Module 3 · Card 1 of 8

Agent tools for sales managers

This is not a tool roundup. The question this module answers: which sales-manager workflows are ready for an agent to drive, and what does it cost you in trust when it gets one wrong? Reversibility is the deciding factor.

Module 3 · Card 2 of 8

Three tiers of autonomy

Drafting agents (Cursor for SQL, Notion AI for write-ups) — you read every output. Triage agents (Clari Copilot, Salesforce Einstein for pipeline scoring) — they propose, you approve. Execution agents (Outreach.io AI, AE.studio's CRM bots) — they act, with guardrails. Each tier has a different cost of being wrong.

Module 3 · Card 3 of 8

What's actually shipping (Q2 2026)

Salesforce Einstein: AI-summarised pipeline reviews referenced by 41% of enterprise sales managers (Salesforce State of Sales 2026). Clari Copilot: forecast-challenge auto-generation in production at 28% of mid-market sales orgs. Microsoft Dynamics + Copilot: CRM hygiene automation in 19% of customer base — most of those running it in approve-each-change mode.

Module 3 · Card 4 of 8

CRM management — start here

Best place to delegate first. Stage-hygiene checks (does every Stage 3 deal have a next step?), missing-field flagging, duplicate contact dedup, opportunity-rep ownership refresh after re-orgs. Reversible, bounded, the agent's wrong-answer cost is minutes of admin to undo.

Module 3 · Card 5 of 8

Proposal creation — the middle ground

Drafting tier. The agent assembles a first-draft proposal from the deal context + your template library. Time saved: 60-80 minutes per proposal in published benchmarks. But the agent doesn't know which clause your legal team will balk at, which pricing structure this buyer's CFO has rejected before, or which competitor's quote is already on the table. Your edits are where the deal is won.

Module 3 · Card 6 of 8

Revenue reporting — the autopilot trap

It is tempting to fully delegate weekly board-level revenue reports to an agent. Don't. The agent will produce a smooth narrative that hides the deal you really need leadership to know about. Use AI to assemble the deck; you write the narrative section that names the deal that's in trouble and what you're doing about it.

Module 3 · Card 7 of 8

Calibrating agent trust over time

Track the disagreement rate — how often you override the agent's suggestion. Below 5% means the agent is on rails and you're not adding value to the review; either widen its scope or stop reviewing. Above 30% means it doesn't have the context yet; pull it back to a lower tier until accuracy improves. The 5-30% band is where leverage lives.

Module 3 · Card 8 of 8

End of module

Next: five questions. Two minutes. Then look at last week's calendar and ask: which 90-minute task could have been a 5-minute review of agent output?

Agent tools for sales managers: CRM, proposals, and the cost of trust