AEs do not lose deals randomly. They lose them in predictable places: discovery that never went deep enough, a stakeholder nobody mapped, a discount conceded too early, a champion who went quiet, a deal that drifted because no one drove the next step. Each of those is a rehearsable moment. These are the five scenarios that fix them.
Unlike SDR practice, which lives in the first 30 seconds of a call, AE practice spans the whole deal. Run these as a rotation, and weight the ones that match where your deals actually stall — your pipeline will tell you which two you need most.
Scenario 1: Deep discovery that surfaces real pain
Most discovery is too shallow. The AE confirms a surface symptom, hears a feature request, and moves to demo. The deal then dies later because the real pain, the real cost, and the real urgency were never established. Deep discovery is the highest-leverage skill an AE can drill.
The setup
AI buyer gives a vague, surface-level problem statement ("our current tool is just kind of clunky"). They will not volunteer the underlying pain unless the AE digs. The rep has to layer questions until the business impact and the cost of inaction are explicit.
What good looks like
- Moving from symptom to impact to quantified cost ("what does that clunkiness actually cost you per quarter?")
- Asking about the decision process, not just the problem
- Resisting the urge to pitch the moment a pain appears
- Success criteria: a documented pain with a number attached and a reason to act now
Scenario 2: The multi-stakeholder buying committee
Enterprise deals are won and lost across a committee, not in a single conversation. The technical evaluator, the economic buyer, and the end-user each want something different — and they conflict. AEs who only sell to one person get blindsided when the deal hits a room they were not in.
The setup
AI buyer shifts personas mid-call: first a detail-hungry technical evaluator, then a skeptical CFO-type who only cares about ROI, then a nervous end-user worried about change. The AE has to address each priority without losing the others.
What good looks like
- Tailoring the message to each stakeholder without contradicting yourself
- Mapping the committee out loud ("who else needs to be comfortable before this moves forward?")
- Equipping the champion to sell internally when you are not in the room
- Success criteria: a clear stakeholder map and a plan to reach the economic buyer
Scenario 3: The late-stage discount squeeze
Procurement asks for a discount at the eleventh hour, often after the decision is effectively made. AEs who cave instantly burn margin and train the buyer to push harder. AEs who hold the line — and trade concessions for value — protect both the deal and the price.
The setup
AI buyer is ready to sign but demands 20% off "or we walk." The pressure is real but partly a test. The AE has to hold value, avoid an immediate concession, and trade rather than give.
What good looks like
- Not discounting reflexively — pausing and asking what is driving the request
- Trading any concession for something in return (term length, case study, faster close)
- Re-anchoring on the value and cost-of-problem established in discovery
- Success criteria: the deal closes near full price, or a concession is traded for real value
Scenario 4: Re-engaging the silent champion
A deal goes quiet. The champion who was excited two weeks ago stops replying. Most AEs send three "just checking in" emails and let the deal rot. The skill is re-engaging without sounding desperate and uncovering what actually changed internally.
The setup
AI buyer is a previously warm champion who has gone cold — distracted, deprioritized, maybe facing internal resistance they did not mention. The AE gets them back on a call and has to diagnose the real blocker.
What good looks like
- Re-engaging with new value or insight, not another "checking in"
- Surfacing the real reason for the silence (budget freeze, competing priority, lost internal support)
- Reviving urgency by reconnecting to the original pain
- Success criteria: the real blocker is named and a path forward is agreed
Scenario 5: Driving the close and the mutual action plan
Deals do not close themselves, and "let me know when you are ready" is not a close. The strongest AEs build a mutual action plan and confidently ask for the business. This scenario drills the moment most reps soften when they should be direct.
The setup
AI buyer is bought-in but passive, waiting to be led: "This looks good, what are the next steps?" The AE has to take control, define the path to signature, and ask for the commitment.
What good looks like
- Proposing a clear mutual action plan with dates and owners
- Asking directly for the business without flinching
- Pre-handling the steps that usually cause slippage (legal, security review, procurement)
- Success criteria: a signed mutual plan with a target close date, not an open-ended "we will see"
How AEs should use these scenarios
Unlike SDR practice, AE practice should follow your pipeline. Look at where your deals actually stall — if you lose late on price, weight Scenario 3; if deals die in committee, weight Scenario 2. Run the two scenarios that match your weakest stage two to three times a week, and rotate the rest in monthly. The point is targeted reps on the exact moment your forecast keeps breaking.
AI roleplay is uniquely suited to AE practice because it can hold a long, branching conversation and switch stakeholder personas on the fly — something a single manager roleplay cannot easily simulate. An AE can rehearse a full multi-threaded deal cycle in 25 minutes, get scored on discovery depth and objection handling, and run it again with a different committee dynamic tomorrow.
“The fastest way to lift win rate is not more pipeline. It is fewer deals lost in the same predictable place. Find the stage where your forecast keeps breaking, and rehearse that exact moment until it stops surprising you.”
