Catalogue › Change and Communications › manager-enablement

Manager enablement

Equips line managers to lead AI adoption in their own teams - a briefing before their team hears anything, conversation guides for purpose and concerns, team norms for approved use and verification, protected practice time, a monthly question and clear routes for ideas and issues - without asking managers to monitor individuals. Use when managers are bypassed or unsure, before a wave reaches their teams, or when teams stall.

Department

Change and Communications

Lifecycle

Enable, Launch, Measure, Scale

Maturity

adopter, builder, scaler, ai-enterprise

Risk tiers

0, 1, 2

Platforms

copilot-chat, m365-copilot, m365-agents, copilot-cowork

Editions

starter, scale, enterprise

Reviewers

none

Status

draft

Copilot Cowork

Change and Communications plugin for Copilot Cowork, or this skill on its own: Download SKILL.md

Licence

CC BY 4.0, by Simon Foote

Purpose

People decide whether to invest effort in AI largely by watching their line manager: whether the manager explains why, makes time to practise, uses the tools themselves and treats mistakes as learning. This skill equips managers at every level to lead that change in their own teams with a briefing, conversation guides, team norms, a simple cadence and clear routes for what they cannot answer. Manager enablement usually goes wrong when managers learn about the change at the same time as their teams, when they are handed a slide deck instead of a conversation, or when they are asked to "drive usage" and start chasing individuals.

When to use

• A Copilot wave, agent or Cowork pilot is about to reach teams and their managers have not been briefed.
• Teams stall after onboarding and managers say they do not know what is expected of them.
• Managers are raising the same questions (time, jobs, monitoring, quality) and giving inconsistent answers.
• Some managers are pressing for usage figures on named people, or using AI use as a performance signal.
• A role-based adoption journey needs managers to play a defined part.

When not to use

• Helping executives use Copilot personally and lead visibly - use copilot-adoption:executive-enablement.
• Copilot scenarios and habits for leaders' own work - use m365-copilot:copilot-for-leaders.
• Designing the adoption journey for one role group - use copilot-adoption:role-based-adoption.
• Formal learning curricula and facilitator materials - use learning-and-skills:ai-role-curriculum.
• The always-on reinforcement calendar after launch - use copilot-adoption:adoption-reinforcement.

Questions to ask first

1. Which managers lead the teams in scope, at which levels, and when will each team be affected?
2. What do managers need to be able to say and do: explain why, set norms, protect practice time, answer concerns, route ideas?
3. What is the agreed position on jobs, use of usage data and approved use, so managers can give the same answer?
4. How much time can managers realistically give, and which existing meetings can carry the work?
5. Do managers have access to the AI experience themselves before their teams, and are they confident using it?

Required inputs

• The communications plan, core messages and agreed answers from change-and-communications:ai-communication-strategy.
• Group-level impact for the teams in scope from change-and-communications:people-change-impact.
• The launch waves from copilot-adoption:launch-plan and starting scenarios from copilot-adoption:scenario-discovery.
• The approved-use guidance, support routes and the route for reporting suspected data exposure.
• Agreement from the sponsor and HR on how AI use relates to performance management.

If managers will not have access before their teams, record it as a risk and move their access earlier where possible. If no position exists on performance management, escalate before managers are briefed, because it is the question they will be asked first.

Decision framework

Define the manager role in five actions, each with a tool. These are the minimum; add only what the impact requires.

Adjust depth by the team's impact rating.

Practitioner approach

Brief managers first, always. Set a rule that no team hears about a change affecting it before its manager has been briefed and has had a chance to ask questions. Where managers lead managers, cascade the briefing through the levels in order, with enough time between for questions to come back up.

Give managers a conversation, not a presentation. A short team conversation guide works better than slides: open with why the organisation is doing this and what it means for this team, show one starting scenario on real work, agree how the team will use AI and check its outputs, and ask what worries people. Managers should leave the conversation with a list of questions to answer or route, not a list of who attended.

Make the norms explicit. Each team agrees, in plain words, which tasks AI is used for, how outputs are checked before they leave the team, what must never be put into a prompt, when use is disclosed to colleagues or customers, and what stays a human judgement. The team norms canvas takes twenty minutes and prevents most of the confusion that otherwise arrives as support tickets.

Protect practice time. People learn AI by using it on real work, and that takes time during the first weeks. Agree with managers how much time their teams will have and when, and help them fit it around peak periods. Where managers cannot release time, record it and tell the adoption lead, because that team will need a later wave.

Equip managers to hear concerns well. Give them the agreed answers on jobs, monitoring, data and quality, and a clear route for what they cannot answer. Ask them to pass on themes, not names. Remind them that a team member's scepticism is information about the change, not a performance issue.

Draw the monitoring line clearly. Managers must not ask for, receive or use individual AI usage data, or treat volume of use as a measure of performance. What managers can ask is how the team is using AI, what is working and what is not. State this in the briefing and in writing, agreed with HR and the DPO.

Support managers as a group. Run short manager cohort sessions for high-impact areas where managers can compare notes, and give each manager one monthly question to ask their team. Encourage managers to use the AI experience themselves on their own work, so they speak from experience. Share what managers learn back into the programme through the change network and the adoption lead.

Outputs

• A manager toolkit: briefing pack, team conversation guide, team norms canvas, practice time agreement, answers to common concerns, escalation routes and the monthly question.
• A manager briefing schedule aligned to waves, recorded in the manager section of the ai-communications-plan template and the enablement section of the adoption-plan.
• A written statement, agreed with HR and the DPO, on what managers may and may not do with AI usage information.
• A monthly theme summary from managers to the adoption lead and change lead.

Microsoft platform considerations

Managers should have the same AI experiences their teams will receive, configured the same way, before the team launches. What aggregate usage reporting can show, and to whom, depends on current Microsoft capability and the organisation's settings; confirm it against references/sources.md and m365-copilot:copilot-usage-reporting, and do not promise managers views that the organisation has chosen not to provide.

Adoption considerations

Manager involvement is the strongest local lever on adoption. Coordinate with Champions so managers know who their local Champion is, and so practice time and peer help reinforce each other.

Knowledge considerations

Managers often know where team knowledge is out of date or held by one person. Give them a route to report it to content owners and to knowledge-management:knowledge-gap-analysis.

Security and privacy considerations

Managers must not request or use individual usage data. Team norms repeat the approved-use rules on confidential and personal data, and every manager knows the route for reporting suspected data exposure. Questions about monitoring and employee data end with HR and the DPO.

Responsible AI considerations

Managers set the expectation that people remain accountable for AI-assisted work and must check it. Where a team uses AI in work that affects decisions about people, managers should know that higher-risk uses go through intake and review, not team norms alone.

Evaluation requirements

Not applicable as solution evaluation. The toolkit is tested with a small group of managers before release, and the briefing is revised after each wave using manager feedback.

Value measures

Owned by the change lead, with baselines at the start of each wave: share of managers briefed before their teams, teams with agreed norms, practice time agreed, and team-level pulse results on clarity of expectations and confidence to raise concerns. Reported by team or business unit above the minimum group size, never as a rating of individual managers.

Scale notes

Starter

A single briefing for all managers, a two-page toolkit and a monthly question, owned by the adoption lead.

Scale

Briefings by wave, cohort sessions for high-impact areas, and a manager channel connected to the change network.

Enterprise

A cascade through management layers with local adaptation, translated toolkits, and manager enablement built into standard leadership development owned by HR.

Hand-offs

• To copilot-adoption:role-based-adoption for the manager's part in each role journey.
• To change-and-communications:change-network so managers and change agents share themes.
• To change-and-communications:behaviour-reinforcement-plan to embed manager routines that sustain new ways of working.
• To change-and-communications:resistance-management when a team's concerns need more than a manager conversation.

Reviewer gates

Not applicable as a formal gate, because manager enablement for approved AI experiences does not release a solution. The statement on usage data is approved by HR and the DPO before managers are briefed.

Failure modes

• Managers learn about the change from their teams and cannot answer basic questions.
• Managers are told to "drive usage", ask for individual figures and turn adoption into surveillance.
• No practice time is agreed, so teams try AI once and return to old habits.
• Toolkits are sent as attachments with no conversation, and few managers use them.

Sources

references/sources.md in this skill lists the source records behind this skill, what each is authoritative for, how volatile it is and what you may do with it. Product availability, licensing, limits and preview status change often, so check the current source before stating them, and cite what you used.

Never

• Ask managers to monitor, rank or chase individuals on AI usage.
• Let AI usage volume become a performance measure for anyone.
• Brief teams before their managers.
• Give managers answers on jobs or monitoring that HR and the DPO have not agreed.