70k CZK – Money S3 has no API, only XML import/export, more work
- Balíkobot integration (carriers): 10–25k CZK – depends on number of carriers and rule complexity
Prices are indicative for one-time development. Plus a service agreement for maintenance and monitoring.
Rule of thumb: If your store works exactly as the plugin maker intended – use the plugin. If you have a simple WooCommerce store with standard processes – try Make. Once you need anything non-standard (custom invoice numbering, specific order status mapping, connection to an internal ERP), go custom.
B2B online stores almost always have specific requirements: individual pricing per customer, invoice payment with terms, orders by company ID with automatic registry checks, bulk orders. Here, you won't get by without a custom solution.
Automatic Product Translations for International Expansion
Problem: Expanding to Slovakia, Poland or Germany? Thousands of product descriptions, categories, filters. Manual translation is unrealistic, a translation agency expensive.
How It Works
- AI translation (Claude, DeepL) connected directly to the e-shop via API
- Translates product descriptions, categories, filters, email templates
- Preserves formatting, technical specs, SEO keywords in the target language
- Humans review and fix exceptions – not translate from scratch
What it saves: Hundreds of hours during expansion. Translation of thousands of products that would take an agency months is done by AI in days + review.
How the Transition to Automation Works
Start where manual work hurts the most. You don't have to automate everything at once.
- Process audit (1–2 days) – we review what you do manually, where errors occur
- Solution design + approval – architecture, priorities, timeline
- First integration implementation – typically orders ↔ accounting system
- Parallel operation – old and new solutions run simultaneously, we compare outputs
- Go live + monitoring – switchover, monitoring the first days
- Expansion – additional integrations (carriers, translations) by priority
Throughout the entire transition, the store runs normally. No downtime, no "big bang". Parallel operation eliminates risk – if something doesn't match, you simply revert to the manual process while the issue is fixed.
And the main bonus: your colleague, instead of copying data, handles customer service, complaints, upsell – things that generate revenue.
How Much Does It Actually Save? A Model Calculation
No made-up case studies. Just math you can recalculate for your own store.
Take a typical online store:
- 30 orders per day
- 2–3 minutes of manual work per order (copying to accounting system, checking, label)
- 22 working days per month
Calculation: 30 × 2.5 min × 22 days = 1,650 minutes per month = 27.5 hours per month
At an hourly labor cost of 300 CZK/h (full employer cost): 27.5 × 300 = 8,250 CZK per month
Connecting the e-shop with the accounting system costs roughly 30–50 thousand CZK.
Payback: 4–6 months. After half a year, automation just saves money.
And that doesn't count cancellations due to wrong stock levels, invoice errors, time spent answering "where's my package" – nor the fact that person can spend those 27 hours per month doing something that generates more business. Customer service, upsell, returns, product range development. Real payback is faster.
Every store is different – on a call we'll calculate it precisely for yours.
When Automation Doesn't Make Sense
I won't sell you something you don't need. Automation isn't worth it if:
- You have fewer than 5 orders per day – manual work takes 15 minutes, the investment won't pay back
- Your processes aren't set up yet and you change them often – stabilize first, then automate
- You plan to close or fundamentally rebuild the store within six months – the investment won't pay back in time
In these cases, it's better to wait. And if you're not sure where you fit – we'll discuss it on a call and I'll tell you straight.
Before: E-shop → Human (copy-paste) → Accounting system → Human (copy-paste) → Carrier
After: E-shop ↔ Middleware ↔ Accounting system + Carriers – all automatic
E-commerce automation doesn't mean firing people. It means stopping the waste of their time on work a machine can handle – and letting them do what generates revenue.
Stock management, invoices, shipping labels, inventory, emails, translations – all of this can run without human intervention today. You don't have to do everything at once. Start with one integration and expand gradually. For what exactly to automate, see part 1.
If you're not sure whether automation is worth it for you – let's go through it together in 30 minutes. We'll calculate it on your numbers.
