If someone does it by copying and pasting, it can be automated
Invoices typed by hand, reports assembled every Monday, emails someone classifies and forwards, data moved from one system to another. All of it runs on its own, with alerts when something fails.
Key takeaways
- The test for whether a task is worth automating: does someone do it more than once a week, always the same way, moving information between two places?
- What AI changed is that a model reads invoices with varying layouts and scanned documents, so data no longer has to arrive in a fixed format.
- Quarl uses self-hosted n8n when the case allows it, and custom code when the logic exceeds what a visual tool expresses well.
- Automating a chaotic process does not organize it: it makes the chaos faster. Around 60% of companies wanting to automate have an ordering problem first.
Operating track record
Three years operating the system in production — not one delivery and an exit.
The most frequent failure in RAG architectures, solved with structured metadata at ingestion.
The proposal arrives within 48 hours with fixed scope, price and date. If scope changes, it is quoted separately and approved first.
The three-minute test
Ask yourself this: does someone do it more than once a week, always the same way, moving information between two places? If yes, it can almost always be automated and almost always pays for itself in under a year.
What AI changed is the kind of task that fits that definition. Automation used to require data arriving in a fixed format. Today a model reads invoices with different layouts, emails written in natural language and scanned documents, and extracts what matters with schema validation. That massively widens what is worth automating.
What we get asked for most
- Invoices and documents typed by hand into an accounting system or a spreadsheet. Today they are read automatically, even scanned or photographed on a phone.
- Recurring reports that someone assembles from three different sources every week.
- Email classification and routing: quotes to sales, support to support, invoices to accounting, with key data extracted.
- Syncing systems that do not talk: online store with inventory, CRM with billing, web forms with the database.
- Alerts and follow-ups that today depend on someone remembering: expirations, outstanding payments, low stock, contracts up for renewal.
- Bulk data loading from files that arrive from suppliers or customers in always-different formats.
How we build it
- Map the process as it is todayWith the person who runs it, not with the org chart. We routinely find three undocumented exceptions, and those are exactly what break badly built automations.
- Build with the exceptions insideAn automation that works for 80% of cases and fails silently on the other 20% is worse than nothing, because nobody checks.
- Monitoring and alertsWhen something fails — and eventually it does — someone finds out the same day, with detail on what was processed and what is pending, and the process can be retried without duplicating.
- Documentation and handoverWritten down: what it does, how to modify it, who to call. The goal is that your team can adjust it without depending on us.
If the process changes every month, if nobody can explain how it really works, or if it depends on judgment calls in every case.
Automating a chaotic process does not organize it: it makes the chaos faster and harder to audit. Around 60% of companies wanting to automate have an ordering problem first. If that is your case, we prefer to say so on the first call.
Ranges and timelines
| Scope | Example | |
|---|---|---|
| One process | Invoice reading into the accounting system. Automated weekly report. | 2–3 weeks |
| Several connected processes | A web order that triggers inventory, invoice, customer notification and team alert. | 4–6 weeks |
| With intelligent document processing | Variable-format documents interpreted with AI rather than rigid templates. | +1–2 weeks |
An automation that saves your team ten hours a week usually pays for itself within months and then keeps working for free. We run that calculation with your real numbers on the call.
What we automate with
Related services
AI agents
LangGraph orchestration, durable state and human approval on steps with consequences.
View serviceIntegrations and APIs
MCP servers against the systems already in operation.
View serviceLanguage models
LLM applications with structured output and tool calling.
View serviceAI consulting
Architecture, evaluation criteria and cost per query before writing code.
View serviceFrequently asked questions
What is the difference between n8n, Zapier and custom development?
Zapier is the easiest and the most expensive at scale: it charges per task executed and the bill climbs fast at volume. n8n does the same but can be self-hosted at fixed cost and handles considerably more complex logic. Custom code is for processes with rules no visual tool expresses well. For most mid-market companies self-hosted n8n is the sweet spot, which is why it is our default.
Do we have to change the systems we already use?
Almost never. The automation connects to what you have through APIs, databases or even exchanged files. If one of your systems is completely closed with no way to connect, we tell you on the first call before you invest in a proposal.
What happens if the automation fails one day?
It will fail eventually: an API changes, a file arrives in a strange format, a service goes down. That is why monitoring is not optional. When something fails, someone on your team gets an alert the same day with detail on what was processed and what is pending, and the process can be retried without duplicating anything. An automation that fails silently is the worst possible outcome.
Does this replace people?
In practice, almost never. It removes mechanical work so the same people handle what requires judgment. Most of our clients automate because their team is underwater, not because they want to shrink it. If your goal is headcount reduction, say so upfront: it changes how the project is designed and what has to be documented.
How reliable is AI document reading?
Very good on documents with reasonable structure, and much better combined with schema validation and business rules. What matters is designing what happens with low-confidence cases: those go to human review rather than into the system. A flow that accepts everything blindly is where automation becomes a problem instead of a solution.
What does it cost to maintain?
With n8n on your own server, roughly USD 20 to 60 monthly in infrastructure for typical volumes. If there is AI document reading, model consumption is added on top depending on volume. We do not charge a subscription on the automation: it is yours and documented so your team can maintain it.
What is the difference from an AI agent?
An automation follows a defined path: if this happens, do that. An agent decides the path based on the goal and context. Automation is cheaper, more predictable and easier to audit, so when the process has clear rules it is the right answer. The agent comes in when the path cannot be enumerated in advance.
Where should we start?
With the process that consumes the most hours and requires the least judgment. Not the most visible one, and not the one the executive asks for: the one that makes someone on your team lose half a day a week to something mechanical. That one usually pays back fastest and builds the confidence to keep going.
Book 15 minutes
Tell us what you are building, or what stopped working. You leave the call with a concrete answer: it can be fixed, it can be built, or it isn't worth it.