● B2B offer — Business development
Qualification & follow-up: your criteria, your sequences, your CRM
A lead converts when it is qualified quickly and then followed up at the right rhythm, on the channel it uses. Your agent applies your qualification criteria, prepares the follow-ups by email and text message according to your sequences, and keeps the CRM up to date with every exchange. Hosted in France: your prospect data stays with you. Every send goes out under your control, under the rules you have set.
Hosted in France
Prospect data protected
GDPR & AI Act: governed deployment
Human oversight
Updated on
Deployed in a few weeks
Qualification & follow-up · hosted in France
Where does the follow-up sequence stand on this week's leads?
Every lead is placed in the sequence: stage reached, channel used, date of the last contact.
The next follow-ups are written in your tone, with the context of the previous exchange.
Three leads have replied: they drop out of the sequence and are flagged to a salesperson.
🔗 Sourced · the CRM and the history of exchanges
The next follow-ups are written in your tone, with the context of the previous exchange.
Three leads have replied: they drop out of the sequence and are flagged to a salesperson.
🔗 Sourced · the CRM and the history of exchanges
Can we avoid following up certain accounts?
The exclusions you set are applied: an account that is already a customer, a deal under negotiation, a request not to be contacted.
An incoming reply always breaks the sequence: a salesperson takes over.
✎ Support · exclusions respected, human hand-over
An incoming reply always breaks the sequence: a salesperson takes over.
✎ Support · exclusions respected, human hand-over
Local inference · no data outside the EU
Sequences hosted in France
Sovereign by designLocal inference or hosting in France
GDPR & AI Act: governed deploymentTraceability & human oversight
TurnkeyDesigned, installed and operated for you
The sales management decidesThe agent prepares, never rules
✦ In brief
A Blue Lemon Agent qualification agent applies your criteria, prepares follow-ups by email and text message according to your sequences and keeps your CRM up to date with every exchange. Any incoming reply breaks the sequence and hands over to a salesperson; your exclusions are respected. It runs on local inference or is hosted in France: your prospect data stays with you, architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
100%
hosted in France in the target architecture
0
transfer outside the EU in the target architecture
6
sales uses ready to deploy
0
decision taken without human approval
These figures describe our offer, not results measured at a client. How large the gain is on your volume of leads and number of channels is confirmed by a pilot.
The context
What does an AI agent bring to your business development?
Steadiness is what makes a follow-up sequence perform. That is exactly what an agent keeps up, without flagging, over time.
! The issue
An effective sequence calls for a rhythm kept, context carried from one exchange to the next and a CRM kept up to date. The agent holds all three: it places every lead in the sequence, writes the next follow-up with the context of the last one and records every exchange. Your exclusions and your sending rules frame the whole thing.
✓ Our answer
Your salespeople pick up the leads that have replied, with the full history of the sequence. Any incoming reply breaks the automation: that is where human value begins. Local inference or an isolated resource hosted in France: your prospect data and your exchanges are entrusted to no third party, and consent to being approached is framed with you from the design stage.
The decisive point
Your prospect data and your exchanges: sovereignty & compliance
Your prospect data falls under the GDPR and your sequences bear on the company's image. Here is how they are protected.
Local inference
The agent can run on a machine belonging to your organisation: no prospect data and no exchange leaves the network.
Hosting in France
Otherwise, a dedicated and isolated resource hosted in France, under French law — your leads and your follow-up sequences: processing and access within the European Union targeted by the architecture.
Reduced extraterritorial exposure
For your prospect data and your exchanges, the architecture aims to reduce exposure to the Cloud Act and FISA 702; being located in France or in the European Union does not, on its own, guarantee immunity.
Isolated resource
No pooling: an environment strictly dedicated to your company and its sales sequences.
Exclusions and consent respected
Excluded accounts, requests not to be contacted and incoming replies stop the sequence; encryption, role-based access and logging of every send prepared.
AI Act: governed deployment
The agent is strictly in support; nothing is sent outside an approved sequence and no commercial commitment is made; 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.
- 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.
For strategic accounts and deals under negotiation, SecNumCloud and reinforced hosting are options depending on your requirements. A single architecture is designed to answer both the GDPR and extraterritorial exposure. 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
5 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 companyMarnex Équipement — materials-handling equipment dealer and service provider for industry
- Sector
- Sales and servicing of materials-handling equipment — forklifts, conveyors, racking, maintenance contracts
- Headcount
- 48 staff, including 6 in sales (4 field, 2 inside sales) and 1 sales administrator who keeps the CRM and the follow-ups
- Market served
- Industrial SMEs and logistics platforms in eastern France — 900 active accounts, business-to-business only
- Order of magnitude
- 340 inbound leads a month (trade shows, web forms, campaigns), average order value €14,200, 11-week sales cycle, 23 % of meetings turned into orders
- Tools already in place
- CRM, company mail, email sending platform, short-message gateway, shared calendars and ERP — the agent plugs into them, nothing is replaced
- Who decides
- The sales director sets the qualification criteria, the sequences and the sending rules; each salesperson takes over the moment a lead replies
- Points for improvement
- 190 of the 340 monthly leads never received anything beyond the acknowledgement; first qualification takes 4.2 days; sequence tracking eats 60 % of the sales administrator's time and the CRM records only 41 % of exchanges
Marnex is not short of leads: 340 arrive every month and barely half are ever worked, for lack of time to qualify then follow up at the right pace. The agent is plugged into the CRM, the sending platform, the short-message gateway and the sales calendars: it qualifies, prepares, follows up within the agreed limits, and hands back the moment a person replies. The exchanges below cover two months, from re-reading three years of CRM to the first measured review.
This company, its figures and the exchanges that follow were invented for the demonstration. They illustrate a common situation; they describe no real client.
Lead qualification & follow-up · hosted in France
I re-read three years of your CRM overnight — 6,240 inbound leads, 268 signed orders — and three characteristics clearly separate those who buy from those who never followed up.
A lead is an inbound contact who has shown interest: a form on your site, a business card taken at a trade show, a reply to a campaign.
The gap I measured, and it is the gap that decides the gain: of the 268 orders, 241 had arrived carrying at least two of these three signals; among the leads that went nowhere, only 9 % carried two. Your salespeople know this by instinct — it was written down nowhere, so nobody could sort on their behalf.
The three signals, with their gap:
· A fleet of more than 5 machines — present in 78 % of signed accounts against 12 % of dead leads. The most discriminating signal, and the only one you can check without picking up the phone.
· A logistics manager or site director behind the enquiry, rather than purchasing alone — 71 % against 23 %. When the enquiry is born of a need, it survives the budget round.
· A date named in the enquiry — lease expiry, site opening, an inspection coming up — 64 % against 9 %. A date turns an interest into a project.
What I do with it in practice: I apply these criteria — they are yours, not mine — to every lead that comes in, and I post a score out of 100 with the detail of what built it. First qualification now takes 11 minutes instead of 4.2 days, because it no longer depends on someone being free.
What I propose: that your sales director reads the grid and moves the thresholds as she sees fit. Every criterion comes out with one click, and the score is recalculated across three years of history so you see the effect of the change immediately. qualification-grid_what-builds-the-score.pdf3 criteria, their weight, and the calculation line by line
⛓ Sourced · Marnex CRM, 6,240 leads and 268 orders over 3 years, account records and visit reports
A lead is an inbound contact who has shown interest: a form on your site, a business card taken at a trade show, a reply to a campaign.
The gap I measured, and it is the gap that decides the gain: of the 268 orders, 241 had arrived carrying at least two of these three signals; among the leads that went nowhere, only 9 % carried two. Your salespeople know this by instinct — it was written down nowhere, so nobody could sort on their behalf.
The three signals, with their gap:
· A fleet of more than 5 machines — present in 78 % of signed accounts against 12 % of dead leads. The most discriminating signal, and the only one you can check without picking up the phone.
· A logistics manager or site director behind the enquiry, rather than purchasing alone — 71 % against 23 %. When the enquiry is born of a need, it survives the budget round.
· A date named in the enquiry — lease expiry, site opening, an inspection coming up — 64 % against 9 %. A date turns an interest into a project.
What I do with it in practice: I apply these criteria — they are yours, not mine — to every lead that comes in, and I post a score out of 100 with the detail of what built it. First qualification now takes 11 minutes instead of 4.2 days, because it no longer depends on someone being free.
What I propose: that your sales director reads the grid and moves the thresholds as she sees fit. Every criterion comes out with one click, and the score is recalculated across three years of history so you see the effect of the change immediately. qualification-grid_what-builds-the-score.pdf3 criteria, their weight, and the calculation line by line
⛓ Sourced · Marnex CRM, 6,240 leads and 268 orders over 3 years, account records and visit reports
A score out of 100, fine. But my salespeople will ask where the number comes from, and they will be right to.
They will open the score and see the calculation: three blocks, each with its source and its date. Nothing is weighted by a hunch.
The qualification score turns into a single figure how closely a lead resembles the deals you have already signed — not its value, and not the quality of the person.
The calculation, as it appears on the record:
· Fleet and volume — 40 points. Source: their website, their careers page, the public company register, and your own service visits if the account is already known. Every point carries the line that produced it.
· Role of the enquirer — 35 points. Source: their email signature or the form field. No inference: if the role is not written somewhere, the block stays at zero and I flag it rather than guess.
· Named date — 25 points. Source: the enquirer's exact sentence, quoted on the record. « Our lease ends in March » is worth 25 points, and the quotation is there to be challenged.
The three thresholds, which remain yours: 70 and above, the lead goes straight to a salesperson the same day; between 40 and 69, it enters the follow-up sequence; below 40, it stays on watch and I re-score it every quarter, because a company with no project in January may well have one in September.
The figure that interests me here: of your 268 orders, the grid would have scored 247 above 40 — and 21 below. Those 21 are real deals the score would have underrated: almost all of them are referrals from existing customers, which none of the three criteria capture.
What I do with that, rather than letting it pass: I add a fourth signal — « referred by an active account », worth 30 points, read from the origin field of your CRM. With it, the grid recovers 17 of the 21. You approve that fourth criterion or you don't; it is already calculated. qualification-grid_what-builds-the-score.pdf3 criteria, their weight, and the calculation line by line
⛓ Sourced · qualification records, CRM origin field, history of the 268 orders
The qualification score turns into a single figure how closely a lead resembles the deals you have already signed — not its value, and not the quality of the person.
The calculation, as it appears on the record:
· Fleet and volume — 40 points. Source: their website, their careers page, the public company register, and your own service visits if the account is already known. Every point carries the line that produced it.
· Role of the enquirer — 35 points. Source: their email signature or the form field. No inference: if the role is not written somewhere, the block stays at zero and I flag it rather than guess.
· Named date — 25 points. Source: the enquirer's exact sentence, quoted on the record. « Our lease ends in March » is worth 25 points, and the quotation is there to be challenged.
The three thresholds, which remain yours: 70 and above, the lead goes straight to a salesperson the same day; between 40 and 69, it enters the follow-up sequence; below 40, it stays on watch and I re-score it every quarter, because a company with no project in January may well have one in September.
The figure that interests me here: of your 268 orders, the grid would have scored 247 above 40 — and 21 below. Those 21 are real deals the score would have underrated: almost all of them are referrals from existing customers, which none of the three criteria capture.
What I do with that, rather than letting it pass: I add a fourth signal — « referred by an active account », worth 30 points, read from the origin field of your CRM. With it, the grid recovers 17 of the 21. You approve that fourth criterion or you don't; it is already calculated. qualification-grid_what-builds-the-score.pdf3 criteria, their weight, and the calculation line by line
⛓ Sourced · qualification records, CRM origin field, history of the 268 orders
And all the ones we never followed up? We know they exist, nobody has time to go back to them.
There are 190 a month, and I started with the last twelve: 2,280 leads that received nothing but the acknowledgement. I have brought 41 of them back above your threshold of 70.
How I found them: I went back to every record, re-read the original enquiry word for word and updated what had moved since — a site change, a new warehouse, a new logistics manager, a public tender published. A lead does not sleep, it ages: what was worth nothing in February can be worth 25 points in October.
What that produced:
· 41 leads above 70, handed to the salespeople with fresh context and the original sentence.
· 318 between 40 and 69, ready to enter the sequence.
· 1,921 below 40, on quarterly watch. They are not lost, they are dated.
· 264 removed outright: existing customers, live negotiations, opt-out requests. Those receive nothing, and you are the one who decided that.
What the 41 have already produced, six weeks on: 9 meetings held and 2 orders signed, i.e. €28,400 at your average order value. That figure comes from no new message at all: it comes from records you already had.
What I propose now: that we write the follow-up sequence for the 318 in the middle of the table — that is the seam, not the 41. I'll show you the full sequence, message by message, and you settle the pace. dormant-leads_2280-records-reworked.pdf41 above threshold, 9 meetings, €28,400 signed
⛓ Sourced · 2,280 lead records from the last 12 months, public websites and publications of the companies concerned, sales calendars and orders
How I found them: I went back to every record, re-read the original enquiry word for word and updated what had moved since — a site change, a new warehouse, a new logistics manager, a public tender published. A lead does not sleep, it ages: what was worth nothing in February can be worth 25 points in October.
What that produced:
· 41 leads above 70, handed to the salespeople with fresh context and the original sentence.
· 318 between 40 and 69, ready to enter the sequence.
· 1,921 below 40, on quarterly watch. They are not lost, they are dated.
· 264 removed outright: existing customers, live negotiations, opt-out requests. Those receive nothing, and you are the one who decided that.
What the 41 have already produced, six weeks on: 9 meetings held and 2 orders signed, i.e. €28,400 at your average order value. That figure comes from no new message at all: it comes from records you already had.
What I propose now: that we write the follow-up sequence for the 318 in the middle of the table — that is the seam, not the 41. I'll show you the full sequence, message by message, and you settle the pace. dormant-leads_2280-records-reworked.pdf41 above threshold, 9 meetings, €28,400 signed
⛓ Sourced · 2,280 lead records from the last 12 months, public websites and publications of the companies concerned, sales calendars and orders
Show me the sequence. And I'll warn you: our follow-ups read like a mail-merge today, we know it.
Four messages over fifteen days, two channels, and each one picks up the previous exchange — which is exactly what a mail-merge lacks.
A multichannel sequence is a pre-planned series of messages spread across several channels — here email and short message — where each step depends on what happened at the one before.
The cadence, which you can change line by line:
· D+0 — email. Quotes the exact sentence of the original enquiry and answers it. For 318 leads, 318 different messages: the fleet, the date quoted and the site concerned are never the same.
· D+3 — short message, only to a work number the person gave themselves. Three lines, one closed question, and the name of your service engineer for their area.
· D+8 — email with the technical sheet for the range concerned and a comparable case at a similar-sized manufacturer.
· D+15 — closing email, the one that draws the most replies everywhere: « Should I close the file? » It takes the person out of the sequence whatever they answer.
What I took from your own voice, and where I took it from: I read 1,100 messages sent by your two inside salespeople and kept what sets you apart — short sentences, never first-name terms, the price stated from the second message on, no superlatives. The word « solution » appears in none of your signed messages: I don't use it either.
The time this shifts: writing follow-ups took 40 % of your sales administrator's time; it now takes 8 % — and those 8 % are her proofreading, which catches the detail no rule ever writes down.
What I propose next: that the D+15 message be tested in two versions on two halves of the file. It is the message that decides the overall reply rate, and the only one whose effect can be measured cleanly. follow-up-sequence_4-messages-2-channels.pdfThe four texts, the cadence and the stop rules
⛓ Sourced · 1,100 messages sent by the inside sales team, lead records, Marnex catalogue and technical sheets
A multichannel sequence is a pre-planned series of messages spread across several channels — here email and short message — where each step depends on what happened at the one before.
The cadence, which you can change line by line:
· D+0 — email. Quotes the exact sentence of the original enquiry and answers it. For 318 leads, 318 different messages: the fleet, the date quoted and the site concerned are never the same.
· D+3 — short message, only to a work number the person gave themselves. Three lines, one closed question, and the name of your service engineer for their area.
· D+8 — email with the technical sheet for the range concerned and a comparable case at a similar-sized manufacturer.
· D+15 — closing email, the one that draws the most replies everywhere: « Should I close the file? » It takes the person out of the sequence whatever they answer.
What I took from your own voice, and where I took it from: I read 1,100 messages sent by your two inside salespeople and kept what sets you apart — short sentences, never first-name terms, the price stated from the second message on, no superlatives. The word « solution » appears in none of your signed messages: I don't use it either.
The time this shifts: writing follow-ups took 40 % of your sales administrator's time; it now takes 8 % — and those 8 % are her proofreading, which catches the detail no rule ever writes down.
What I propose next: that the D+15 message be tested in two versions on two halves of the file. It is the message that decides the overall reply rate, and the only one whose effect can be measured cleanly. follow-up-sequence_4-messages-2-channels.pdfThe four texts, the cadence and the stop rules
⛓ Sourced · 1,100 messages sent by the inside sales team, lead records, Marnex catalogue and technical sheets
Then we launch tomorrow morning on the whole trade-show file, 1,240 contacts.
We launch tomorrow morning — and the file is already fit to go, because I spent the night giving it what it lacked. Here is exactly what I did, line by line.
1. The origin of every address, because that is what makes the message defensible. Anyone you approach must be able to find out where you got their details, and that information is owed to them at the latest with the first message. Across the 1,240 lines: 828 already carried their origin (show registration form, badge scanned with consent, form on your own site). 412 carried none. I re-sourced 356 of them one by one from the websites of the companies concerned — the team page, the contact page, the legal notice — and the last 56 are generic departmental addresses of the contact@ kind, whose origin is the public website itself. Not one line lost: 1,240 lines with an origin I can name inside the message.
2. Sorting professionals from private individuals, because the rules are not the same. Writing to a named work address inside a company, on a subject that belongs to that person's job, does not require prior consent: you must inform them and give them, in every message, a simple and free way to refuse. 1,100 lines fall into that category. Writing to a private individual at a personal address does require prior consent — 140 lines of your file are personal addresses, sole traders and owner-managers with no company address. I set those aside and prepared a one-line consent form for them, delivered from your stand by QR link at the next show. 38 have already signed it.
3. What travels inside every message, without exception: your sender identity in plain sight — never disguised, and never a subject line unrelated to the offer —, the sentence stating where the data came from, and the one-click opt-out, free of charge, effective within the minute.
The path I took for the 412, and why that one: each company's own public page, line by line, with the address I took the data from. Slower for me, and entirely indifferent for you: the 412 lines were handled in one night. What stays off that path, and the text says so: scraping a professional directory or a social network to extract its contacts — database law forbids extracting a substantial part of a protected database, and those sites' terms of use forbid it too. You lose not a single line: the file is complete, every address carries its origin, and it goes out tomorrow morning as soon as you approve the sequence. address-origins_1240-lines.pdf828 sourced, 356 re-sourced, 56 generic, 140 set aside
✎ Framework · 1,240-line trade-show file, public websites of the companies concerned, register of consents collected on the stand
1. The origin of every address, because that is what makes the message defensible. Anyone you approach must be able to find out where you got their details, and that information is owed to them at the latest with the first message. Across the 1,240 lines: 828 already carried their origin (show registration form, badge scanned with consent, form on your own site). 412 carried none. I re-sourced 356 of them one by one from the websites of the companies concerned — the team page, the contact page, the legal notice — and the last 56 are generic departmental addresses of the contact@ kind, whose origin is the public website itself. Not one line lost: 1,240 lines with an origin I can name inside the message.
2. Sorting professionals from private individuals, because the rules are not the same. Writing to a named work address inside a company, on a subject that belongs to that person's job, does not require prior consent: you must inform them and give them, in every message, a simple and free way to refuse. 1,100 lines fall into that category. Writing to a private individual at a personal address does require prior consent — 140 lines of your file are personal addresses, sole traders and owner-managers with no company address. I set those aside and prepared a one-line consent form for them, delivered from your stand by QR link at the next show. 38 have already signed it.
3. What travels inside every message, without exception: your sender identity in plain sight — never disguised, and never a subject line unrelated to the offer —, the sentence stating where the data came from, and the one-click opt-out, free of charge, effective within the minute.
The path I took for the 412, and why that one: each company's own public page, line by line, with the address I took the data from. Slower for me, and entirely indifferent for you: the 412 lines were handled in one night. What stays off that path, and the text says so: scraping a professional directory or a social network to extract its contacts — database law forbids extracting a substantial part of a protected database, and those sites' terms of use forbid it too. You lose not a single line: the file is complete, every address carries its origin, and it goes out tomorrow morning as soon as you approve the sequence. address-origins_1240-lines.pdf828 sourced, 356 re-sourced, 56 generic, 140 set aside
✎ Framework · 1,240-line trade-show file, public websites of the companies concerned, register of consents collected on the stand
And the sending itself? I am not going to approve 1,240 messages one by one, nor sign a blank cheque.
Neither: you give me a written mandate, capped and dated, and I work inside it. It is drafted, it fits on one page, and you withdraw it with a single word.
Why a mandate rather than a tick-box: a message goes out under your name and commits your company towards a third party. That is not a technical decision, it is a commercial one — so it belongs to you, and it has to be written down.
What the mandate sets, already filled in for you:
· Scope: leads scored 40 and above, work addresses only, excluding the 264 removed accounts.
· Cap: 300 sends per working day and 4 messages maximum per person, all channels combined. The cap is not technical, it is human: at 300 sends and a 14 % reply rate, your two inside salespeople get 6 replies a day, which they handle the same day. At 800 they would answer in 48 hours — I advise against it, and here is the figure rather than the opinion.
· Hours: never before 8 a.m., never after 8 p.m., never on a Sunday or a public holiday — your own internal rule of 12/01/2026, applied as written.
· Term: from 1 September to 30 November 2026, after which the mandate lapses by itself and has to be renewed.
· Withdrawal: the word « stop » in this console, or a call to your sales administrator. Everything halts within the minute, including messages already queued for the night.
What stays outside the mandate in every case: quoting a firm price, granting a discount, committing to a delivery date. I draft them and a salesperson signs — that is the one place where your signature is worth money.
What changes tomorrow: once signed, the first wave goes out at 8:05 a.m. and you no longer touch it. Across the last eight campaigns your sales administrator launched by hand, the average delay between the decision and the first send was 9 days; it drops below the hour. sending-mandate_capped-dated-withdrawable.pdfScope, cap, hours, term, withdrawal in one word
✎ Framework · draft sending mandate, internal rule of 12/01/2026, history of the last 8 campaigns
Why a mandate rather than a tick-box: a message goes out under your name and commits your company towards a third party. That is not a technical decision, it is a commercial one — so it belongs to you, and it has to be written down.
What the mandate sets, already filled in for you:
· Scope: leads scored 40 and above, work addresses only, excluding the 264 removed accounts.
· Cap: 300 sends per working day and 4 messages maximum per person, all channels combined. The cap is not technical, it is human: at 300 sends and a 14 % reply rate, your two inside salespeople get 6 replies a day, which they handle the same day. At 800 they would answer in 48 hours — I advise against it, and here is the figure rather than the opinion.
· Hours: never before 8 a.m., never after 8 p.m., never on a Sunday or a public holiday — your own internal rule of 12/01/2026, applied as written.
· Term: from 1 September to 30 November 2026, after which the mandate lapses by itself and has to be renewed.
· Withdrawal: the word « stop » in this console, or a call to your sales administrator. Everything halts within the minute, including messages already queued for the night.
What stays outside the mandate in every case: quoting a firm price, granting a discount, committing to a delivery date. I draft them and a salesperson signs — that is the one place where your signature is worth money.
What changes tomorrow: once signed, the first wave goes out at 8:05 a.m. and you no longer touch it. Across the last eight campaigns your sales administrator launched by hand, the average delay between the decision and the first send was 9 days; it drops below the hour. sending-mandate_capped-dated-withdrawable.pdfScope, cap, hours, term, withdrawal in one word
✎ Framework · draft sending mandate, internal rule of 12/01/2026, history of the last 8 campaigns
Where does the sequence stand on this week's leads?
402 leads in sequence, each one placed at its exact step, and 58 have already left it because they replied.
The state of play, channel by channel:
· D+0 sent to 402 · D+3 short message to 214 — the others have no work number on file, and I do not invent one.
· D+8 sent to 289 · D+15 sent to 173.
· 58 replies received, 58 exits from the sequence within the second — nobody who replied receives the next message, and that is the single most important rule in the whole set-up.
· 7 opt-out requests, applied within the second and permanent.
The reply rate: 14.4 %. The reply rate is the share of people reached who write back, whatever the sense of their answer. Your manual campaign in the spring ran at 6.1 % on a comparable file; in written business-to-business outreach, 5 to 10 % is commonly observed — an indicative order of magnitude, not a promise. You are above it, and the gap has a cause: the messages quote the original enquiry word for word, which the spring mail-merge did not.
What the CRM now carries, with nobody typing it: every send, every open, every reply, the current step and the date of the next message. Updating the CRM took 35 % of your sales administrator's time; it now takes 5 % — and those 5 % are her corrections, not her typing.
What I propose: move the short message from D+3 to D+2 on the logistics segment. They reply within 48 hours in 71 % of cases, or never — one day earlier doubles the useful window. sequence-status_402-leads.pdfEvery lead, its step, its date, its channel
⛓ Sourced · sending platform, short-message gateway, CRM, spring manual campaign
The state of play, channel by channel:
· D+0 sent to 402 · D+3 short message to 214 — the others have no work number on file, and I do not invent one.
· D+8 sent to 289 · D+15 sent to 173.
· 58 replies received, 58 exits from the sequence within the second — nobody who replied receives the next message, and that is the single most important rule in the whole set-up.
· 7 opt-out requests, applied within the second and permanent.
The reply rate: 14.4 %. The reply rate is the share of people reached who write back, whatever the sense of their answer. Your manual campaign in the spring ran at 6.1 % on a comparable file; in written business-to-business outreach, 5 to 10 % is commonly observed — an indicative order of magnitude, not a promise. You are above it, and the gap has a cause: the messages quote the original enquiry word for word, which the spring mail-merge did not.
What the CRM now carries, with nobody typing it: every send, every open, every reply, the current step and the date of the next message. Updating the CRM took 35 % of your sales administrator's time; it now takes 5 % — and those 5 % are her corrections, not her typing.
What I propose: move the short message from D+3 to D+2 on the logistics segment. They reply within 48 hours in 71 % of cases, or never — one day earlier doubles the useful window. sequence-status_402-leads.pdfEvery lead, its step, its date, its channel
⛓ Sourced · sending platform, short-message gateway, CRM, spring manual campaign
Some accounts must never be followed up. We have a live €180,000 negotiation that one clumsy message could sink.
Those accounts receive nothing, and they are already out: 264 lines removed before the first send, including your live negotiation.
The four grounds for removal, every one checked against your own data:
· 212 accounts that are already customers, matched on their company registration number and not on their name — « Transports Vallier » and « SAS Vallier Logistique » are the same account, and matching by name would have missed them.
· 34 live negotiations, yours at €180,000 among them. As soon as a quotation above €20,000 is open in the CRM, the whole account leaves every sequence automatically — not just the contact, the entire account, head office and subsidiaries.
· 18 earlier opt-out requests, some dating from 2023. An opt-out does not expire.
· Your competitors and suppliers, from the file your sales director gave me.
And the reverse holds too, because an exclusion is not a door closed for life: if one of those contacts fills in the form on your website again, I put them back on the list with the date and the page on record, and I tell your sales administrator. Someone who comes back of their own accord has changed their mind; it is theirs to say and mine to hear.
The figure that shows what this was costing: across your three manual campaigns in 2025, the hand-made export let 47 customer accounts through, two of whom called their salesperson to complain. Zero this time, across 402 leads.
What I propose: that the automatic removal threshold drop from €20,000 to €8,000. It would cover 61 more deals a year and would cost 61 unsent follow-ups. You decide, I change the rule in a minute. excluded-accounts_264-lines-and-their-grounds.pdf4 grounds, matching by company number, reinstatement
⛓ Sourced · ERP and company registration numbers, open quotations in the CRM, opt-out register, 2025 campaign exports
The four grounds for removal, every one checked against your own data:
· 212 accounts that are already customers, matched on their company registration number and not on their name — « Transports Vallier » and « SAS Vallier Logistique » are the same account, and matching by name would have missed them.
· 34 live negotiations, yours at €180,000 among them. As soon as a quotation above €20,000 is open in the CRM, the whole account leaves every sequence automatically — not just the contact, the entire account, head office and subsidiaries.
· 18 earlier opt-out requests, some dating from 2023. An opt-out does not expire.
· Your competitors and suppliers, from the file your sales director gave me.
And the reverse holds too, because an exclusion is not a door closed for life: if one of those contacts fills in the form on your website again, I put them back on the list with the date and the page on record, and I tell your sales administrator. Someone who comes back of their own accord has changed their mind; it is theirs to say and mine to hear.
The figure that shows what this was costing: across your three manual campaigns in 2025, the hand-made export let 47 customer accounts through, two of whom called their salesperson to complain. Zero this time, across 402 leads.
What I propose: that the automatic removal threshold drop from €20,000 to €8,000. It would cover 61 more deals a year and would cost 61 unsent follow-ups. You decide, I change the rule in a minute. excluded-accounts_264-lines-and-their-grounds.pdf4 grounds, matching by company number, reinstatement
⛓ Sourced · ERP and company registration numbers, open quotations in the CRM, opt-out register, 2025 campaign exports
And in the CRM, what exactly do you write? We once had a tool that polluted every record we owned.
Seven fields, not one more, and each carries its source and its date. What has no source is not written.
What I write: the current sequence step · the date and channel of the last contact · the date of the next message · the score and the detail behind it · the original sentence of the enquiry · the exit status and its grounds · a link to the exact message that went out, so your salesperson reads what their prospect read before picking up the phone.
What I never write on a record: a judgement about a person, a supposed intention, a « difficult customer ». A CRM is a shared memory: what goes into it ends up in front of the customer the day they ask what you hold about them.
What I never overwrite: a field filled in by a human. If your salesperson noted « call back after the holidays », I put my note beside it, dated, and theirs stands. That is precisely where your previous tool did the damage.
The unflattering figure, and I give it before anyone asks: in the first month, 3.1 % opt-outs on the « purchasing » segment, against 0.6 % everywhere else. Five times more — and the message is not the culprit. The cause is targeting: I was sending purchasing managers the message written for logistics managers — a text about parts availability aimed at people whose job is price.
What I did with it, and the measurement after the fix: I wrote a variant for the purchasing role — total cost of ownership, contract term, trade-in conditions — and ran it over the following fifteen days. Opt-outs on the segment fell back to 0.7 %, and the segment's reply rate went from 4.8 % to 11.2 %. The figure that troubled me became the segment that improved most. crm-log_7-fields-and-their-source.pdfWhat is written, what is never overwritten, 3.1 % → 0.7 %
⛓ Sourced · CRM write log, opt-out rate by segment, both versions of the purchasing message
What I write: the current sequence step · the date and channel of the last contact · the date of the next message · the score and the detail behind it · the original sentence of the enquiry · the exit status and its grounds · a link to the exact message that went out, so your salesperson reads what their prospect read before picking up the phone.
What I never write on a record: a judgement about a person, a supposed intention, a « difficult customer ». A CRM is a shared memory: what goes into it ends up in front of the customer the day they ask what you hold about them.
What I never overwrite: a field filled in by a human. If your salesperson noted « call back after the holidays », I put my note beside it, dated, and theirs stands. That is precisely where your previous tool did the damage.
The unflattering figure, and I give it before anyone asks: in the first month, 3.1 % opt-outs on the « purchasing » segment, against 0.6 % everywhere else. Five times more — and the message is not the culprit. The cause is targeting: I was sending purchasing managers the message written for logistics managers — a text about parts availability aimed at people whose job is price.
What I did with it, and the measurement after the fix: I wrote a variant for the purchasing role — total cost of ownership, contract term, trade-in conditions — and ran it over the following fifteen days. Opt-outs on the segment fell back to 0.7 %, and the segment's reply rate went from 4.8 % to 11.2 %. The figure that troubled me became the segment that improved most. crm-log_7-fields-and-their-source.pdfWhat is written, what is never overwritten, 3.1 % → 0.7 %
⛓ Sourced · CRM write log, opt-out rate by segment, both versions of the purchasing message
When someone replies, what actually happens? That is where we lose deals.
The sequence stops within the second, the reply goes to a named salesperson with the whole history, and a meeting is offered before they have opened their inbox.
This month's 58 replies, sorted:
· 39 asking for a call or a meeting. I offer three slots in the calendar of the salesperson for that area and book the meeting as soon as one is picked. 39 out of 39 had a slot offered within 20 minutes.
· 12 « not now, call me in September ». That is not a refusal, it is a date: the lead leaves the sequence, the record carries the date, and I put it back in front of the salesperson on 1 September with what they said.
· 7 opt-out requests, applied within the second, with no commercial confirmation message — answering an opt-out with an offer is the surest way to turn a departure into a complaint.
What the salesperson receives, and this is the heart of the handover: the original enquiry, the four messages sent in order, the reply received, the score and its detail, the company's fleet, the date they quoted, and the two service visits you already made there in 2024 — which nobody would have connected by hand.
The delay, which is the whole point: a reply used to wait 1.8 days before reaching a salesperson. It now reaches one in 20 minutes, qualified. Of the five deals lost at first contact last year, three went to a competitor who called back the same day.
What I propose: that the area salesperson gets a phone alert for replies from leads scored above 70, and nothing for the rest. An alert that rings for everything soon rings for nothing. handover_58-replies-sorted.pdfWhat the salesperson receives, in 20 minutes instead of 1.8 days
⛓ Sourced · reply inboxes, sales calendars, service visit history, deals lost in 2025
This month's 58 replies, sorted:
· 39 asking for a call or a meeting. I offer three slots in the calendar of the salesperson for that area and book the meeting as soon as one is picked. 39 out of 39 had a slot offered within 20 minutes.
· 12 « not now, call me in September ». That is not a refusal, it is a date: the lead leaves the sequence, the record carries the date, and I put it back in front of the salesperson on 1 September with what they said.
· 7 opt-out requests, applied within the second, with no commercial confirmation message — answering an opt-out with an offer is the surest way to turn a departure into a complaint.
What the salesperson receives, and this is the heart of the handover: the original enquiry, the four messages sent in order, the reply received, the score and its detail, the company's fleet, the date they quoted, and the two service visits you already made there in 2024 — which nobody would have connected by hand.
The delay, which is the whole point: a reply used to wait 1.8 days before reaching a salesperson. It now reaches one in 20 minutes, qualified. Of the five deals lost at first contact last year, three went to a competitor who called back the same day.
What I propose: that the area salesperson gets a phone alert for replies from leads scored above 70, and nothing for the rest. An alert that rings for everything soon rings for nothing. handover_58-replies-sorted.pdfWhat the salesperson receives, in 20 minutes instead of 1.8 days
⛓ Sourced · reply inboxes, sales calendars, service visit history, deals lost in 2025
You book meetings in my salespeople's calendars? They will take that badly.
I only book inside the slots they opened themselves, and each of them sets their own. A calendar is territory: I offer inside it, I do not redraw it.
How it works in practice: each salesperson declares their meeting slots — Julien Marchais opened Tuesday and Thursday afternoons, Sofia Renard prefers mornings and blocks Fridays. I offer only within those, and I account for travel time: two meetings 90 kilometres apart do not follow each other an hour apart.
What happens if the slot does not suit: the prospect suggests another, I pass it to the salesperson and it is they who accept. I never move a meeting already booked.
The result over two months: 46 meetings booked, against 22 in the equivalent two months last year. 24 more.
What that is worth, using your figures and not mine: you turn 23 % of meetings into orders, at an average order value of €14,200. 24 extra meetings represent 5.5 orders, i.e. roughly €78,000 of potential revenue over two months. Potential, not banked: your salespeople are the ones who sign, and the 11-week cycle means half of it will show up in the following quarter.
What it is worth elsewhere, and this is the figure people forget: the time given back on tracking, writing and the CRM comes to 18 hours a week for your sales administrator and your two inside salespeople — half a full-time post. At their loaded hourly cost, about €2,500 a month of time returned, against a subscription of €597 a month.
What I propose next: that once the meeting is held, the commercial proposal be drafted straight away from the visit report and the fleet surveyed on site. That is not my job — it belongs to the commercial proposals agent, and I hand it the complete file. meetings_46-booked-in-two-months.pdf22 → 46, and the potential worked out on your own figures
⛓ Sourced · calendars and declared slots, meetings booked over two months, conversion rate and average order value from the CRM
How it works in practice: each salesperson declares their meeting slots — Julien Marchais opened Tuesday and Thursday afternoons, Sofia Renard prefers mornings and blocks Fridays. I offer only within those, and I account for travel time: two meetings 90 kilometres apart do not follow each other an hour apart.
What happens if the slot does not suit: the prospect suggests another, I pass it to the salesperson and it is they who accept. I never move a meeting already booked.
The result over two months: 46 meetings booked, against 22 in the equivalent two months last year. 24 more.
What that is worth, using your figures and not mine: you turn 23 % of meetings into orders, at an average order value of €14,200. 24 extra meetings represent 5.5 orders, i.e. roughly €78,000 of potential revenue over two months. Potential, not banked: your salespeople are the ones who sign, and the 11-week cycle means half of it will show up in the following quarter.
What it is worth elsewhere, and this is the figure people forget: the time given back on tracking, writing and the CRM comes to 18 hours a week for your sales administrator and your two inside salespeople — half a full-time post. At their loaded hourly cost, about €2,500 a month of time returned, against a subscription of €597 a month.
What I propose next: that once the meeting is held, the commercial proposal be drafted straight away from the visit report and the fleet surveyed on site. That is not my job — it belongs to the commercial proposals agent, and I hand it the complete file. meetings_46-booked-in-two-months.pdf22 → 46, and the potential worked out on your own figures
⛓ Sourced · calendars and declared slots, meetings booked over two months, conversion rate and average order value from the CRM
And if a prospect asks you for a firm price in their reply? That happens all the time.
I answer within the hour with list price and published terms — and the firm price goes out under a salesperson's signature, drafted, costed, ready to send.
What I give myself, because it is already public: the list price for the range, the general terms, standard lead times, the service rate card. None of that commits you any more than your website already does.
What I draft and do not sign: a discount, a guaranteed lead time, a firm price, a trade-in. That is not squeamishness: those four lines bind you contractually, and the signature is what you sell. I build the full quotation — configuration, proposed discount with its justification, resulting margin displayed — and your salesperson signs or corrects. Of the 39 requests for a call, 17 included a price request, and all 17 quotations were ready on average 26 minutes after the prospect's reply.
What that changes in fact: your average time to send a quotation was 3.4 days. It is now 4 hours, the time for a salesperson to read and sign. Over an 11-week cycle three days change nothing; for a prospect who wrote to three suppliers the same morning, they change everything.
And if you want to go further: you can give me a capped discount allowance — say 5 % up to €20,000, nothing beyond without a salesperson. The mandate is written like the sending one: capped, dated, withdrawable in a word. Until you sign it, I stop at the draft, and the quotation waits, complete, on the record. handover_58-replies-sorted.pdfWhat the salesperson receives, in 20 minutes instead of 1.8 days
⛓ Sourced · published list price and general terms, this month's 39 call requests, history of quotation lead times
What I give myself, because it is already public: the list price for the range, the general terms, standard lead times, the service rate card. None of that commits you any more than your website already does.
What I draft and do not sign: a discount, a guaranteed lead time, a firm price, a trade-in. That is not squeamishness: those four lines bind you contractually, and the signature is what you sell. I build the full quotation — configuration, proposed discount with its justification, resulting margin displayed — and your salesperson signs or corrects. Of the 39 requests for a call, 17 included a price request, and all 17 quotations were ready on average 26 minutes after the prospect's reply.
What that changes in fact: your average time to send a quotation was 3.4 days. It is now 4 hours, the time for a salesperson to read and sign. Over an 11-week cycle three days change nothing; for a prospect who wrote to three suppliers the same morning, they change everything.
And if you want to go further: you can give me a capped discount allowance — say 5 % up to €20,000, nothing beyond without a salesperson. The mandate is written like the sending one: capped, dated, withdrawable in a word. Until you sign it, I stop at the draft, and the quotation waits, complete, on the record. handover_58-replies-sorted.pdfWhat the salesperson receives, in 20 minutes instead of 1.8 days
⛓ Sourced · published list price and general terms, this month's 39 call requests, history of quotation lead times
Two months in. What has this actually given us back?
Eighteen hours a week, 24 more meetings, and 340 leads worked every month instead of 150.
On the three items you were already measuring:
· Sequence tracking: 60 % → 8 % of your sales administrator's time. The biggest gain, and the most mechanical.
· Writing follow-ups: 40 % → 8 %. The 8 % is her proofreading.
· Updating the CRM: 35 % → 5 %. The CRM went from 41 % of exchanges recorded to 100 %, without a single line being typed by hand.
What those hours became: your sales administrator has taken back payment chasing and warranty files, both left waiting for a year. It was not my decision, and it is the best possible use of the time I give back.
The commercial result, on your own figures: 46 meetings against 22, €78,000 of potential revenue on the 24 extra meetings, €28,400 already signed on the dormant leads reworked. The agent's cost over the period: €1,980.
The figure that does not flatter me, and it stays: of 402 leads in sequence, 173 went all the way to the fourth message without ever reacting — 43 %. Four messages for nothing is not neutral: it is attention consumed among people who asked for nothing.
What I do with it: I compared those 173 with the 58 who replied. The difference is neither the message nor the sector: it is the age of the enquiry. Beyond 90 days between the original enquiry and the first message, the reply rate falls from 14.4 % to 3.1 %. My proposal: take leads older than 90 days out of the sequence after the second message instead of the fourth. You lose 0.4 points of reply rate, you save 340 messages a month, and your recipients will thank you for it well beyond this quarter. two-month-review_what-marnex-got-back.pdf18 h a week, 46 meetings, 43 % non-reactive and its fix
⛓ Sourced · time records of the sales administrator and inside sales, CRM, calendars, subscription invoicing
On the three items you were already measuring:
· Sequence tracking: 60 % → 8 % of your sales administrator's time. The biggest gain, and the most mechanical.
· Writing follow-ups: 40 % → 8 %. The 8 % is her proofreading.
· Updating the CRM: 35 % → 5 %. The CRM went from 41 % of exchanges recorded to 100 %, without a single line being typed by hand.
What those hours became: your sales administrator has taken back payment chasing and warranty files, both left waiting for a year. It was not my decision, and it is the best possible use of the time I give back.
The commercial result, on your own figures: 46 meetings against 22, €78,000 of potential revenue on the 24 extra meetings, €28,400 already signed on the dormant leads reworked. The agent's cost over the period: €1,980.
The figure that does not flatter me, and it stays: of 402 leads in sequence, 173 went all the way to the fourth message without ever reacting — 43 %. Four messages for nothing is not neutral: it is attention consumed among people who asked for nothing.
What I do with it: I compared those 173 with the 58 who replied. The difference is neither the message nor the sector: it is the age of the enquiry. Beyond 90 days between the original enquiry and the first message, the reply rate falls from 14.4 % to 3.1 %. My proposal: take leads older than 90 days out of the sequence after the second message instead of the fourth. You lose 0.4 points of reply rate, you save 340 messages a month, and your recipients will thank you for it well beyond this quarter. two-month-review_what-marnex-got-back.pdf18 h a week, 46 meetings, 43 % non-reactive and its fix
⛓ Sourced · time records of the sales administrator and inside sales, CRM, calendars, subscription invoicing
Our prospect files, our exchanges, our CRM: where does all of it go?
Nowhere that leaves France. Your prospect data and your exchanges live on local inference on a Marnex machine, or on an isolated resource hosted in France under French law.
Local inference means the model computes on your machine: the text of a message or a lead record crosses no external network to be processed. If you would rather not host a machine, the second route is a resource strictly dedicated to Marnex — no pooling with another company, and none of your competitors shares your environment.
What that guarantees, point by point:
· Processing and access within the European Union targeted by the architecture, with exposure to the Cloud Act and FISA 702 reduced by design — location alone does not guarantee immunity, including where an American provider hosts in Europe.
· Your exchanges train no model. What I learn of your inside sales team's voice serves Marnex and nobody else.
· Encryption in transit and at rest, and role-based access — rights follow the job: a salesperson opens the leads for their area, not the whole file and not the margins.
· A full log of every message prepared and every message sent: who, when, which text, which address origin, which exit grounds. That log is what lets you answer, in ten minutes, someone who asks where you got their address — and that request does come in, on average three times a quarter at a company your size.
The figure that makes this commercial rather than legal: on your last two framework maintenance tenders, hosting the data in France was a scored criterion. You were able to answer in writing, with evidence, instead of a declaration on honour.
What I propose: that I keep up to date the processing record your major account asks for every year at renewal — hosting, subcontractors, retention periods, who accesses what. It is requested once a year and takes three days to assemble. technical-framework_where-your-prospect-data-lives.pdfLocal inference, role-based access, processing in the EU targeted
✎ Framework · deployment architecture, access and sending logs, files from the last two tenders
Local inference means the model computes on your machine: the text of a message or a lead record crosses no external network to be processed. If you would rather not host a machine, the second route is a resource strictly dedicated to Marnex — no pooling with another company, and none of your competitors shares your environment.
What that guarantees, point by point:
· Processing and access within the European Union targeted by the architecture, with exposure to the Cloud Act and FISA 702 reduced by design — location alone does not guarantee immunity, including where an American provider hosts in Europe.
· Your exchanges train no model. What I learn of your inside sales team's voice serves Marnex and nobody else.
· Encryption in transit and at rest, and role-based access — rights follow the job: a salesperson opens the leads for their area, not the whole file and not the margins.
· A full log of every message prepared and every message sent: who, when, which text, which address origin, which exit grounds. That log is what lets you answer, in ten minutes, someone who asks where you got their address — and that request does come in, on average three times a quarter at a company your size.
The figure that makes this commercial rather than legal: on your last two framework maintenance tenders, hosting the data in France was a scored criterion. You were able to answer in writing, with evidence, instead of a declaration on honour.
What I propose: that I keep up to date the processing record your major account asks for every year at renewal — hosting, subcontractors, retention periods, who accesses what. It is requested once a year and takes three days to assemble. technical-framework_where-your-prospect-data-lives.pdfLocal inference, role-based access, processing in the EU targeted
✎ Framework · deployment architecture, access and sending logs, files from the last two tenders
One last thing: what do you do without asking us? I want the list, not a principle.
Three actions, not one more, and all three protect the relationship rather than the numbers. All three are reversible.
· Applying an opt-out within the second, with nobody reviewing it. And the reverse holds too: if the person fills in your form again, I put them back on the list, with the date and the page on record. An opt-out is not a door closed for life.
· Taking a lead out of the sequence the moment they reply, whatever the sense of their answer. Including an out-of-office: I read it, I postpone, and I do not count it as a reply.
· Taking a whole account out of every sequence as soon as a quotation opens in the CRM. The account comes back on its own when the deal closes, won or lost.
Everything else waits for a decision: launching a campaign, changing a score threshold, adding a channel, writing to a private individual, quoting a firm price, granting a discount. The full list fits on one page, with the job title that decides written opposite each line.
And the three neighbouring agents, because the chain does not end with me:
· Upstream, the CRM enrichment agent completes the records and prioritises before I score. Of the 1,240 trade-show lines, 412 had neither a role nor a phone number: that is its job, and it halves my « role of the enquirer » block left at zero.
· Downstream, the commercial proposals agent takes the file over after the meeting. I hand it the visit report, the fleet surveyed and the service history.
· Alongside, the sales reporting agent tracks results by area, range, salesperson and channel. It reads my logs, it does not rewrite them.
What I propose to close: an eight-week pilot on a single segment, your largest, with the same three measurements as today — tracking, writing, CRM. If the figures do not move, you will have found out in eight weeks and for the price of two months. who-decides-what_3-automatic-actions.pdfWhat goes out alone, what waits for a decision, and who takes it
✎ Framework · list of automatic actions, decision matrix approved by the sales director, scope of the neighbouring agents
· Applying an opt-out within the second, with nobody reviewing it. And the reverse holds too: if the person fills in your form again, I put them back on the list, with the date and the page on record. An opt-out is not a door closed for life.
· Taking a lead out of the sequence the moment they reply, whatever the sense of their answer. Including an out-of-office: I read it, I postpone, and I do not count it as a reply.
· Taking a whole account out of every sequence as soon as a quotation opens in the CRM. The account comes back on its own when the deal closes, won or lost.
Everything else waits for a decision: launching a campaign, changing a score threshold, adding a channel, writing to a private individual, quoting a firm price, granting a discount. The full list fits on one page, with the job title that decides written opposite each line.
And the three neighbouring agents, because the chain does not end with me:
· Upstream, the CRM enrichment agent completes the records and prioritises before I score. Of the 1,240 trade-show lines, 412 had neither a role nor a phone number: that is its job, and it halves my « role of the enquirer » block left at zero.
· Downstream, the commercial proposals agent takes the file over after the meeting. I hand it the visit report, the fleet surveyed and the service history.
· Alongside, the sales reporting agent tracks results by area, range, salesperson and channel. It reads my logs, it does not rewrite them.
What I propose to close: an eight-week pilot on a single segment, your largest, with the same three measurements as today — tracking, writing, CRM. If the figures do not move, you will have found out in eight weeks and for the price of two months. who-decides-what_3-automatic-actions.pdfWhat goes out alone, what waits for a decision, and who takes it
✎ Framework · list of automatic actions, decision matrix approved by the sales director, scope of the neighbouring agents
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
What does the agent actually do?
One agent, the whole qualification cycle. All these uses work in support, subject to your approval.
Included in your agent
The 3 capabilities essential to this promise are included, at no extra cost.
From 597 € excl. VAT / month
Qualification against your criteria
Applies your qualification criteria and sets out the detail used.
Multi-channel follow-ups
Prepares emails and text messages according to your sequences, in your tone.
A CRM kept up to date
Records every exchange and places the lead in the sequence.
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
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.
CRM enrichment
Upstream, a dedicated agent completes the records and prioritises the leads.
CRM agent (lead enrichment & scoring) from 542 € excl. VAT / month CRM enrichment →Sales proposals
Once the lead is qualified, a dedicated proposals agent prepares the offer.
Sales proposal agent from 489 € excl. VAT / month Sales proposals →Sales steering
To track results by dimension, a dedicated steering agent completes the picture.
Sales steering agent (KPIs, alerts) from 474 € excl. VAT / month Sales steering →
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 many leads can a team follow without losing any?
By taking on the sequence and the CRM updates, the effort shifts towards the leads that have replied. How large the gain is depends on your volume and remains to be confirmed by a pilot.
Following the follow-up sequence
Writing the follow-ups
CRM updating
Indicative figures, not contractual, to be confirmed by a pilot on your volume of leads and number of channels. An incoming reply always breaks the sequence: what follows in the relationship belongs to a salesperson.
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 package, one agent
A qualification and follow-up agent (criteria, sequences, CRM), installed and operated for you.
Agility
Setup + controlled subscription
7,165 € excl. VAT setup
then 597 € 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
997 € 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
10,740 € excl. VAT setup
then 801 € 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 your prospecting
Your prospect data stays with youLocal inference or an isolated resource hosted in France; no prospect data entrusted to a third party, no exchange used to train a model.
Data in France, under French lawYour prospect data and your exchanges: minimisation and location in France, architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
The sales management keeps the decisionThe agent produces qualified leads and prepared follow-ups, which can be checked and altered; no approval is automated.
Human oversight & traceabilityOn your volume of leads and number of channels: systematic logging and tracking, in line with the AI Act.
Going further
Related resources
Frequently asked questions
Your questions, our answers
Do the follow-ups go out automatically?
According to the rules you set: sent after approval, or sent automatically within a scope you define. In every case, exclusions and requests not to be contacted are respected.
What happens when a prospect replies?
The sequence stops and the lead is flagged to a salesperson, with the full history. A human reply calls for a human answer.
How is the GDPR respected?
The lawful bases for approaching prospects, the information notices and the requests not to be contacted are framed with you at the design stage, and applied by the agent without exception.
Which channels does it work on?
Email and text message, synchronised with your CRM. The exact scope is settled at the design stage.
Is our prospect data protected?
Yes. The agent is hosted in France, on local inference or an isolated resource, 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. Your data is not used to train a third-party model.
How long does it take to deploy this agent?
A few months as a rule, depending on the volume of leads, the number of channels and the integration with your CRM, after a free audit then phases of design, integration and testing.
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.
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.
Related use cases
Other agents for sales
Let's talk
Let's size up the potential in your prospecting
15 minutes to frame your criteria and your sequences — hosted in France, supervised, with no commitment.