+33 (0)1 87 66 00 65 · Monday to Friday, 9am–6pm Free audit (15 min)
● Business offer — Human resources

Preparing appraisals: your frameworks, your materials ready

A good appraisal is prepared, and preparing it mostly means gathering what already exists: the company's framework, the previous record, the objectives set. Your agent brings those elements together in your template, recalls the points left open and proposes the questions from your framework. Hosted in France: your employees' data stays with you. The agent expresses no assessment — evaluating a person rests with the manager, and with them alone.

Hosted in France Employee data protected GDPR & AI Act: governed deployment Human oversight

Updated on

Deployed in a few weeks
Appraisal preparation · hosted in France
Prepare the material for Wednesday's annual appraisal.
Material prepared in your company framework: sections in the usual order, with the corresponding questions.
The record of the previous appraisal is recalled, along with the points left open and the objectives that had been set.
The assessment areas are empty: they are yours.
✎ Action · material ready, assessment yours
Can you tell me whether the objectives were met?
I recall the objectives as they were worded and the factual elements you have recorded through the year.
Judging whether they were met belongs to the manager: it concerns a person and may affect their career.
⚖ AI Act · no automated assessment of a person
Local inference · no data outside the EU
HR data 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 manager decidesThe agent prepares, never rules
✦ In brief

A Blue Lemon Agent appraisal preparation agent brings together in your framework what already exists — the previous record, the objectives set, the points left open — and proposes the corresponding questions. It expresses no assessment and rates no one: evaluating an employee may affect their career and rests with the manager. It runs on local inference or is hosted in France: your employees' data is entrusted to no one, 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
HR 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 number of appraisals is confirmed by a pilot.

The context

What does an AI agent bring to preparing your appraisals?

The quality of an appraisal comes from its preparation. Gathering the framework, the history and the objectives takes time; the conversation itself takes no less.

! The issue

A well-prepared appraisal is one where the manager arrives with the history in mind and the company framework in front of them. The agent brings those elements together — the previous record, the objectives set, the points left hanging — in your template, so that the manager's time goes to the conversation rather than to reconstructing the file.

Our answer

The material arrives complete and faithful to your framework, with the history recalled and the assessment areas empty. That is an essential distinction: the agent prepares the material, the manager makes the judgement. Local inference or an isolated resource hosted in France, logging: your employees' data does not leave the company and no assessment is produced automatically.

The decisive point

Data relating to your employees: sovereignty & compliance

Data relating to your employees is among the most sensitive a company handles. Here is how the architecture of our agents protects it.

Local inference

The agent can run on a machine belonging to your organisation: no employee data leaves the network, no record passes through a public cloud.

Hosting in France

Otherwise, a dedicated and isolated resource hosted in France, under French law — your appraisal frameworks and your records: processing and access within the European Union targeted by the architecture.

Reduced extraterritorial exposure

For data relating to your employees, 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 appraisal frameworks.

No assessment produced

The agent gathers and recalls; the assessment areas stay empty and every preparation is logged, with access restricted to the managers concerned.

AI Act: governed deployment

The agent is strictly in support; no evaluation or rating of an employee is automated; 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 events logged, their content, their retention period and who may access them are defined for the deployment chosen.
For appraisals tied to disciplinary proceedings or a reorganisation, 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

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.

Appraisal preparation · hosted in France
Four flags this morning on your appraisals.
· Twenty-three employees have had no appraisal for more than two years against the frequency set by your procedure note of 12/01/2025. Four are beyond six years.
· Fourteen training requests made at appraisal have never received an answer, written or spoken. The oldest dates from 2022.
· Of last year's 118 records, 72 carry objectives dated the day of the appraisal. They were not set during the year: they were written during the meeting.
· Eleven employees have changed manager since their last appraisal. Their new manager does not have the previous record. morning-watch_4-flags.pdf4 flags · 14 requests with no answer
⛓ Source · procedure note of 12/01/2025, 118 records, appraisal register, org chart
The fourteen training requests with no answer?
Fourteen people were asked what they wanted, and nobody answered them.
What I record: fourteen requests written in the record's dedicated section. None carries any follow-up — no agreement, no refusal, no "to be revisited", no referral to the training plan.
What I do not say: that they should have been accepted. A request can be refused, and a reasoned refusal is an answer.
What I do record: of the fourteen, six employees restated the same request at the following appraisal. Four dropped it. And four had no following appraisal.
What that does to the coming meeting: the manager will reopen a form containing a 2022 request that was never answered. If they do not know before going in, they learn it in front of the employee — with nothing to say.
What I propose: the list of fourteen, with the date of the request and the fact that no follow-up appears. Before the meeting, not during it. 14-requests_0-answers.pdf6 restated · 4 dropped · 4 with no next appraisal
⛓ Source · 118 records, training section, appraisal register
How did you spot all that? And who do you tell?
I watch, continuously, what you have opened to me: appraisal records, the register of dates, your procedure note, and the org chart.
Routing follows what can still be of use: an overdue frequency goes to HR, not to the manager — it is an organisation, not an individual fault; a request with no answer to the manager, before the meeting, never after; a change of manager to the new manager, with what the employee is entitled to see; an objective dated the day of the meeting to whoever designs the form.
With a chase: 7 days before the meeting, then the day before. Then a monthly summary: by department and type of gap, never by employee and never by manager.

Three operating rules, and the first governs the rest. The appraisal is written and signed by the manager, in the room — no rating proposed, no judgement from an earlier appraisal carried forward: an opinion that outlives whoever formed it can no longer be challenged by anyone. What I hand over before they walk in is complete: the objectives as they were recorded, with their dates, the training requests left unanswered and for how long, the history of reporting lines, and the date of the last appraisal. The manager walks in knowing; they no longer discover a 2022 request in front of the employee. And what you have not opened to me is invisible: I name it rather than assume it does not exist.
✎ Framework · no wording, no rating, no judgement carried forward
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, several kinds of appraisal to prepare. 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 526 € excl. VAT / month

Material in your framework

Prepares the document with the company's sections and the corresponding questions.

Recalling the history

Takes up the previous record, the objectives set and the points left open.

Points to clarify

Gathers the factual elements recorded, without drawing any conclusion from them.

Controls and safeguards These 5 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 Apply transparent criteria and avoid automated decisions Preserve human validation, recourse and audit trail Protect HR data with fine-grained access and retention periods
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.
Integrate with the HR information system/ATS and measure quality, turnaround time and fairness

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.

Does your need fall outside this?

In 15 minutes we identify the most relevant agent — without oversizing the project.

Book the free audit Build your agent
The gain

How much conversation time can a manager recover?

By gathering the history and the framework, the effort shifts from reconstructing the file to the conversation itself. How large the gain is depends on your volume and remains to be confirmed by a pilot.

Reconstructing the file
Today · done by hand
History gathered
Laying out the material
Today · done by hand
Material pre-filled
Re-reading the previous objectives
Today · done by hand
Objectives recalled
Indicative figures, not contractual, to be confirmed by a pilot on your number of appraisals. Assessing the employee, judging whether objectives were met and any conclusion about their career belong to the manager. The agent produces no judgement about a 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 package, one agent

An appraisal preparation assistant (frameworks, history, questions), installed and operated for you.

Agility

Setup + controlled subscription

4,220 € excl. VAT setup
then 526 € 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
Order →
The simplest Serenity

All inclusive, no setup fee

761 € 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
Order →
100% Sovereign

On site, you own it

7,905 € excl. VAT setup
then 741 € excl. VAT/month · + hardware from 1,058 € (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)
Order →
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 appraisals

Your employees' data stays with youLocal inference or an isolated resource hosted in France; no record and no HR data entrusted to a third party, no element used to train a model.
Data in France, under French lawData relating to your employees: minimisation and location in France, architecture designed to reduce exposure to extraterritorial legislation, location alone not being enough to guarantee immunity.
The manager keeps the decisionThe agent produces prepared appraisal materials, which can be checked and altered; no approval is automated.
Human oversight & traceabilityOn your number of appraisals: systematic logging and tracking, in line with the AI Act.
Frequently asked questions

Your questions, our answers

Does the agent assess employees?
No, never. It gathers the framework, the history and the factual elements recorded; the assessment areas stay empty. Evaluating a person may affect their career: it rests with the manager, and the European AI Act requires that human control.
Does the agent say whether objectives were met?
It recalls the objectives as they were worded and the elements you have recorded. Judging whether they were met belongs to the manager — and it is that judgement which drives what follows from the appraisal.
Who has access to the preparations?
Only the managers concerned, under the rights you define. Every preparation is logged, and access can be revoked.
Is the HR data protected?
Yes. Hosting 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. The retention periods you have defined apply.
Does it adapt to our existing frameworks?
Yes. Your sections, your order and your questions are used as they stand. The agent does not propose its own framework: yours is the product of your social dialogue, and it does not rewrite it.
How long does it take to deploy this assistant?
A few weeks as a rule, depending on your frameworks and the integration with your HR tool, after a free audit then a phase of design, integration and testing.
Let's talk

Let's size up the potential in your appraisals

15 minutes to frame your frameworks and your safeguards — hosted in France, supervised, with no commitment.