CrewHu CSAT: surveys, urgent alerts, and churn flags

CrewHu is where your partners tell you the truth. A survey answered "Negative" on a Tuesday is the earliest, cheapest warning you will ever get — and in most MSPs it dies in a CrewHu dashboard nobody opens. Connect CrewHu to Current and every survey lands against the right company, turns into an urgent follow-up for the person who owns the account, and becomes a permanent signal inside company health, the Relationship Health tile, the AI company read, and your SBR packs.

One token does all of it. Current imports your full survey history first, then keeps up with new surveys every 15 minutes.

The connection and the alert settings are managed by a Tenant Admin or sales leadership. Anyone with CRM access can see the connection status; partners never see any of this.

1. 1
   Copy your CrewHu API token

   In CrewHu, go to Settings → Integrations → API and copy the token.
2. 2
   Open the CrewHu card in Current

   In Current's left sidebar, open Integrations (it sits under Admin), find the CrewHu card, and press Manage. One caveat today: the Integrations link only appears in the sidebar for Tenant Admins. A sales manager who isn't also a Tenant Admin can still get there — go to the /integrations page directly, and the CrewHu card and its controls render for them.
3. 3
   Paste the token

   Paste it into the CrewHu API token box and press Save token. Current checks it live against CrewHu before storing anything, so a bad token fails immediately instead of silently. The field is write-only — the token is kept server-side and is never shown back to you or to the browser.
4. 4
   Let the history import run

   Saving the token starts the backfill straight away. The drawer shows an "Importing survey history" progress bar with a running count; it works in the background, so you can close the drawer and keep working. Panels fill in as history lands.
5. 5
   Confirm the connection

   Press Test connection any time. Current reports how many surveys the token can see — a fast way to prove the token is still good without touching your data.

A CSAT survey is only useful once Current knows which account it belongs to. Company names in CrewHu rarely match your PSA exactly ("ACME Inc." vs "Acme"), so Current tries five things in order and stops at the first confident answer.

Anything that can't be matched confidently is skipped rather than guessed at, and counted. If any surveys were skipped, the CrewHu drawer says so and gives you two buttons: Run matching diagnosis, which dry-runs the matcher over the 100 most recent surveys and shows how many landed on each tier plus up to ten examples that fell through; and Re-run full sync, which re-fetches and re-matches your entire history.

CrewHu scores a survey Positive, Neutral, or Negative. Under Alerts & churn flag you choose which of those raise a follow-up (both Negative and Neutral are on by default). When a qualifying survey lands, Current does four things in one pass:

• ＋Sends a bell notification to the company's owner and to sales leadership.
• ＋Emails the same people — each person can mute that email in their own notification preferences; the bell always fires.
• ＋Creates an urgent follow-up to-do for the account owner, due a set number of days from when the alert fires (2 by default, adjustable from 1 to 30). If the company has no owner yet, no to-do is created — the bell still goes to leadership. Because it's a normal Current to-do, it rides through to the owner's Microsoft To Do like any other — if they've connected Microsoft 365.
• ＋Writes an entry on the company's timeline, credited to CrewHu CSAT, so the record shows what happened and when.

One bad survey is a bad day. A pattern is a churn signal. Current raises a churn-risk flag on a company when both of these are true inside the same window: there are at least a set number of negative surveys, and negatives make up at least a set percentage of that window's surveys. Out of the box that's 2 negatives in the last 90 days and at least 25% negative — so a company with two bad surveys out of three flags, and one with two bad out of forty doesn't.

All three numbers are yours to set in the drawer: the window (7–365 days), the negative count (1–50), and the negative percentage (0–100). The flag raises once, alerting the owner and leadership the same way a survey alert does, and clears itself automatically when the company recovers or the bad surveys age out of the window — so a relapse later can raise it again.

For how CSAT combines with tickets, email, and activity into a single risk read, see "Churn and sentiment scoring: reading the risk flags". For how the token is protected, see "How Current stores credentials and secrets".