GuidesPublished 9 min read

CSM Onboarding Plan: Gate on Scored Calls

three ascending steps toward the light
Listen to this article · 13:13 · AI-generated narration
0:00 / 13:13
Chapters

TL;DR

Build your CSM onboarding plan around four gates instead of three dates. The gates are passing scored reps on the kickoff call, the low-adoption check-in, delivering bad news, and the QBR. Each pass transfers one named responsibility (solo kickoffs, an owned book of accounts, escalation comms, renewal ownership), and each failed attempt is re-run, not waved through.

  • Gate on a passing scored rep on a named conversation, never on days served.
  • Every gate blocks one named responsibility until the CSM passes.
  • Write rubric rows a manager can verify from the transcript alone.
  • Read scores per rep per scenario. Never blend them into one readiness number.
  • Run gate one, the kickoff call, with a new hire this week.

Why a Dated 30-60-90 Plan Is a Countdown

Pull up your current customer success onboarding plan. Look at what moves a new CSM from the first milestone to the second. If the answer is the calendar, the plan measures attendance. A 30-60-90 plan with dates but no gates is a countdown, not a ramp.

The case for dates deserves a fair hearing. Time-in-role is simple to plan around. A new hire knows where they stand, and capacity planning gets a clean handover date. If your CSMs ramp at the same pace, gates cost you nothing: everyone passes on schedule and the dates hold. If they do not ramp at the same pace, a gate shows you in practice instead of on a live account.

A date also certifies the wrong thing. Completing product modules and passing a knowledge check proves the CSM can recall the material. It does not prove they can hold a kickoff together when the customer's admin asks to skip straight to configuration. We count training only when it certifies observable behavior under pressure, and that behavior has to be scored from a transcript.

Keep the dates as planning targets. Stop using them as the trigger. Responsibility transfers when the transcript says it should.

What Should a CSM Onboarding Plan Gate On?

Gate on a passing scored rep on a named service conversation. Attach each pass to one responsibility the CSM does not hold until they pass. We recommend four CSM ramp gates, run in order: kickoff call, low-adoption check-in, delivering bad news, and QBR.

At each checkpoint the manager records one decision: advance, repeat, or escalate. Advance means the scored rep passed and the responsibility transfers that week. Repeat means a scheduled re-run on the same scenario. Escalate means a failed re-run, which opens a written remediation plan with the CS director.

A gate that blocks nothing named is a suggestion, not a gate. The last column of the table below is what makes each gate real: it states who holds the work until the CSM passes.

GateNamed scenarioA pass transfersUntil then
1Kickoff callSolo kickoff callsManager or senior CSM leads every kickoff
2Low-adoption check-inAn owned book of accountsAccounts stay with the current owner
3Delivering bad newsEscalation comms without a manager on the lineManager delivers or joins every bad-news call
4QBRRenewal ownershipManager owns the renewal conversation

Gate One: The Kickoff Call

We treat the kickoff as the call that sets terms for the relationship. We score it against the Sandler up-front contract because the behavior is checkable. Before the first agenda item, the CSM states purpose, time, both parties' agendas, and the acceptable outcomes, including that we're not ready to proceed yet is acceptable. The customer then agrees out loud.

Rubric rows: (1) Up-front contract set and agreed before the first agenda item. (2) Handoff confirmed: the CSM restates why the customer bought, in the customer's words, and the customer confirms or corrects it. (3) Success criterion named: the customer states one outcome they will judge the first phase by. (4) Next step closed: a dated session with named attendees is booked before the call ends.

A failing attempt opens with Great to meet everyone, I have our standard onboarding deck, so let's jump in. It closes with I'll send a recap and we'll find a time. Rows one and four are absent, and the transcript proves it. A kickoff without an agreed outcome is a product tour with a calendar invite. For scenario variants, see our customer kickoff call roleplay guide.

Gate Two: The Low-Adoption Check-In

This gate certifies that the CSM can surface a problem without leading with blame or with a fix. We score it with SPIN implication questions. The CSM builds the cost of low usage before naming any remedy.

Rubric rows: (1) Opens with the usage fact and the customer's original goal, with no blame. (2) Asks at least one implication question before proposing anything, such as What happens to the reporting goal if the team keeps working outside the tool? (3) The customer names the cost in their own terms. (4) One adoption action is agreed, with a named owner and a check date.

SPIN is a discovery sequence. It does not score the close, so row four carries the commitment. A failing attempt sounds like Your logins dropped, so I'd suggest we book a retraining session for the team. The fix arrives before the customer has said why the gap matters to them. A fix proposed before the customer names the cost is the CSM's plan, not the customer's. Our low product adoption conversation drill covers persona variants.

Gate Three: Delivering Bad News

No named model is needed here. The behaviors are plain and every one shows up in the transcript.

Rubric rows: (1) The news is stated within the first two sentences, before any cushioning. (2) Ownership: the CSM says what happened and what our side is doing, with no blame on another team. (3) Known, unknown, next update: confirmed facts are separated from open questions, and the CSM commits to a time for the next update. (4) The customer gets room to respond, and the CSM paraphrases the customer's concern before answering it.

Use talk/listen ratio as a diagnostic on gate three, not a pass bar. If the CSM's share climbs right after the news lands, open the transcript at that moment and read what they said. The ratio tells you where to look, and the rubric rows decide the verdict.

A failing attempt sounds like So there's been a lot going on with the engineering roadmap, a few things have shifted, and we're still looking at options. If the customer has to ask what the news is, the attempt fails.

Gate Four: The QBR

The QBR gate certifies that the CSM can build a renewal case from the customer's own words. We score it with Gap Selling, which quantifies the distance between the customer's current state and future state.

Rubric rows: (1) The customer states their current state against the goal set at kickoff, in their own numbers. (2) The customer states a target for the next period. (3) The gap is agreed out loud: So the distance between where you are and where you said you need to be is this. Is that right? (4) The renewal path is surfaced. The CSM asks who else weighs in on the renewal and when, and books the next step.

When the customer has already hit the kickoff goal, the gap does not disappear. It moves to the next-period target from row two, and row three scores agreement on that new gap. If the customer will not name a next target, there is no agreed gap, and the attempt fails row three. A failing attempt sounds like Here's your usage for the quarter, and here are the features we shipped. The customer never speaks to the gap. Renewal ownership belongs to the CSM who can get a customer to agree on the gap out loud. More scenarios are in our quarterly business review roleplay drills.

How Do You Score Gates Without One Blended Readiness Number?

Read every score per rep per scenario. Never average the four gates into one figure. A strong kickoff score says nothing about how the same CSM delivers bad news, and a blended number hides exactly the gap you need to see.

Write the rubric rows against your own CS call stages and standards. Feedback graded on a generic rubric is feedback a CSM can rationally discount. Scoring against the organization's own methodology is what makes the debrief land.

Scored evidence also corrects the manager. Intuition alone overrates confident reps and underrates quiet ones, and a transcript does neither. A CS readiness certification is a transcript a second manager could re-score and reach the same verdict.

If you run gates in XL Roleplay, your call stages, rubrics, and standards are loaded first, and every session is scored against them. Each session is recorded, timed, and transcribed. Flags in the report link back to the exact moment in the transcript, which gives you the evidence behind an advance or repeat decision.

What Does the Gate Program Cost a Manager's Week?

We recommend 30 minutes to run each weekly session: 5 for setup, 15 for the drill, and 10 for the debrief. Add about 10 minutes of prep beforehand to pick the scenario from a live account and choose the one rubric row you will debrief.

On top of that, we recommend a 15-minute 1:1 built around one flagged moment from a scored session. It runs in addition to your account-review 1:1, never as a replacement for it. A coaching block survives only as a protected recurring calendar commitment, separate from pipeline and account reviews.

Keep each debrief to one moment and one behavior. Let the CSM self-diagnose before you give your verdict. End with a scheduled re-run. A debrief that covers three problems is entertainment, not coaching.

Run the Gate-One Drill With a New Hire This Week

Prep (about 10 minutes): pick a real account that kicked off recently. Write a one-paragraph persona for the customer admin who was not on the sales calls. The admin opens with Can we skip the slides and go straight to configuration? Choose row one, the up-front contract, as the debrief row.

Run (30 minutes). Setup, 5 minutes: give the CSM the scenario and all four gate-one rubric rows. Drill, 15 minutes: the CSM runs the kickoff while you play the admin, and you keep pushing to skip ahead. Debrief, 10 minutes.

Pass bar: all four rubric rows are evidenced in the transcript, clearly enough that a manager who did not hear the call could verify each one. A failing attempt sounds like Sure, we can skip ahead, let me pull up the admin console. That reply sets no contract, confirms no handoff, and names no success criterion.

Debrief script: Find the moment the admin asked to skip the slides. What did you agree to there, and what did the customer agree to? Let the CSM answer first, then name the one behavior: an agreed up-front contract before the first agenda item. Book the re-run before the debrief ends, or the gate quietly turns back into a date. When the CSM passes, record advance. They lead the next kickoff solo.

Frequently asked questions

Can an experienced CSM hire skip the gates?

No gate is skipped, but none has to wait. We recommend running all four as a baseline in the first week and transferring each responsibility as soon as its gate is passed.

How many failed attempts before a gate escalates?

We recommend a scheduled re-run after the first failure and escalation to a written remediation plan after a failed re-run. Use the same scenario each time so the per-scenario trend stays comparable.

Do the gates need AI roleplay, or can a manager play the customer?

Either works if the session is recorded, transcribed, and scored against the same rubric rows. The gate is the transcript evidence, not the tool.

Does customer success manager training still need product modules?

Yes. Modules prepare the CSM to attempt a gate, but they do not certify readiness. Keep them as prerequisites to attempting a gate, never as the gate itself.

All insights