Blog · Published · Updated
ERP Order Entry: When n8n Is Enough
Short answer
You turn an order written in the customer's words into item codes and quantities with rules: matching to the catalogue, converting units and deciding on substitutes. Unclear lines go to an operator. An agent is needed only when the agreed path is not enough. At Syntalith an automation costs from €3,000 net, and with 1,000 orders a month at 6 minutes each it pays back in 1.1 months.
How we workDetails ↓
- 01Call30 minutes with an engineer, free
- 02PrototypeA working system on your process and data, free
- 03ImplementationTarget and fixed price in the contract; 50% at signing
- 04Acceptance50% once the agreed target is met
- net for the automation of one process, build paid once
- from €3,000
- from signing the contract to acceptance
- 2–6 weeks
- free prototype on your cases
- €0
We belong to the OpenAI and Anthropic partner programmes and learn through their official courses. What this means ↓
Author: Syntalith · 3 min read
Statistics
Orders and automation: statistics 2026
1.1 mo.
payback of a €3,000 automation at 1,000 orders a month: €3,490 of work today, €706 with checking and the model. Scenario.
21 of 21
According to a Syntalith test (2026), a local order intake demonstrator passed 21 of 21 cases: 12 catalogue matches, 5 unit conversions and 4 substitute decisions.
9 of 9
According to a Syntalith test (2026), 9 of 9 new cases checked on 9 October 2026 gave the expected result, and a trial with a negative quantity showed missing input validation.
20.0%
According to Eurostat (2025), 20.0% of EU companies with at least 10 employees used AI.
Source: Eurostat, 20% of EU enterprises use AI technologies (2025) ↗
55.0%
According to Eurostat (2025), 55.0% of large EU companies with 250 or more employees used AI.
Source: Eurostat, Use of artificial intelligence in enterprises (2025) ↗
Figures checked: 22 October 2026
What to do with three boxes of gloves
A salesperson receives an order: „three boxes of gloves”. The code first looks for a confirmed product in the catalogue. A full name gives a match, and a shortened or ambiguous description stays a candidate to check. Three boxes of a hundred give three hundred units in the draft, and the operator sees the conversion. An unknown unit and an unavailable product that needs a substitute go to the operator, because consent to substitutes is not consent to any item code.
In the demonstrator these decisions are three separate functions: catalogue matching, reading the unit and the substitute decision. Each can be right or wrong independently. The operator picks the item code, confirms the unit and records the reason. Repeating the same decision is rejected, and an order with an unresolved line does not go to export.
- 01Cataloguea full name gives a match, a short form stays a candidate
- 02Unitthree boxes of a hundred are three hundred units
- 03Substituteonly from the allowed list, otherwise the operator decides
- 04Operatorpicks the item code, the unit and the reason
- 05Exportonly orders with no open lines
- 25 September 2026: 12 matches, 5 conversions and 4 substitute decisions, 21 of 21 passed
- 9 October 2026: 9 new cases, all with the expected result
- A negative quantity passes through the conversion function, so input validation has to be added
- No model calls, no n8n and no external order system
What order intake automation costs and when it pays back
- Today
- 1,000 orders × 6 min = 100 h × €34.90 = €3,490
- With the system, a person checks the output in 20% of the time
- €698
- Model, 1 call per case, Claude Sonnet 5.5
- €8
- Cost of one case
- €3.49 today, €0.71 with the system
- Saving
- €2,784 a month
Payback of €3,000
1.1 months
The volume, the time and the checking share are the scenario's assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat. Model: 2,500 input tokens and 400 output tokens per call at Anthropic's rates, converted at the ECB rate of 1.1205 USD per euro on 22 October 2026.
The count covers the work around a case: searching, reading, writing and passing it on. What stays is the time to check the output, 20% of today's work in the scenario. Before we quote, we measure both times on 20 to 100 cases your team has already handled, in a €0 test that takes 5 working days.
The confidence field in the demonstrator holds constants written into the code. In a rollout, we measure reading confidence separately on the client's orders.
Do you need n8n, AI or a feature of your current system
If orders arrive in a file with fixed columns and correct item codes, check the import in your current order system: a model has nothing to recognise there. When data has to move between several tools and the conditions are known, an n8n flow is enough: the If node splits cases by conditions, and HTTP Request calls an API. AI is worth testing when a customer writes a free-text message from which the lines have to be extracted. Converting boxes into units is an ordinary rule.
A file with fixed columns and item codes
the import in your current system
configuration
Known conditions, several tools
a flow with rules and operator decisions
from €3,000
Free-text messages from customers
an AI step that reads the lines before the rules
from €3,000
Where to start
Start with one order channel and writes to a test environment. To prepare: the current catalogue, units and pack sizes, and sample orders with the correct result. The demonstrator's screens are on the page order intake. We build this kind of rollout as part of AI automations. For the free process review, a description of one case from recent weeks and the tools it went through is enough. You do not need to send confidential documents, and you get a written plan within 2 working days.
Worked example · Scenario
How we count the payback: order intake
We run the same count for your orders in the free process review.
€3,723
a month for handling 800 orders by hand
Payback: 1 mo.
8 minutes to retype an order into the software and checking in 15% of the old time are our assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat, for companies with 10 or more employees. Model cost: reading the order in our system (about €0.0006) plus matching the lines (Claude Sonnet 5.5, about 3,000 + 500 tokens), at the ECB rate of 1.1205 USD per euro on 22 October 2026.
See the full calculation →
