Implementation · Copilot Studio

An agent your teams can keep evolving.

Copilot Studio lets you build agents and workflows connected to your sources, inside your Microsoft environment — and beyond it, through APIs and MCP connectors to your other tools. They are built with the people who will use them: by the end, the capability sits with you.

Formats

Four ways in, depending on your maturity.

Many organisations start with scoping alone. It is often the best investment: it stops a badly framed project before it starts.

Scoping · 2 hrs

Check the case holds

A session to assess your use case: how often the need actually arises, the quality of the sources, the expected gain. With no commitment to go further.

The business owner
and one technical profile
Workshop · 1 day

Build a first agent

A simple agent, built end to end in one day, on an already scoped case with sources available.

4 to 8 people
from the team involved
Programme · 2 to 4 weeks

Go live

The full format: scoping, source preparation, build, testing and handover. Short sessions spread out.

A business team
and a named champion
Ongoing support · 3 to 6 months

Roll out several agents

After the first agent, your teams build the next ones themselves, with support on demand.

Internal champions
already trained
Programme

From idea to live agent, with your teams at every step.

The outline below is the full programme. The one-day workshop covers steps 4 and 6 on an already prepared case.

01

Choosing the right approach before choosing the tool

Microsoft now offers several ways to build an agent, from a few clicks to a line of code. Picking the wrong approach costs more than picking the wrong use case.

  • SharePoint agent — an assistant sitting on a document library, created by a business user in minutes. Often enough.
  • Copilot Studio — when you need defined topics, precise triggers, or actions into other systems
  • Agent or workflow — an agent answers and decides in conversation; a workflow runs an automated sequence with no one to talk to. The same studio produces both, and confusing them is the most common mistake.
  • Developed approaches — when the need goes beyond no-code. The right technical profile then joins the engagement, and the handover rule stays the same.
  • Governance — who creates, who approves, who monitors: a question to settle before the first agent, not after the tenth
02

Choosing the right use case

Not every agent idea is worth it, and some don't warrant an agent at all. This step usually saves more than it costs.

  • Frequency and volume — a monthly need doesn't warrant an agent; a daily one does
  • Nature of the request — recurring questions on a stable corpus, or a task that differs every time
  • What belongs elsewhere — a well-organised FAQ, a document template or a simple automation is sometimes enough
  • Choosing the first case: the one with the cleanest sources, not the one with the loudest need
03

Preparing the knowledge sources

An agent answers only as well as what it is given to read. This is the most decisive step, and the most often neglected.

  • Source inventory — SharePoint sites, document libraries, pages, reference files
  • Corpus quality — outdated versions, duplicates, contradictory documents: what must be settled before connecting anything
  • Access rights — the agent inherits permissions; who should see what
  • Formats that work well, and those that don't: scans, complex tables, slide decks
04

Building the agent in the workshop

The agent is built in the session, with your teams at the controls. They build, guided step by step — which is what makes taking it over possible afterwards.

  • Instructions and scope — what the agent must do, what it must refuse, and in what tone it answers
  • Topics and triggers — the intents it recognises and the phrasings that activate them
  • Variables and entities — hold information through the conversation, extract a reference or a date
  • Actions — connect the agent to Power Automate to raise a ticket, send a notification, write to a list
  • Beyond Microsoft — API and MCP connectors into your business tools (CRM, ERP, document base, third-party service), so the agent reads and writes where your data actually lives
  • Model choice — several LLMs are available: you keep the one that handles the task at the best cost, and swap it if another does better, without rebuilding the agent
  • Escalation — define when the agent must hand over to a human instead of improvising
05

Testing and hardening

An agent that works in a demo often fails on contact with reality. This step prevents the backlash.

  • Real scenarios brought by users, not by the builder
  • Edge cases — out-of-scope questions, ambiguous phrasing, requests in several languages
  • Measuring what fails — misunderstanding rate, off-topic answers, topics never triggered
  • Fixing topics, enriching phrasings, adjusting sources
06

Publishing and handing over

The part that separates an implementation from a delivery. Without it, the agent rarely outlives six months.

  • Publishing — Teams, SharePoint, or a dedicated channel depending on usage
  • Governance — who may modify the agent, and how a change is approved
  • Operating documentation — written during the workshops, by your teams
  • Reading the metrics — sessions, resolution rate, most-triggered topics, unanswered questions
  • Full handover: by the end, your teams add a topic or a source with no outside help
Discuss this format — 30 min, free →
Use cases

What it looks like, function by function.

A first selection of typical cases. Scoping will tell whether yours calls for an agent or something simpler.

Internal support · HR

Assistant on HR procedures

Answer routine questions on leave, expenses and internal procedures, drawing on the existing SharePoint documents.

Outcome · Repetitive questions absorbed, the HR team on the real cases
Real estate · Property management

Extraction across a contract portfolio

Query a body of leases: deadlines, specific clauses, obligations per asset, with a link back to the source document.

Outcome · Contractual information available without reopening every lease
Manufacturing · Technical support

Assistant on product documentation

Answer field questions from manuals, datasheets and procedures, escalating to an expert when needed.

Outcome · A first line of answers available around the clock
Marketing · Content

SEO and GEO article writer

Produce articles aligned with your keywords, your tone and your existing pages — optimised both for classic search and for generative answers (GEO).

Outcome · A publishing rhythm you can sustain, without outsourcing the brand voice
Legal · Contracts

Contract drafter

Generate a draft contract from your approved templates and the parameters of the deal, with the mandatory clauses and the points left open for decision.

Outcome · A compliant first draft in minutes, legal review focused on what matters
Sales · Pre-sales

Assistant on the offer and references

Quickly find a product reference, a pricing condition or a comparable client case in the sales documentation.

Outcome · Answers given in the meeting rather than after checking
“In two seminars, Jonathan showed us the concrete potential of AI for our business. A clear, well-taught approach built on practical cases: directly applicable day to day.”
Arthur Michelino · Manager, Olea
Discuss this format — 30 min, free →
Get in touch

Got a use case in mind? Let's check together that it holds.

The direct route

Book 30 minutes

A scoping call: your context, your teams, the format that fits best.

Book 30 minFree · No commitment
In person or remoteEnglish · French · SpanishFrance & internationalIn-house
In writing

Describe your need

Three lines is enough: reply within 24 h.