Solution · Prospecting and follow-ups

Sales outreach automation: fewer emails, better targeted

An AI agent spots companies that have just entered the commercial register, scores how well they fit your offer and drafts a short, personalised first email. You read every message before it goes out.

  • Every message approved by a person
  • Sender clearly identified
  • An easy way to say no

How automated sales outreach works

Every working day, the Swiss Official Gazette of Commerce (SOGC, or FOSC/SHAB) publishes the new entries in the commercial register: company name, registered office, purpose. For an accounting firm, an agency or a B2B supplier, these are potential clients at the very moment they are getting organised.

A prospecting agent reads these notices and other public sources, keeps the companies that match your criteria (sector, canton, activity) and notes why. For each one, it drafts a short first email built on a specific fact, then one or two follow-ups. You read and approve every message.

The same agent also looks after your existing contacts: it prepares a follow-up when a quote gets no answer and keeps your pipeline up to date, in a simple spreadsheet or in your CRM.

In Switzerland, mass advertising by email without prior consent is banned (Unfair Competition Act, UWG/LCD art. 3 para. 1 lit. o). Our approach: a small number of messages, chosen one by one and checked by a person, with a clear sender and an easy way to say no.

The problem

Why prospecting comes last

No time to prospect

Client work comes first, so prospecting keeps getting put off.

Quotes with no reply

A quote goes out, then nothing. Nobody follows up at the right moment, and the deal fades away.

Mass mailings cause problems

The same message sent to hundreds of addresses risks ending up as spam and, without consent, breaks the law.

Interactive demo

Try the prospecting agent

Choose an offer, a target sector and a canton. The agent spots newly registered companies (fictional), scores how well they fit and drafts a short, personalised first email plus two follow-ups, all waiting for approval. The demo also explains the Swiss rule on mass emails.

Sales outreach agent

Spots new companies and writes each one a tailored message. You approve every send.

Idle
Where does the AI run?
agent · prospecting Fictional data
New registrations
—
After your filters
—
Kept · score ≥ 70
—

      Choose your offer, a sector and a canton, then run the agent.

      The Swiss rule. In Switzerland, the Unfair Competition Act (art. 3 para. 1 let. o) bans mass advertising by email without prior consent. Our approach: few messages, truly individual, approved by a person, with the sender clearly identified and an easy way to say no.

      Everything runs in your browser: nothing is sent or stored. Simplified demo: answers are prepared in advance or computed by simple rules in your browser. In production, a language model suited to your data does this work, and someone on your team signs off.

      I want this for my business

      What’s happening, step by step

      1. New companies appear Input

        The agent reads the new registrations published in the SOGC, the Swiss Official Gazette of Commerce. Public data, updated every working day.

      2. Your criteria Code

        Simple code keeps only the sectors and cantons you serve. No AI here: clear filters, set with you.

      3. The AI rates the fit AI

        A language model reads each company’s registered purpose and gives it a fit score, with its reasons in plain words.

      4. A tailored message AI

        For each company kept: a short, personal first email and two spaced follow-ups, in the language and tone you chose.

      5. Swiss rules checked Code

        Code checks every message: identified sender, a free and easy way to say no, an individual message, a capped volume (Unfair Competition Act, art. 3 para. 1 let. o).

      6. You approve Human

        Nothing goes out without your say-so: you read each sequence, adjust it if needed, and decide.

      Run the demo: each step lights up here.

      ModelApertus 70B · Swiss open source

      Data processedA server in Switzerland or yours

      Picture a sales assistant who reads the new entries in the commercial register every morning, spots the companies that genuinely need what you offer, and drafts a personal first email plus two follow-ups for each. No mass mailing: a handful of well-targeted messages that you read and approve one by one.

      example 20 min → 1 min to qualify a company and prepare its first email, review included
      example FR · DE · IT · EN each message in the language of the company you are writing to
      example 0 emails sent without your approval

      Words to know

      SOGC
      The Swiss Official Gazette of Commerce. Every working day it publishes the commercial register entries: new companies, changes, deletions.
      Fit score
      A mark from 0 to 100 showing how closely a company matches your ideal client. The AI must always give its reasons.
      Sequence
      A first email, then two spaced follow-ups (day 4, day 10). A reply or a “no thanks” stops everything.

      Why it’s safe

      • Nothing goes out without your click: every sequence waits for your approval before the first email.
      • Code checks every message before you do: identified sender, one-word opt-out, individual message, capped volume (Unfair Competition Act).
      • A “no thanks” stops everything: the address joins the opt-out list, checked before every send.
      • Public data only (SOGC, Zefix, the company’s own website), and you choose where the AI runs.

      The agent’s pipeline

      1. Input Public sources New SOGC registrations, completed with Zefix and the company’s own website (the contact address it publishes).
      2. Code Filters Canton of the registered office, legal form, keywords in the registered purpose. Existing clients and the opt-out list excluded.
      3. AI Score and reasons The model reads the registered purpose and returns schema-validated JSON: score 0–100, reasons, angle.
      4. AI Drafting Day 0, day 4 and day 10 emails in the company’s language, from your offer, your examples and your tone.
      5. Code Swiss rules check Real sender, an opt-out line in every email, individual content, daily cap, no duplicates.
      6. Human Approval and sending Approval queue, then sent from your mailbox (Microsoft 365 or Google Workspace) at a limited pace.

      AI reads and writes. Code checks. Humans decide.

      Stack

      n8nSOGC · ZefixPythonJSON SchemaApertus · GPT · Claude · MistralMicrosoft 365 · Google WorkspacePostgreSQL (tracking, opt-out list)

      Guardrails

      • Few messages: a daily sending cap, three emails at most per company, no bought lists.
      • No means no: a reply or a “no thanks” stops the sequence. The address joins the opt-out list, checked before every send.
      • Justified score: if the JSON doesn’t match the schema (score and reasons), the company is set aside, not guessed.
      • Human approval: the agent drafts, a named person reads and sends. Everything is recorded in the audit log.

      What the agent produces (JSON)

      {
        "source": "SOGC · 2026-10-27",
        "company": "Café Zéphyrine Sàrl",
        "town": "Morges (VD)",
        "purpose": "Running a café-restaurant and catering service.",
        "contact": "info@cafe-zephyrine.example",
        "fit_score": 91,
        "reasons": [
          "Registered 3 days ago",
          "Sàrl: bookkeeping required by law",
          "VAT likely above CHF 100,000 turnover"
        ],
        "sequence": [
          {
            "day": 0,
            "language": "en",
            "status": "awaiting_approval"
          },
          {
            "day": 4,
            "only_if": "no_reply"
          },
          {
            "day": 10,
            "only_if": "no_reply",
            "last": true
          }
        ],
        "checks": {
          "sender": "marc@alvelia-fiduciaire.example",
          "opt_out_in_every_email": true,
          "individual_message": true,
          "daily_volume": "5/20"
        },
        "human_approval": "pending"
      }

      Choose the engine

      How we set it up

      1. Day 1 Free audit 30 minutes to define your ideal client, your cantons, your offer and what must stay human.
      2. Weeks 1–2 Test on your data We tune filters, scoring and messages on recent registrations in your area. You review the first drafts.
      3. Weeks 2–4 Go-live We aim for a first automation in service 2 to 4 weeks after the audit. The agent is connected to your email, with your sending cap and approval rules.
      4. Every month Care Replies, opt-outs and message quality monitored, adjustments and a monthly report.

      What we need from you

      • Your offer in a few lines, plus two or three emails you have already sent
      • Your ideal client: sectors, cantons, size
      • The mailbox of the person who will sign the messages
      • Your current client list, so they are never prospected

      What we set up

      An agent that prepares your outreach, one message at a time

      Your targets, your tone, your approval rules.

      01

      Spotting new companies

      Every working day, the agent reads the new SOGC entries and other public sources, and filters them by your sector and cantons.

      SOGCPublic sources
      02

      Fit score

      Each company gets a score against your criteria, with the reason in one sentence.

      03

      First email and follow-ups

      A short message explaining why you are writing to this company, then one or two spaced-out follow-ups. Every message waits for your approval.

      Human approval
      04

      Quote follow-ups

      A quote still unanswered after the delay you set, say 7 days? The agent prepares a polite follow-up, approved by you or sent automatically if you choose.

      05

      Pipeline kept up to date

      Contacts, status, replies and next steps are recorded in a shared spreadsheet or in your CRM, through its API or an export.

      06

      Do-not-contact list

      Any company that replies “no thanks” is taken off the list and not contacted again.

      How we work

      How it goes

      1. Free audit

        In 30 minutes we define your offer, targets and cantons, and look at how you follow up on quotes.

      2. Test on real data

        We test the agent on recent real notices. You see the companies it picks and the drafts, and adjust the tone.

      3. Go-live

        We aim for a first automation in service 2 to 4 weeks after the audit.

      4. Monthly care

        We adjust the criteria based on the replies you get and send you a report every month.

      The legal framework

      The Swiss rule on mass emails

      Art. 3 para. 1 lit. o of the Unfair Competition Act (UWG/LCD) bans mass advertising by telecoms (email, text message…) sent without prior consent, without a correct sender or without saying how to refuse it easily and free of charge. The exception for existing customers does not cover new prospects. So we stay clear of mass advertising and apply these safeguards to every message anyway.

      Individual messages, not mass mailings

      A few companies, chosen one by one, and a message written for each, checked by a person. No bulk sending.

      A clearly identified sender

      Every email goes out from your address, with your name, your company and your full contact details.

      An easy no, respected

      Every message says how to refuse further contact, free of charge. A simple “no thanks” is enough: the company is not contacted again.

      Frequently asked questions

      Is email prospecting legal in Switzerland?

      It is possible, as long as it isn’t mass advertising without consent. The law sets no numerical threshold: the same template sent to hundreds of addresses, even with the first name changed, may well count as mass advertising. This page is not legal advice.

      Where do the suggested companies come from?

      From public sources, mainly the new entries published in the Swiss Official Gazette of Commerce (SOGC). The agent adds public information about each company, such as its website.

      Are the emails sent automatically?

      No: every prospecting email is read and approved by a person. Only quote follow-ups can go out automatically, if you choose, case by case.

      What does a first email look like?

      A few sentences: why you are writing to this company, what you offer and one simple question. It starts from a specific fact, such as a recent registration, and follows your tone.

      Does it work with our CRM?

      We connect the tools you already use, including your CRM, through their API, an export or email. Without a CRM, a shared spreadsheet is enough to start.

      How much does it cost?

      It is priced on quote, after a free audit. The price depends on the sources to monitor, the number of companies to score each month and the tools to connect. Request a quote.

      NotHuman

      How many of your quotes go unanswered each month?

      In a free 30-minute audit, we look at your targets and follow-ups and propose a concrete plan.

      Contact

      A question? A project?

      A real person reads every message. No commitment.

      We only use this information to reply to you. See our privacy policy.

      A personal reply, usually within 2 working days.