Skip to main content

​
Overview

A playbook is a reusable guide your reps follow during a conversation, a talk track broken into sections, a set of objection handlers, and upsell prompts. Playbooks are created and edited in the workspace settings under Playbooks, then assigned to conversation types so the right guide surfaces automatically. Each playbook has a category, a type, an enabled/disabled flag, and a version number.

​
How it works

Every playbook stores three kinds of content:
  • Script: an ordered list of sections (each a title plus formatted body) that make up the talk track.
  • Objections: pairs of “Customer Says” triggers and a suggested response, with optional tips. These are meant to surface when the customer raises a concern during a live call.
  • Upsells: opportunities with talking points, meant to surface when relevant keywords are detected.
Playbooks are tagged with a category (Script, Objection Handling, Upsell, General) and a type (Sales call, Demo, Success, Implementation, Support, General). The type decides how the playbook is scored: sales playbooks by win rate, Success playbooks by renewals against churn, and Implementation playbooks by calls that kept go-live on track. Script bodies support variables inserted with the {{ menu, system variables (contact, user, org) are always available, and you can define custom variables on the Variables tab. The detail editor is organized into tabs: General, Script, Objections, Upsells, Settings (auto-scroll, live mode, AI suggestions toggles), Variables, Conversations (executions that used this playbook), and Suggestions (the playbook’s score and the changes drafted from its calls). You can generate content with AI for the whole playbook or for a single script section.

​
Rules: which playbook starts on which call

Settings → Playbooks → Rules decides which playbook starts on its own. A rule says “on phone calls or meetings, when these conditions hold, start this playbook”: the call’s channel, the journey stage its records are in, or a field on those records (for example a company’s lifecycle stage). The first matching rule wins, and the playbook starts live as soon as the call’s transcript begins. Reps can still switch when the mid-call card suggests another playbook. Above the rules, Suggested rules come from your last 90 days of calls: where a playbook did clearly better than running none on the same kind of call, with the calls behind it. Add rule puts it first; Dismiss hides it for good.

​
Generate a playbook

Generate with AI opens a dialog that starts from Sales, Success or Implementation, each with three templates (Discovery, Demo, Negotiation; QBR, Renewal, Save call; Kickoff, Training, Go-live check) or a blank prompt. Under Learn from, choose what the draft reads from your workspace: won and lost calls, the call library, scorecard reviews, docs and pages, your help center, and deals and other records that match your prompt. The preview shows what it read and, under each section, the calls and pages it learned from. Build playbook puts the draft in the editor; nothing is saved until you save.

​
Keep it improving

Each time a call on the playbook gets an outcome or a review, Custral re-scores it. When at least five calls since your last change have an outcome or a review, and the score has dropped ten points (or a new playbook is doing worse than calls with no playbook), the playbook shows Ready to update, the list shows Update ready, and workspace admins get one notification. Draft changes writes up to four changes from the playbook’s calls. Each one says why (“came up on 6 of 8 lost calls”), shows the proposed change, and lists the calls, reviews and step stats it learned from. Apply edits the playbook; Edit first lets you change the wording before applying. Nothing changes on its own. Implementation playbooks are scored from the onboarding behind each call: the onboarding linked to the call, or to the call’s contact or their company. It counts as on time when its phase moved to a live or done option by its go-live date, and as slipped when it went live late, is still not live past the date, or is blocked. Custral uses the time the phase was last changed as the go-live time, so editing the phase again later moves it. Snooze a week hides the nudge.
Outcomes are your workspace’s dispositions. They count toward the score when their name reads as a win or a loss for the playbook’s type (for example “Won”, “Renewed”, “Churned”, “Went live”, “Delayed”). Outcomes like “No answer” count for neither.

Groups & assignments

Bundle playbooks into groups and route them to conversation types.

Analytics

Track executions, completion rate, adherence, and step performance.

​
Using it

1

Create a playbook

On the Playbooks screen, click Add Playbook, give it a name, optional description, category, and type, then create it. You land on the playbook’s detail editor.
2

Add your content

Use the Script tab to add sections, the Objections tab for objection handlers, and the Upsells tab for opportunities. Save each tab with its Save button. Optionally click Generate with AI to draft content from a prompt.
3

Enable and assign it

On the General tab, toggle Enabled. Then route it to the right conversations from the Groups & Assignments screen.
Generating content with AI on the Script, Objections, or Upsells tab replaces existing content in those areas: review the prompt before generating.

​
Troubleshooting

When reporting an issue, include the request id shown in the error toast. It links the in-app error to the server logs.