AE Performance Review: Metrics and Development Plans

An AE quarterly performance review is a documented look backward that produces one focused plan forward. The review establishes what happened and why the team believes it happened; the one-page 30/60/90 coaching plan names the observable behavior, practice, evidence, support, and checkpoints that come next.

OPM’s primary performance-management cycle frames performance as planning, monitoring, developing, rating, and rewarding rather than rating alone. It also describes standards as measurable, understandable, verifiable, equitable, and achievable and emphasizes ongoing feedback. This article adapts those process principles to an AE quarterly review; it does not import federal employment rules into a private employer.

Separate the quarterly review from other meetings

Do not convert the review into a deal-by-deal forecast meeting. Do not call a disciplinary process “coaching,” and do not promise that three reviews automatically lead to or prevent a formal action. Separate channels protect the development conversation and the people involved.

Set the quarterly schedule, rubric, evidence sources, response route, and decision uses before the period. If reviews inform compensation, promotion, territory, account access, or employment action, qualified owners must approve the process. For weekly developmental cadence, use the sales 1:1 template.

Freeze the evidence packet before scoring

At the cutoff, create a read-only packet available to the rep and manager before independent drafts. Record extraction time, source query/report version, filters, timezone, currency, quota/territory version, exclusions, missing data, corrections, and packet owner. Preserve the original; append corrections rather than silently replacing it.

• Role and opportunity: role profile, segment, tenure/ramp status, territory/account allocation, quota and changes, leave or approved adjustments handled through qualified channels.
• Outcomes: approved attainment basis, won/lost/open population, pipeline creation, forecast variance, and customer outcomes relevant to the role.
• Motion: stage progression, dated next steps, buying-group coverage, qualification evidence, forecast changes, handoffs, and CRM quality.
• Behavior artifacts: representative calls, emails, plans, notes, mutual actions, proposals, follow-ups, and manager observations chosen under a stated sampling method.
• Growth and support: prior coaching bet, practice artifacts, checkpoints, enablement completed, manager commitments, blockers raised, resources provided, and unresolved disputes.

Sample across the whole period, deal types, outcomes, and buyer contexts. Do not cherry-pick the last call, largest loss, favorite win, or only calls the software happened to record. Note missing recordings and data quality as limitations rather than treating absence as poor behavior. Apply the controls in CRM hygiene before relying on CRM fields.

Define every metric and rating anchor

Every metric needs a contract: name, purpose, owner, formula, grain, population, inclusions/exclusions, source fields, period/cutoff, currency/timezone, quota/crediting version, treatment of splits, reversals and reopened deals, missing-data rule, correction path, and effective date.

For attainment, state numerator and denominator, credit timing, split rules, currency, services/renewals, reversals, ramp treatment, and approved adjustments. For win rate, state count or value, eligible stages, won/lost/open treatment, and period assignment. For forecast variance, state forecast snapshot, category, horizon, actual, and absolute or signed formula. The AE forecast accuracy guide covers that metric; AE quota planning covers quota context.

OPM’s primary performance-standards guidance describes quality, quantity, timeliness, and cost-effectiveness as general measures. Not everything requires a numeric proxy. If discovery quality is qualitative, define observable anchors and the sampling method instead of inventing a “talk ratio” target.

Write rating anchors before seeing the rep’s results. Use behaviorally specific language: “In the sampled opportunities, the rep documented a buyer-confirmed decision process and dated owner/next step under the rubric in X of Y eligible cases,” with the approved expected range. Avoid “top performer,” personality adjectives, or comparisons to whoever the manager remembers.

Separate outcomes, motion, behavior, and growth

Quota attainment is important, but a rep cannot practice “attain more.” Diagnose from outcome to motion to observable behavior. Conversely, activity volume is not automatically good behavior; assess relevance, quality, permission, and outcome. Use sales coaching metrics to keep a developmental measure close to the coached action.

Do not infer motivation, attitude, confidence, culture fit, or commitment from a dashboard. Describe observable work and impact. Separate facts (“three sampled opportunities lacked a dated next step at cutoff”) from inference (“the rep does not care”). Invite counterevidence and correction.

Calibrate with bias and context controls

EEOC’s primary performance-evaluation guidance recommends communicating standards, applying them consistently, explaining whether they were met, including relevant factual detail, and notifying employees when standards change. Its disability guidance highlights explicit expectations, clear standards, accurate measures, reliable feedback, and consistent application. These are useful controls, not a guarantee that a review is lawful or unbiased.

Before the meeting, the manager drafts independently. A calibration reviewer checks: same communicated rubric/version; same evidence window and sampling; fact versus inference; opportunity/context; leaves or accommodations handled only through appropriate channels; standard changes; unsupported personality language; counterevidence; similar application across appropriately comparable cases; and rating-to-evidence traceability.

Calibration is not forced ranking or making ratings fit a distribution. It is a consistency and evidence check. Keep original rating, proposed change, reason, evidence, reviewers, final rating, and date. Do not expose unrelated employees’ confidential information to justify a score.

Provide a route for the rep to dispute a fact, interpretation, or application of the standard. EEOC’s primary complaint guidance recommends identifying the concern, reviewing relevant material and comparisons where appropriate, correcting discrimination if found, explaining outcomes, and considering documentation. Follow employer policy and qualified HR/legal guidance for actual complaints.

Run the quarterly review meeting

1. Open with scope. Confirm period, role expectations, metric/rubric versions, review uses, confidentiality, response route, and packet limitations.
2. Rep reflection first. Ask what outcomes, motion, behaviors, growth, context, and manager support the rep sees. Do not require agreement as proof of coachability.
3. Review evidence by layer. State facts, then interpretations, then invite correction/counterevidence. Record unresolved items.
4. Confirm the summary. Name strengths, required standards, development areas, context, and evidence. Avoid introducing surprise allegations without a proper review path.
5. Select the coaching bet. Choose an observable behavior that plausibly addresses the constraint and can be practiced safely.
6. Write and respond. Complete the plan, manager support, checkpoints, rep comments, disagreements, escalation/response route, and signatures or acknowledgments.

A signature should mean receipt or agreed meaning under employer policy; do not assume it proves agreement. Label it clearly. Give the rep a copy. Record who may access it, where it is retained, correction procedure, and review date.

Choose one coaching bet

A coaching bet has five parts: one observable behavior, the job context where it applies, a practice method, an artifact, and a measure close to the behavior. Example: “Before every eligible first demo, state the buyer-confirmed problem and success criterion, ask the buyer to correct it, and capture the confirmed version in the CRM note.”

The bet is not “increase win rate 10%,” “be more strategic,” or “improve discovery.” Those are outcomes or abstractions. It is also not an excuse to omit other communicated standards. The one-bet method focuses developmental practice while normal role expectations continue.

Choose with the rep where possible. Check that the behavior is job-relevant, within the rep’s influence, observable, ethically and legally appropriate, supported by systems and manager time, and unlikely to create a harmful gaming incentive. Define what evidence would disconfirm the diagnosis.

Write the one-page 30/60/90 plan

At each checkpoint, record planned artifacts, completed artifacts, evidence reviewed, rep learning, manager feedback, support delivered, blockers, corrections, and next action. Use the broader sales coaching plan for program design, but keep this quarterly artifact to one page plus evidence references.

Worked example: improve discovery evidence

Frozen diagnosis: The rep’s outcome was below the communicated quarterly standard. Territory and quota context were recorded. In an approved sample of eight eligible first-discovery-to-demo opportunities, three notes contained a buyer-confirmed problem, impact, and success criterion before the demo. Five moved forward with a seller-authored summary that was not explicitly confirmed. The evidence supports a discovery-to-demo handoff bet; it does not prove this behavior alone caused the outcome.

Bet: Before each eligible first demo, the AE states the problem, impact, and success criterion; asks the buyer to correct or confirm them; and stores the confirmed wording in the CRM next-step note. Near measure: number of eligible sampled demos with all four artifacts ÷ eligible sampled demos, using the frozen rubric and sampling rule.

• Day 30: Review three contrasting call examples; rehearse five scenarios; manager scores two rehearsals with the sales call scorecard; revise phrasing. Evidence: rubric, recordings, feedback notes.
• Day 60: Apply to at least the number of naturally occurring eligible calls available; review a pre-agreed representative sample, including a difficult case; audit CRM artifacts. Do not create unnecessary calls to hit a count.
• Day 90: Re-sample under the same definition; compare behavior evidence with baseline; review qualitative buyer corrections and opportunity motion; decide close, continue, or choose a new diagnosis.

Manager commitment: one rehearsal in week one, biweekly sample feedback within two working days, an approved example library, and removal/escalation of CRM workflow blockers. Rep response: “I agree the artifacts are missing; I disagree that this explains two procurement-led losses. Add those as counterexamples.” The final file preserves both.

Printable review and coaching template

Print the block or copy it into the approved system. Link evidence rather than pasting sensitive transcripts or customer data into the review. Follow access, retention, correction, and deletion policy.

Audit follow-through and close the cycle

At days 30, 60, and 90, audit two parties. Rep: planned practice, artifacts, application, reflection, corrections, and blockers. Manager: scheduled coaching, timely feedback, examples/resources, process fixes, escalation, and documentation. A missed manager commitment is not rep noncompliance.

Use a plan completion ledger: item, owner, due date, status, evidence, blocker, change reason, and next action. If source data or expectations changed, document version and decide whether comparison remains valid. Never move a target after seeing results without recording the change and its effect.

At day 90, close with evidence: behavior applied consistently under the contract; partially applied and continue with revised support; diagnosis not supported and choose a different bet; or formal concerns require a separate qualified process. Do not let the coaching plan drift indefinitely or quietly become discipline.

Where Gangly fits—and does not

Gangly’s own product copy positions it around call preparation, live guidance, reviewed notes, CRM hygiene, and follow-through. Those artifacts may contribute to a frozen evidence packet and coaching practice when approved. That is first-party positioning, not independent proof that Gangly improves performance, removes bias, or should rate employees.

Do not let software make final employment ratings, infer protected or sensitive traits, diagnose motivation, or treat a generated summary as objective. Validate source completeness, permissions, recording rules, transcript accuracy, model behavior, correction, retention, access, and deletion. Human managers remain accountable, with qualified HR/legal involvement as required.

A good AE review is not the most confident rating. It is the clearest chain from communicated standard to frozen evidence, calibrated interpretation, focused practice, reciprocal commitments, and documented follow-through.