Questions

The questions we get asked before the first call ends.

Including the ones with an inconvenient answer.

Do we have to replace our existing systems?

No. The model sits over what you already run and connects the hand-offs between them. Replacing a working system is expensive and rarely the reason things are slow — the gaps between systems usually are.

Can the AI send something to a customer on its own?

No, and this is a design rule rather than a setting. Anything customer-facing, money-spending or publishing is blocked from automatic execution and waits for a person. Only read-only steps may run automatically, and every automation runs as a dry-run before it can run at all.

Do you certify us to ISO 9001?

No. We are not a certification body and cannot issue a certificate. What we do is produce the documented information a certification audit asks for — process maps with owners and indicators, a risk register, document control, and a PDCA loop — generated from the live model rather than written by hand. The audit itself is between you and your certification body.

How is this different from an ERP or CRM?

An ERP covers the same ground as many separate modules, with AI added later as an assistant. Here the operating model is AI-native from the start, marketplace connectors and payroll are included rather than sold as apps, and ISO records are generated rather than maintained separately. The practical difference is that you switch cards on as you need them instead of implementing a suite.

What happens to our data?

Credentials and tokens are held in encrypted fields, never in plain text or in the codebase. Records are archived where you choose — a shared server folder, OneDrive or Google Drive — and each form carries a confidentiality class under document control. The discovery form explicitly asks what must not leave the company, so we design around it rather than discover it later.

What size of company does this fit?

The same engine fits a 20-person firm and a 500-person one, because cards are added or removed per company rather than shipped as a fixed suite. What matters more than headcount is whether you have at least one repetitive process with a real owner — without that, there is nothing to model.

What if AI turns out not to be the answer for our process?

Then we say so. Scoping exists to answer exactly that with a go / no-go, and it comes before the invoice rather than after it. Quite often the honest finding is that the process needs fixing, not automating — automating a broken process just produces errors faster.

Which languages do you work in?

English, Turkish, Arabic, German, French and Spanish — including the outreach service, where campaigns are written per market rather than translated once and reused everywhere.

Do we own what you build?

Your process model, your forms, your records and your data are yours, and they are archived in a location you control. That is also why stopping is a real option rather than a threat — there is no version of this where walking away costs you the documentation.

How much does it cost?

It depends on how many processes are in scope, how well defined they already are, what has to be connected and how many people the workflow touches. We share the figures in a short call once we know which process we would be working on. There is no lock-in and no minimum term.


Related

See it running on your own process.

If your question is not here, it is probably the interesting one. Ask it.

Request a demoRead the FAQ