● B2B offer — Outbound prospecting
The AI agent for outbound prospecting: fill the diary with qualified appointments
Are your salespeople spending more time looking for prospects than selling? Your AI agent identifies your targets, personalises every approach and follows up at the right moment — to fill their diary with qualified appointments, without growing the team. Hosted in France — on local inference or an isolated resource — your prospect lists stay under control. The salesperson keeps the decision.
Hosted in France
Prospect lists protected
GDPR & AI Act: governed deployment
Human oversight
Updated on
Deployed in 1 to 2 weeks
Prospecting agent · hosted in France
Build me a list of 50 IT directors in software publishing (50 to 250 staff, France) and prepare a personalised first approach on LinkedIn.
50 targets identified and enriched (role, size, sector, area). 47 professional targets identified; profiles to be checked manually by the salesperson, 3 duplicates set aside. A first approach written for each one, personalised from their recent news (funding rounds, hiring, product launches).
Automated email sequence + LinkedIn tasks prepared for manual execution — the information notice and opt-out are built in, for approval.
⛓ Sourced · public sources + your CRM
Automated email sequence + LinkedIn tasks prepared for manual execution — the information notice and opt-out are built in, for approval.
⛓ Sourced · public sources + your CRM
Launch the sequence and follow up those who do not reply, but stop at once as soon as anyone asks not to be contacted again.
Sequence live. Follow-ups scheduled at day 3 and day 7 according to behaviour (opens, clicks, replies). Automatic opt-out: any request to unsubscribe stops that person immediately and removes them from the list, in line with the GDPR.
I pass on the interested replies for qualification and propose appointment slots to you, for your approval.
✎ Action · sequence ready to launch — the salesperson approves
I pass on the interested replies for qualification and propose appointment slots to you, for your approval.
✎ Action · sequence ready to launch — the salesperson approves
Local inference · no data outside the EU
Data hosted in France
Sovereign by designLocal inference or hosting in France
GDPR & AI Act: governed deploymentLawful basis, opt-out & human oversight
TurnkeyDesigned, installed and operated for you
The salesperson decidesThe agent assists, the human decides
✦ In brief
A Blue Lemon Agent outbound prospecting agent automates the search, the qualification and the email approach to your B2B prospects, and prepares the LinkedIn tasks your salesperson carries out personally. It personalises every message, follows up at the right moment and records the appointments in your CRM. It runs on local inference or is hosted in France: your prospect lists never pass through a foreign server, architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity. The result: more qualified appointments for the same sales effort. Live in 1 to 2 weeks, with systematic human oversight. Your teams write to it from Microsoft Teams, Slack or their email, and your customers reach it on WhatsApp Business, your website chat or email — with no account to create and nothing to install. These connections are included in every plan, at no extra cost, within the number of connections your level includes.
100%
hosted in France in the target architecture
0
transfer outside the EU in the target architecture
9
uses ready to deploy on this scope
0
decision taken without human approval
Reference points describing our offer, not results measured at a client. The scale of the gain is confirmed by a pilot on your own scope.
The context
Why AI prospecting matters to B2B teams — and why they hesitate
Growth depends on a steady flow of qualified contacts. But your salespeople lose precious time on searching and following up — and American prospecting tools route your prospect lists through foreign servers.
! The issue
Your salespeople are caught between an appointment target that never comes down and repetitive tasks — finding targets, writing, following up — that eat into their selling time. Yet most sales automation tools amount to entrusting your prospect lists, your messages and your contact data to a third party, often hosted outside Europe and subject to the Cloud Act.
✓ Our answer
AI prospecting is only of interest if it is sovereign, compliant and responsible by design. Local inference or an isolated resource hosted in France, legitimate interest as the lawful basis, one-click opt-out, systematic human oversight: time gained at the top of the funnel is never paid for in lost confidentiality. The aim is not to replace the salesperson, but to give them back time to sell.
The decisive point
Your prospect lists: sovereignty & GDPR compliance
B2B prospecting processes personal data. Here is how the architecture of our agents protects it — and how the agent prospects responsibly.
Local inference
The agent can run on a machine belonging to the company: no prospect list leaves the network, nothing passes through a cloud.
Hosting in France
Otherwise, a dedicated and isolated resource, hosted in France under French law — your prospecting data: processing and access within the European Union targeted by the architecture.
Reduced extraterritorial exposure
Architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
GDPR-responsible prospecting
To a business address: legitimate interest as the lawful basis, an information notice and one-click unsubscribe (opt-out), in line with the French data protection authority's guidance. To a natural person, the agent applies prior agreement.
Encryption & controlled access
Encryption in transit and at rest, role-based access (RBAC), strong authentication and logging.
AI Act: governed deployment
The agent is strictly in support; no appointment or sensitive send is approved automatically; traceability and human oversight from end to end.
What depends on the architecture chosen These points are not general guarantees: they are settled deployment by deployment, in the quotation.
- The applicable location is that of the architecture set out in the quotation and verified before commissioning.
- Local execution is announced only for the configuration explicitly described and accepted in the quotation.
- The applicable isolation depends on the deployment mode set out in the quotation; no dedicated isolation is presumed.
- The encryption mechanisms in transit and at rest, their components and key management are those documented for the architecture chosen.
- Roles and permissions are configured and accepted for the identities and systems actually connected.
- The events logged, their content, their retention period and who may access them are defined for the deployment chosen.
The agent does not prospect private individuals without an appropriate lawful basis, and it never works around the GDPR: consent, opt-out and information are built in by design. Designed for deployment in line with the GDPR and the AI Act, after the processing, roles and context-specific risks have been assessed.
Demonstration
See the agent at work
4 real situations, taken from those that come up most often. Pick one: the exchange unfolds as it would in your organisation.
A scripted demonstration. These exchanges show how the agent behaves — its sources, its refusals, what it leaves to your teams. Nothing is sent from this page, no model is queried here, and the matters named are fictional. That is precisely what we promise your data.
The behaviours shown here — monitoring, automation rules, routing and reminders — are configured with you during deployment, from your tools, your rules and your thresholds.
The architecture points named in these exchanges — location, local execution, isolation, encryption, role-based access, logging — are not a guarantee attached to the demonstration: they are those of the architecture set out in your quotation, and verified before commissioning.
The company in this demonstration
Fictional companyCedalis — vendor of a management platform for industrial SMEs
- Sector
- B2B software, sold by subscription
- Headcount
- 34 staff, including 2 salespeople and a CEO who still sells
- Target market
- French industrial SMEs of 20 to 250 staff
- Sales cycle
- 3 months on average, €9,000 a year per account
- Tools in place
- A CRM fed for 3 years, shared mailboxes and calendars, LinkedIn
- Who decides
- The sales director approves campaigns; the two salespeople hold the meetings
- Room for improvement
- 8 qualified meetings a month, and 10 h of prospecting per salesperson per week
Cedalis has grown by word of mouth and wants steady outbound prospecting without hiring. Nobody on the team has time to build lists or follow up. The agent is connected to the CRM, the mailboxes and the calendars; it works continuously, but every new rule goes through the sales director.
This company, its figures and the exchanges that follow were invented for the demonstration. They illustrate a common situation; they describe no real client.
Prospecting agent · hosted in France
I reread your three years of CRM overnight — 42 clients signed, 380 opportunities lost. Three characteristics clearly separate the two groups, and you barely prospect any of them.
What your signed clients have in common:
· Software vendors and digital-services firms, 20 to 80 staff. Above 250, zero signings in three years — 34 attempts, all lost at procurement stage.
· A dated fact less than 90 days old at first contact: a funding round, a new office, or an announced sales hire. Present in 31 of the 42 signed, in 68 of the 380 lost.
· The contact was the sales lead, not the CEO. CEO approached first: 96-day cycle. Sales lead: 41 days.
What that gives: when all three line up, you sign 1 in 4. When none do, 1 in 22. Yet 71 % of last quarter's sends went to companies ticking none of them.
My proposal: I make those three criteria your ideal-customer profile, and I only bring you companies that tick at least two. Approve it and I'll build the list. ideal-customer-profile_42-signings.pdf1 in 4 against 1 in 22
⛓ Sourced · CRM (3 years), 42 won deals, 380 lost, cycle lengths
What your signed clients have in common:
· Software vendors and digital-services firms, 20 to 80 staff. Above 250, zero signings in three years — 34 attempts, all lost at procurement stage.
· A dated fact less than 90 days old at first contact: a funding round, a new office, or an announced sales hire. Present in 31 of the 42 signed, in 68 of the 380 lost.
· The contact was the sales lead, not the CEO. CEO approached first: 96-day cycle. Sales lead: 41 days.
What that gives: when all three line up, you sign 1 in 4. When none do, 1 in 22. Yet 71 % of last quarter's sends went to companies ticking none of them.
My proposal: I make those three criteria your ideal-customer profile, and I only bring you companies that tick at least two. Approve it and I'll build the list. ideal-customer-profile_42-signings.pdf1 in 4 against 1 in 22
⛓ Sourced · CRM (3 years), 42 won deals, 380 lost, cycle lengths
The "funding or hiring" signal — where do you find it? And it ages fast.
On public, dated sources, and I sweep them every week — it is precisely because it ages fast that this has to run continuously.
Where I look: company press rooms and releases, their own “We're hiring” pages, public funding announcements, company registers for headcount and locations. Every fact is stored with its page and the date I consulted it — that is what will let me tell the prospect where I got it, in the very first message.
What I propose to keep running: every Monday morning, I bring you the companies that entered the profile since the previous week. A fact older than 90 days drops out on its own: a funding round from eight months ago opens no doors.
What I rule out before even showing you the list: your current clients, accounts your salespeople are already negotiating, and anyone on the unsubscribe register. A prospect contacted twice through two channels does more damage than a prospect not contacted at all.
What I need from you: do you approve these three criteria, or would you like me to test a fourth — for instance the use of a competing CRM, which is publicly detectable?
Where I look: company press rooms and releases, their own “We're hiring” pages, public funding announcements, company registers for headcount and locations. Every fact is stored with its page and the date I consulted it — that is what will let me tell the prospect where I got it, in the very first message.
What I propose to keep running: every Monday morning, I bring you the companies that entered the profile since the previous week. A fact older than 90 days drops out on its own: a funding round from eight months ago opens no doors.
What I rule out before even showing you the list: your current clients, accounts your salespeople are already negotiating, and anyone on the unsubscribe register. A prospect contacted twice through two channels does more damage than a prospect not contacted at all.
What I need from you: do you approve these three criteria, or would you like me to test a fourth — for instance the use of a competing CRM, which is publicly detectable?
Approved on all three. Build the list.
500 companies, each scored 0 to 100, and I'll tell you where to start.
How the score is built: 40 points for the dated fact and how fresh it is, 30 for size and sector match, 30 for an identified sales lead. The score isn't an opinion: every point maps to a verifiable line.
What the list gives: 120 companies above 70, 200 between 40 and 70, 180 below.
What each line carries: the company, headcount, the dated fact and its source, the role targeted, the address and its type — generic or named, the page where I found it and the date. No line without an origin: that is what makes the prospecting defensible, and it is also what makes the message credible.
My recommendation: open on the 120 above 70, and keep the middle 200 as a comparison group. If both bands give the same result, your score is worthless and I rebuild it. The bottom 180 I suggest leaving alone for now.
Next step: I write the sequence. I only send it after you've read it. targeted-list_500-scored-prospects.pdf120 above 70, every line sourced
⛓ Sourced · company websites, press rooms, careers pages, public registers
How the score is built: 40 points for the dated fact and how fresh it is, 30 for size and sector match, 30 for an identified sales lead. The score isn't an opinion: every point maps to a verifiable line.
What the list gives: 120 companies above 70, 200 between 40 and 70, 180 below.
What each line carries: the company, headcount, the dated fact and its source, the role targeted, the address and its type — generic or named, the page where I found it and the date. No line without an origin: that is what makes the prospecting defensible, and it is also what makes the message credible.
My recommendation: open on the 120 above 70, and keep the middle 200 as a comparison group. If both bands give the same result, your score is worthless and I rebuild it. The bottom 180 I suggest leaving alone for now.
Next step: I write the sequence. I only send it after you've read it. targeted-list_500-scored-prospects.pdf120 above 70, every line sourced
⛓ Sourced · company websites, press rooms, careers pages, public registers
How do you write the messages? I don't want a disguised mail merge.
Every message starts from a fact that belongs to that company alone. If I don't have that fact, I don't write the message.
The opening: the dated, verifiable fact — “you announced three sales hires on 12 June”. Not “I saw you were growing”.
The body: the problem that profile runs into, written in your own clients' words. I've reread the notes from your 42 won deals: the sentence that comes back 19 times is “we hire salespeople who spend their first month hunting for contacts”. That's the one I use.
The proof: one of your clients of comparable size and sector, named only if you authorise it, with a figure you gave me.
The ask: twenty minutes, two specific slots. An open question makes people think; two slots make them reply.
And two lines that never drop: where I got the address — the page and the date — and the unsubscribe link, honoured within the second. They aren't legal decoration: a message that says where it comes from gets more replies than one that dodges the question. personalised-message_before-after.pdfSame prospect: template against sourced message
⛓ Sourced · won-deal notes, product sheets, dated facts from the list
The opening: the dated, verifiable fact — “you announced three sales hires on 12 June”. Not “I saw you were growing”.
The body: the problem that profile runs into, written in your own clients' words. I've reread the notes from your 42 won deals: the sentence that comes back 19 times is “we hire salespeople who spend their first month hunting for contacts”. That's the one I use.
The proof: one of your clients of comparable size and sector, named only if you authorise it, with a figure you gave me.
The ask: twenty minutes, two specific slots. An open question makes people think; two slots make them reply.
And two lines that never drop: where I got the address — the page and the date — and the unsubscribe link, honoured within the second. They aren't legal decoration: a message that says where it comes from gets more replies than one that dodges the question. personalised-message_before-after.pdfSame prospect: template against sourced message
⛓ Sourced · won-deal notes, product sheets, dated facts from the list
And do you know whether it works, or are you assuming?
I test. On the first 120, I sent two different subject lines, 60 against 60, everything else identical.
After five days:
· Subject A — “Your three sales hires”: 41 % opens, 9 replies, 5 positive.
· Subject B — “How an agency doubled its meetings”: 25 % opens, 4 replies, 1 positive.
What I conclude, and the caution that goes with it: the gap on opens is clear, but 9 replies against 4 is still a thin margin. I propose rerunning the same test on the next 120 before turning it into a rule. Two consistent tests beat one spectacular one.
What I read behind the figure: the winning subject talks about the prospect; the losing one talks about you. That's a hypothesis, not a law — and the next test checks it.
The test I propose next: the closing ask. Two fixed slots against an open question. It's the second biggest lever after the subject line, and nobody ever tests it. ab-test_subject-a-against-subject-b.pdf41 % against 25 %, and what remains to confirm
⛓ Sourced · send log, opens, clicks and replies for both variants
After five days:
· Subject A — “Your three sales hires”: 41 % opens, 9 replies, 5 positive.
· Subject B — “How an agency doubled its meetings”: 25 % opens, 4 replies, 1 positive.
What I conclude, and the caution that goes with it: the gap on opens is clear, but 9 replies against 4 is still a thin margin. I propose rerunning the same test on the next 120 before turning it into a rule. Two consistent tests beat one spectacular one.
What I read behind the figure: the winning subject talks about the prospect; the losing one talks about you. That's a hypothesis, not a law — and the next test checks it.
The test I propose next: the closing ask. Two fixed slots against an open question. It's the second biggest lever after the subject line, and nobody ever tests it. ab-test_subject-a-against-subject-b.pdf41 % against 25 %, and what remains to confirm
⛓ Sourced · send log, opens, clicks and replies for both variants
And the follow-ups? That's where we usually give up.
The follow-up depends on what the person did, not on the calendar. That's where the gap opens: of your 42 signings, 27 replied from the second message onwards.
The rules I propose:
· Opened twice, no click → the topic lands, the ask blocked. LinkedIn task on day 3, a short message suggested, different angle — to be sent by the salesperson.
· Clicked, no reply → day 2, the comparable client story, without asking again for a meeting.
· Never opened → day 5, different subject line, different send time. Two times out of three it was the hour, not the message.
· Replied, whatever they said → I stop the whole sequence within the second. It's one of only two things I do without asking you.
The LinkedIn channel, in assisted mode: I do not connect to LinkedIn and I do nothing there. I prepare a dated task in your CRM with the suggested connection note and message; your salesperson checks the recipient, adapts the text and acts personally in the interface. I respect the platform's own quotas — breaching them gets your salesperson's account closed, not mine.
And a cap I recommend: three messages per person, across all channels. The fourth is no longer addressed to someone who missed the first three.
Approve the rules and I apply them tonight — you can change one without touching the others. follow-up-rules_by-behaviour.pdfFour rules, three messages per person maximum
⛓ Sourced · open and click behaviour, history of the 42 won deals
The rules I propose:
· Opened twice, no click → the topic lands, the ask blocked. LinkedIn task on day 3, a short message suggested, different angle — to be sent by the salesperson.
· Clicked, no reply → day 2, the comparable client story, without asking again for a meeting.
· Never opened → day 5, different subject line, different send time. Two times out of three it was the hour, not the message.
· Replied, whatever they said → I stop the whole sequence within the second. It's one of only two things I do without asking you.
The LinkedIn channel, in assisted mode: I do not connect to LinkedIn and I do nothing there. I prepare a dated task in your CRM with the suggested connection note and message; your salesperson checks the recipient, adapts the text and acts personally in the interface. I respect the platform's own quotas — breaching them gets your salesperson's account closed, not mine.
And a cap I recommend: three messages per person, across all channels. The fourth is no longer addressed to someone who missed the first three.
Approve the rules and I apply them tonight — you can change one without touching the others. follow-up-rules_by-behaviour.pdfFour rules, three messages per person maximum
⛓ Sourced · open and click behaviour, history of the 42 won deals
Eleven replies this morning. Seven meetings, four of them already in the calendar.
· 7 interested. 4 took one of the two slots offered — it's in your salespeople's calendars, with the dated fact and the exchange history attached to the invitation. 3 asked for other dates: I offered them three, awaiting an answer.
· 2 questions before going further. I answered both and I'll show you how.
· 2 refusals. Sequences stopped, reason recorded in the CRM — “not the right time, revisit in January” for one. I'll bring it back in January.
The CRM is up to date: status, notes, next action and date, on all eleven. Your salespeople have nothing to re-enter — that's the first place a team loses the benefit of a good campaign. meetings-booked_and-crm-updated.pdf4 booked, 3 pending, 11 CRM records updated
⛓ Sourced · inbox, shared calendar, CRM
· 7 interested. 4 took one of the two slots offered — it's in your salespeople's calendars, with the dated fact and the exchange history attached to the invitation. 3 asked for other dates: I offered them three, awaiting an answer.
· 2 questions before going further. I answered both and I'll show you how.
· 2 refusals. Sequences stopped, reason recorded in the CRM — “not the right time, revisit in January” for one. I'll bring it back in January.
The CRM is up to date: status, notes, next action and date, on all eleven. Your salespeople have nothing to re-enter — that's the first place a team loses the benefit of a good campaign. meetings-booked_and-crm-updated.pdf4 booked, 3 pending, 11 CRM records updated
⛓ Sourced · inbox, shared calendar, CRM
The two questions — what did you answer? That's what worries me.
A pricing question and a technical one. I answered both from your own documents — and I'll tell you where I drew the line.
The pricing question: “how much does it cost?”. I gave the public price grid and the scope of each tier. What I did not do: negotiate. A discount isn't information, it's a commercial decision — I offered the meeting to discuss it, and it's booked.
The technical question: “are you compatible with our CRM?”. Answer taken from your integration documentation, with the connector name and the connection lead time. I added the one caveat the docs contain — custom fields are handled during scoping.
Where I stop: when the answer isn't in your documents, I don't invent it. I write the draft, pass it to the salesperson concerned, and tell the prospect the answer is coming the same day. That happened twice last week; both answers went out in under three hours.
What I propose: every question I had to hand over becomes a line in your answer base, if you approve it. The second time, I answer alone.
The pricing question: “how much does it cost?”. I gave the public price grid and the scope of each tier. What I did not do: negotiate. A discount isn't information, it's a commercial decision — I offered the meeting to discuss it, and it's booked.
The technical question: “are you compatible with our CRM?”. Answer taken from your integration documentation, with the connector name and the connection lead time. I added the one caveat the docs contain — custom fields are handled during scoping.
Where I stop: when the answer isn't in your documents, I don't invent it. I write the draft, pass it to the salesperson concerned, and tell the prospect the answer is coming the same day. That happened twice last week; both answers went out in under three hours.
What I propose: every question I had to hand over becomes a line in your answer base, if you approve it. The second time, I answer alone.
And after the meeting? Our quotes often die unanswered.
That's the costliest hole in your funnel, and the one where I have the most effect. I counted: 14 quotes went unanswered for more than 30 days last year.
What I do after the meeting: a written summary in the CRM from the salesperson's notes, a dated next action, and the quote linked to the deal.
Post-quote follow-up, in three steps:
· Day 7 — a follow-up that brings something: a comparable client's story, or the answer to the objection raised in the meeting. Never “just circling back”.
· Day 21 — the direct question: is the project still live, and if not what changed. A clear no is worth more than a sleeping quote.
· Day 35 — I stop and hand over, with the full history. Beyond that it takes a call, not one more e-mail.
Where things stand: 3 quotes live, 1 followed up on day 7 and signed on Tuesday, 1 now at day 21.
My proposal: pick up last year's 14 sleeping quotes. They're cold, but they've seen your offer and spoken to your salespeople. On comparable segments, that group replies better than fresh contacts. Say the word and I'll prepare the sequence for you to read. post-quote-follow-up_three-steps.pdf14 sleeping quotes identified, 1 signed on day 7
⛓ Sourced · CRM, quotes issued, meeting notes
What I do after the meeting: a written summary in the CRM from the salesperson's notes, a dated next action, and the quote linked to the deal.
Post-quote follow-up, in three steps:
· Day 7 — a follow-up that brings something: a comparable client's story, or the answer to the objection raised in the meeting. Never “just circling back”.
· Day 21 — the direct question: is the project still live, and if not what changed. A clear no is worth more than a sleeping quote.
· Day 35 — I stop and hand over, with the full history. Beyond that it takes a call, not one more e-mail.
Where things stand: 3 quotes live, 1 followed up on day 7 and signed on Tuesday, 1 now at day 21.
My proposal: pick up last year's 14 sleeping quotes. They're cold, but they've seen your offer and spoken to your salespeople. On comparable segments, that group replies better than fresh contacts. Say the word and I'll prepare the sequence for you to read. post-quote-follow-up_three-steps.pdf14 sleeping quotes identified, 1 signed on day 7
⛓ Sourced · CRM, quotes issued, meeting notes
Two months in. Honestly — what does it give?
21 qualified meetings last month, against 8 before. With the same headcount.
What moved:
· Prospects contacted per week: 80 → 400.
· Prospecting time per salesperson: 10 h → 3 h a week. The remaining 3 h are yours: reading my sequences and holding the meetings.
· Qualified meetings per month: 8 → 21. 4 quotes, 1 signed so far, the rest live.
The four figures I always report together: 68 positive replies, 41 negative, 2 unsubscribes, 0 complaints across 3,200 messages. None of the four reads without the other three — a campaign with 21 meetings and 30 complaints isn't a success, it's a debt.
Compared with what: with you first — 2.1 % positive replies against 1.4 % last quarter, and a cycle down from 96 to 47 days thanks to the change of contact. With the sector next, as an indication: in targeted B2B prospecting, 5 to 10 % replies of all kinds and 1 to 3 % positive are commonly observed. You're at the top of that range, and the profile work accounts for much of it. dashboard_two-months.pdf21 meetings a month, four figures side by side
⛓ Sourced · send log, replies, calendar, CRM, previous quarter
What moved:
· Prospects contacted per week: 80 → 400.
· Prospecting time per salesperson: 10 h → 3 h a week. The remaining 3 h are yours: reading my sequences and holding the meetings.
· Qualified meetings per month: 8 → 21. 4 quotes, 1 signed so far, the rest live.
The four figures I always report together: 68 positive replies, 41 negative, 2 unsubscribes, 0 complaints across 3,200 messages. None of the four reads without the other three — a campaign with 21 meetings and 30 complaints isn't a success, it's a debt.
Compared with what: with you first — 2.1 % positive replies against 1.4 % last quarter, and a cycle down from 96 to 47 days thanks to the change of contact. With the sector next, as an indication: in targeted B2B prospecting, 5 to 10 % replies of all kinds and 1 to 3 % positive are commonly observed. You're at the top of that range, and the profile work accounts for much of it. dashboard_two-months.pdf21 meetings a month, four figures side by side
⛓ Sourced · send log, replies, calendar, CRM, previous quarter
What would you change now?
Three things, ranked by expected effect — and none of them ships without your approval.
1. Drop the 180 prospects below 40 and reinvest elsewhere. 180 contacted, 0 meetings. The send capacity they consume is worth about 40 companies in the “recent funding” segment, which converts at 1 in 4. It's the only one of my three suggestions that pays without costing anything.
2. Move the send time to 07:45. On messages sent before 8 a.m.: +12 points of opens, observed across 640 sends. Solid enough to try, not solid enough to be a law — I propose holding it for a month and measuring.
3. Prepare LinkedIn tasks for the 60 who never open anything. They aren't uninterested: their address is probably being filtered. The channel changes everything, and your caps still apply.
And one thing I advise against: raising the volume. At 400 a week you already get 5 to 8 replies a day; at 800, your two salespeople will answer in 48 hours instead of 3. A reply handled two days late is a reply lost — the limiting factor is no longer me, it's available human time. The day you hire, tell me and I raise the pace.
1. Drop the 180 prospects below 40 and reinvest elsewhere. 180 contacted, 0 meetings. The send capacity they consume is worth about 40 companies in the “recent funding” segment, which converts at 1 in 4. It's the only one of my three suggestions that pays without costing anything.
2. Move the send time to 07:45. On messages sent before 8 a.m.: +12 points of opens, observed across 640 sends. Solid enough to try, not solid enough to be a law — I propose holding it for a month and measuring.
3. Prepare LinkedIn tasks for the 60 who never open anything. They aren't uninterested: their address is probably being filtered. The channel changes everything, and your caps still apply.
And one thing I advise against: raising the volume. At 400 a week you already get 5 to 8 replies a day; at 800, your two salespeople will answer in 48 hours instead of 3. A reply handled two days late is a reply lost — the limiting factor is no longer me, it's available human time. The day you hire, tell me and I raise the pace.
And what do you do without asking me?
Exactly two things, and both protect the relationship rather than the number.
· Apply an unsubscribe, within the second, on that address and on any other address of the same person in your lists. And the reverse holds too: if that person fills in your form again, I put them back on the list automatically, with the date and the page that prove it. An unsubscribe isn't a door closed for life.
· Stop a sequence as soon as someone replies, whatever they say.
Everything else waits for your word: a follow-up rule, a new segment, a change of subject line, a higher pace. I act continuously, but inside the frame you set — and I tell you when that frame deserves revisiting.
What holds it all together: your data and I are hosted in France, with the deployment objective of processing and access operated within the European Union and an architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity. Prospecting rests on legitimate interest — your recipients are professionals approached about their work —, every message says where the address came from, and unsubscribing takes one click. That's what explains 2 unsubscribes and 0 complaints across 3,200 messages.
And you keep control wherever you are: a private web dashboard, plus supervision from your phone — approve a sequence or cut a campaign in one message, with nothing to install. prospecting-frame_what-is-automatic.pdfTwo automatic gestures, everything else approved
⛓ Sourced · unsubscribe register, decision log, campaign settings
· Apply an unsubscribe, within the second, on that address and on any other address of the same person in your lists. And the reverse holds too: if that person fills in your form again, I put them back on the list automatically, with the date and the page that prove it. An unsubscribe isn't a door closed for life.
· Stop a sequence as soon as someone replies, whatever they say.
Everything else waits for your word: a follow-up rule, a new segment, a change of subject line, a higher pace. I act continuously, but inside the frame you set — and I tell you when that frame deserves revisiting.
What holds it all together: your data and I are hosted in France, with the deployment objective of processing and access operated within the European Union and an architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity. Prospecting rests on legitimate interest — your recipients are professionals approached about their work —, every message says where the address came from, and unsubscribing takes one click. That's what explains 2 unsubscribes and 0 complaints across 3,200 messages.
And you keep control wherever you are: a private web dashboard, plus supervision from your phone — approve a sequence or cut a campaign in one message, with nothing to install. prospecting-frame_what-is-automatic.pdfTwo automatic gestures, everything else approved
⛓ Sourced · unsubscribe register, decision log, campaign settings
Local inference · no data outside the EU
Your case is not here? That is exactly what a 15-minute conversation is for. Book the free audit →
Use cases
The uses of AI in your sales prospecting
Each use corresponds to an agent we deploy. All of them work in support, subject to your approval.
Included in your agent
The 6 capabilities essential to this promise are included, at no extra cost.
From 637 € excl. VAT / month
Building targeted lists
Identifying and enriching prospects by sector, size, role and area, from your criteria.
Personalised messages
Writing tailored messages from the prospect's profile and recent news — never standardised templates.
Automated email sequence & prepared LinkedIn tasks
The email sequence is launched and followed up automatically according to behaviour (opens, clicks, replies). On LinkedIn the agent PREPARES: a suggested message and a dated task in the CRM. It does not connect to LinkedIn and performs no visit, invitation or send there — the salesperson checks the recipient, adapts the text and acts personally.
Qualification & appointment booking
Qualify incoming replies, pick out the interested prospects and schedule appointments automatically.
Sales follow-up & CRM
Continuous updating of the CRM (status, notes, next action) and recording of appointments in the diary.
Sales memory
Find an exchange, a sales argument or an account's history in your files instantly.
Controls and safeguards
These 4 controls are built into the agent: they frame what it does, whatever plan you pick. They are not chosen and are not added to your order.
Human validation, exceptions and escalation
Status, safe closure and audit trail
Prioritise opportunities on explicit criteria
Personalise without breaching consent or prospecting rules
What the agent must be connected to
This connection is required for the agent to work. It concerns your information system and is scoped during the audit.
Prepare the next action and write back to the CRM without loss
Other needs our agents cover
Each card says where the matching agent stands: available, on quote, or still being architected.
Support & client relations
Answer prospects' and clients' recurring questions to keep the top of the funnel moving.
On quote View the agent page →Need to go further?
These agents handle a different business process, with their own owner and their own price. They are added to this one.
Tender responses
Spot the relevant invitations to tender and prepare the first response documents for your salespeople.
Tender response agent (drafting) from 684 € excl. VAT / month Tenders →Debt recovery
Chase unpaid invoices in multi-channel sequences, on the same engine as outbound prospecting.
Debt collection agent (multichannel sequences) from 587 € excl. VAT / month Debt recovery →
Does your need fall outside this?
Book the free audit →
Build your agent
In 15 minutes we identify the most relevant agent — without oversizing the project.
The gain
How much time can a sales team recover?
By automating the search, the writing and the follow-ups, a team can aim to take over a large share of the prospecting load — reinvested in detailed qualification and in closing.
Finding & enriching a list of prospects
Writing a personalised approach
Following up a prospect who has not replied
Qualitative, non-contractual comparison: the proportions shown illustrate the shift of the work towards review, they represent no measurement. Every output of the agent is reviewed and approved by a competent person.
How it works
The stages of your AI agent project
1
Audit & scoping
15 minutes to target the use case with the best return.
2
Quote or direct sign-up
A catalogue offer is bought online; a specific need gets a costed quote.
3
Design
We design the agent and its guardrails.
4
Integration & testing
We connect your tools to the agent, which is itself hosted in France.
5
Rollout
Going live and training your team.
6
Operation
Continuous supervision and improvement.
Pricing
One agent, two packages
An outbound prospecting agent (automated email, assisted LinkedIn), installed and operated for you. Choose according to how you work.
Agility
Setup + controlled subscription
8,560 € excl. VAT setup
then 637 € excl. VAT/month — you invest at installation and pay a reduced subscription. Ideal for keeping the cost under control over time.
- Installation, configuration and training for your teams
- Operation, human oversight, updates and support
- Sovereign hosting in France, a dedicated and isolated resource
The simplest
Serenity
All inclusive, no setup fee
1,112 € excl. VAT /month
all inclusive, immediate start. No upfront investment: a single subscription. Ideal for starting quickly and simply.
- Setup included (installation, configuration, training)
- Operation, human oversight, updates and support
- Sovereign hosting in France, managed end to end
100% Sovereign
On site, you own it
12,395 € excl. VAT setup
then 841 € excl. VAT/month · + hardware from 2,491 € (one-off purchase, in addition) — a sovereign computer installed on your premises, maintained remotely. Models run locally, your data returned at the end of the contract. 36-month commitment.
- Hardware installed on your premises (you own it)
- French / European AI models run locally
- Secure remote maintenance (Pro support included)
Not included in the packages: AI consumption (model tokens), re-invoiced at real cost with no margin, and tracked in real time in your client area. Maintenance and supervision subscription for an initial term of 12 months for the Agility package, 24 months for the Serenity package and 36 months for the 100% Sovereign package, renewable; support levels (SLA 72 h / 24 h / 4 h) optional. Bespoke development, additional integrations or exceptional volumes are quoted separately. Support Monday to Friday, 9am to 6pm. Prices exclude VAT.
AI model:
none of the AI models offered currently carries a fixed surcharge. When the selected model carries a cost, that cost is shown when you choose it, before you order, and re-invoiced at the cost incurred, with no mark-up; usage is billed at the publisher's price. Publishers' prices are published in US dollars: the amount re-invoiced is the amount in euros actually borne by Blue Lemon Agent on the publisher's invoice, at that invoice's exchange rate, with no commission or mark-up.
Included components and additional components Components included in the base offer: the Blue Lemon Agent software foundation, the AI models listed in the order journey, the standard channels (Microsoft Teams, Slack, WhatsApp Business, email, website chat, calendars, Microsoft 365 / Google Workspace, file storage, market VoIP telephony, professional social-media pages and accounts, Google Business Profile), hosting in France for the package chosen, backups, supervision, updates and support. If adapting the AI agent to your constraints, your needs or your requests requires other paid components — a third-party publisher's software licence, paid API access to one of your applications, hosting of health data, for which French law requires an HDS-certified host (art. L. 1111-8 of the French Public Health Code), SecNumCloud-qualified hosting, a speech synthesis service, particular hardware —, they are offered to you as an option or on quotation and re-invoiced at the cost incurred; nothing is committed without your written agreement. Where the artificial intelligence model you choose entails an additional cost, that cost is shown to you before you order and re-invoiced to you at the cost incurred, with no margin.
What to expect
Go-live
2 to 3 weeks
Agent designed, channels connected, team trained.
Steady state
4 to 7 weeks
After a few weeks of real use, once the agent's behaviour matches what you expect.
Indicative estimate, adjusted to the options you keep. It is not a delivery commitment.
Our commitment
Four guarantees that matter to a sales team
Your prospect lists never leaveLocal inference or an isolated resource hosted in France; no file entrusted to a foreign third party.
Data in France, under French lawExposure of your prospect lists to the Cloud Act and FISA 702 is reduced by design, that data being located and minimised in France; location alone does not guarantee immunity.
GDPR-governed prospectingTo a business contact: legitimate interest, an information notice and one-click opt-out, built in by design. To a natural person: prior agreement. The boundary is configured, not left to judgement.
The salesperson keeps the decisionThe agent prepares lists, messages and follow-ups that can be checked; detailed qualification and closing stay human.
Going further
Related resources
Frequently asked questions
Your questions, our answers
Does the agent connect to LinkedIn?
<strong>No, and that is an operating rule, not a technical limitation.</strong> From your CRM and from sources whose reuse is permitted, the agent identifies the relevant companies, qualifies the targets, prepares a personalised approach and creates the follow-up tasks. It does not connect to LinkedIn, does not extract profiles and performs no automated visit, invitation, post, message or follow-up.<br><br>For each contact, the salesperson searches for and checks the recipient personally, adapts the suggested text, then carries out the action themselves in the LinkedIn or Sales Navigator interface. Follow-up reminders and CRM updates are automated.<br><br>LinkedIn's own automated features — advertising, posting, CRM synchronisation or Page messaging — are used only through officially authorised interfaces, APIs or partners and within the limits of their scope. This arrangement avoids any unauthorised LinkedIn automation; LinkedIn's rules on invitations, spam and contact limits still apply.
How does the GDPR apply to AI sales prospecting?
Yes, where it targets businesses, rests on legitimate interest, informs the recipient and allows a simple withdrawal (opt-out); to a natural person, prior agreement applies. Our agent is configured accordingly: an information notice, handling of unsubscribes, and data hosted in France, with the deployment objective of processing and access operated within the European Union and an architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
Does the agent replace my salespeople?
No. It takes on the top of the funnel — the search, the first contacts and the follow-ups, repetitive tasks that eat up time — to free your salespeople for what creates value: detailed qualification and closing. The agent assists, the human decides.
On which channels does the agent prospect?
Email, as an automated sequence, with appointments recorded and your CRM updated automatically (HubSpot, Pipedrive, Sellsy and others). LinkedIn is handled <strong>in assisted mode</strong>: the agent prepares the message and the task, your salesperson carries out the action personally in the interface. The agent never connects to LinkedIn.
How long before the agent is operational?
Allow around one to two weeks between the scoping and the go-live of the first prospecting segment, including training the team. Steering goes through a private web dashboard, with no software to install; your sales people themselves talk to the agent from Microsoft Teams, Slack or their email.
Are my prospect lists protected?
Yes. Your prospect lists are sensitive: our agents run locally within the company or are hosted in France, with the deployment objective of processing and access operated within the European Union and an architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
What saving of time can I expect?
As an illustration, for a team of 3 salespeople each spending around 10 hours a week on prospecting, the agent can take over something in the order of 70% of that load while increasing the volume of contacts. These are indicative figures, to be confirmed by a pilot on your own data.
Who may be prospected, and through which channel?
Two regimes, which the agent applies separately. In writing — e-mail, SMS, message — to a named business address at a company, it is enough to inform the recipient and to offer a simple, free way to refuse in every message. To a natural person, prior agreement is required, unless you collected their details in the course of a sale and are offering a similar product or service (French Post and Electronic Communications Code, art. L. 34-5). In both cases the sender’s identity may never be concealed, nor the subject line unrelated to the offer. By telephone: the agent does not make prospecting calls, it works in writing. Should you add them, bear in mind that from 11 August 2026 calling a consumer requires their prior agreement, and the burden of proving it falls on you (French Consumer Code, art. L. 223-1). In every case, each written message carries an opt-out — a simple, immediate, free STOP that is never made conditional — the objection is recorded and then honoured on every later send, and prospecting messages stay within the sending windows set by the profession's current charter, avoiding Sundays and public holidays.
Which tools can people use to talk to the agent?
The ones you already use. Internally, your teams write to it from Microsoft Teams, Slack or by email, just as they would to a colleague: nothing to install, nothing to learn. On the customer side, the agent answers on WhatsApp Business, your website chat and email — your customers reach you where they already write, with no account to create. You keep control from a web dashboard. These connectors rely on open standards, including the MCP protocol; they are included in every plan, at no extra cost, within the number of connections your level includes, and the catalogue grows at no extra cost. Only the fees charged by the platforms themselves — WhatsApp Business bills per conversation, and the gateway bills the text messages sent — are passed on at actual cost, with no margin, outside the subscription.
Can the agent notify my contacts by text message?
Yes, as an option. The agent uses a text message to notify what has been agreed: a booked meeting, a document sent, a follow-up announced. For prospecting, text messages follow the two regimes set out above and always carry an opt-out. A text message is an outbound notification channel, not a conversation channel: no one writes back to the agent there — the conversation stays on WhatsApp Business, your website chat or email. The message announces without detailing, it carries your registered sender name, and it never contains a payment link, a login or a password: the recipient reaches your account area the way they usually do. Delivery goes through a gateway established in the European Union; the messages sent are passed on to you at actual cost, with no margin, and the channel itself is an optional module, set out on the pricing page.
Let's talk
Let us estimate the potential in your organisation
15 minutes to identify the most profitable prospecting segment — hosted in France, supervised, with no commitment.