01Layer 01

The bottleneck sits behind the enquiry.

Enquiries rarely fade away because they were poor. They fade because nobody follows up systematically. Time to the first real reply moves the close rate more than price or design does — and it is the most expensive lever if you try to solve it by hiring.

What gets built

Provia — the CRM module inside Reventis

An intermediary firm earns on two fronts: new business and the book of business. Provia covers both — the path from enquiry through conversation and quote to the close, and the care that follows, with reminders, touchpoints and renewals. Built in rather than bolted on is a complete communication suite: WhatsApp and the other channels converge in one team-capable inbox — several staff work in the same thread, every message sits on the customer record, and customer contacts do not live on private phones.

First contact in the messenger, within minutes

The automatic first contact runs in exactly that inbox: an assistant collects the missing details — and nothing else. Standard cases such as rejections or appointment requests are decided deterministically, without ever asking the model. When a human takes over, they see the full history without switching tools.

An output filter before every reply

Two-stage: price commitments, assurances and model slip-ups are blocked before they reach the customer. The assistant explains, the human commits.

Follow-up that forgets nothing

Inside Provia: a status model, reminders, multi-step follow-up chains with caps and quiet periods, a daily list of what needs doing — and an alert when someone has been waiting too long for a reply.

Quotes and documents

On-brand quotes at the press of a button, multilingual and in several formats, digital signature with an audit trail. With feedback into Provia, so that an issued quote becomes measurable instead of disappearing into an inbox.

Instant pricing without a provider interface

Where your product provider offers no pricing interface — the normal case in many industries — a pricing engine is built from controlled measurements that determines the price for a bounded segment to the cent. The real deliverable is the doctrine behind it: where the data does not support a reliable price, the engine returns no price at all and hands the case back to a member of staff. A price quoted on the spot is the strongest lever in a first conversation — a wrong one burns the customer for good, because they have already passed the number on.

What it does not do

The assistant does not advise, does not quote prices and commits to nothing. That is not a limitation but the precondition for letting it near customers at all.
The system does not sell. It makes sure nobody is forgotten.

What you bring

Your own messenger business number, a team willing to work with a status model, and the willingness to settle the handover rules once, properly: when does a human take over?

Ongoing modules for this layer

Operations: Hosting, monitoring, updates, backups, small adjustments.
Provia: The CRM module: your own instance with pipeline, follow-up and the complete communication suite including team-capable WhatsApp.
Quotes: Individual quote PDFs at the press of a button, on brand, in seconds instead of half an hour.
Signature: Your customer signs the intermediation agreement digitally via a link, recorded and traceable.
Assistant: Messenger assistant, follow-up automation, model and messaging costs.
Data upkeep: Re-measuring pricing engines, keeping tariff data current. An add-on outside the plans — quantified in the concept.

Build · one-off · €7,000 – 15,000

What it costs →

How the work is done

Four construction decisions that were actually made this way in the system — open to questioning, not merely claimed.

No number beats a wrong number
Where the data doesn't support an answer, the system hands the case to a human instead of guessing.
Nothing is discarded silently
Anything suspicious is flagged and reviewed, not deleted. A genuine enquiry dropped in silence is the most expensive mistake such a system can make.
Automation commits to nothing
No assistant promises a price, no report changes an ad account. The model explains, the human commits.
Handover-ready from day one
Documentation is enforced, not written on the side. An outside developer understands the system in two days.

Does this fit?

It starts with a conversation about the goal, the scope and what is realistic. If the system doesn't fit, that is said at the end of that conversation — not in an invoice.

Request a first conversation All layers