# AI health assessment

***

<figure><img src="https://1065233822-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fttka6LKL3bc1LsaDzuI5%2Fuploads%2FcTxckBNVoXpSzvErWy2O%2FScreenshot%202026-04-20%20at%2020.40.47.png?alt=media&#x26;token=5c4bb86c-955d-4302-877b-eefe666e5041" alt=""><figcaption></figcaption></figure>

### AI Assessments

**Where to find it:** Every customer profile → Assessment panel · Settings → AI **Who can access it:** All team members

***

#### What it is

AI Assessments is the intelligence engine at the core of Startdeliver. It reads all available context for every customer — product usage, experience feedback, support activity, relationship engagement, revenue data, and more — and produces a plain-language assessment of where each customer stands and what your team should do next.

The assessment runs continuously and automatically. It does not need to be triggered. It does not require a CSM to fill in a form. When new data comes in — a survey response, a delayed ticket, a drop in usage, a week of silence — the assessment updates.

***

#### What the assessment produces

On every customer profile, the Assessment panel shows:

**Status** — a plain-language verdict on the customer's current state. Not a number. Not a colour alone. A label that means something: Strongly Thriving, Actively Engaged, Neutral, Needs Attention, At Risk.

**Summary** — one or two sentences describing what is driving the status. What the AI actually found, stated plainly. "Strong engagement with deep adoption, proactive communication, and expansion signals." Or: "Usage declining across core features, last survey response poor, no CSM contact in 6 weeks."

**Trend** — how the assessment has moved over the past weeks and months. Whether things are improving, holding steady, or deteriorating.

**Take Action** — a recommended next step tied specifically to what the AI found for this customer. Not a generic playbook item — a specific, contextual recommendation.

**See full analysis** — the complete breakdown of every signal the AI read, the data behind each one, and the reasoning behind the assessment.

<figure><img src="https://1065233822-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fttka6LKL3bc1LsaDzuI5%2Fuploads%2FPrIJBph1CoYJceyblcgw%2FScreenshot%202026-04-20%20at%2022.10.16.png?alt=media&#x26;token=7d3013cd-209e-415b-bc10-bb0ee96cb04d" alt=""><figcaption></figcaption></figure>

***

#### What the AI looks for

The AI reads the full picture of the customer relationship. Key signals include:

**Product engagement** — are users active? Is adoption deepening or narrowing? Are the right users using the right features?

**Customer sentiment** — what are survey responses telling you? Is sentiment improving or declining? Are there signals of an advocate or an early warning of churn?

**Support health** — are there open or delayed tickets? Is the customer getting timely help or quietly struggling?

**Relationship activity** — when did your team last meaningfully engage? Is the customer active in Shared Space? Are emails and messages flowing or has communication gone quiet?

**Revenue context** — what is the account worth? When does it renew? Is seat utilisation growing or shrinking?

The AI does not treat these signals as separate scores to average. It reads them together. A customer with strong usage but declining sentiment and a two-month communication gap is a specific situation — the AI identifies the pattern and tells your team what it means.

***

#### Flags the AI raises

Beyond the general assessment, the AI identifies specific patterns worth flagging:

**Churn risk** — a customer showing early warning signs: declining usage, poor feedback, unresolved support issues, or communication that has gone quiet. Surfaced proactively so your team can act before the customer raises it.

**Expansion opportunity** — a customer with strong adoption, positive sentiment, and growing usage who may be ready for an upgrade, an additional product, or a wider rollout.

**Reference and advocate** — a customer with consistently strong feedback across multiple contacts. Worth a conversation about a case study, reference call, or partnership.

**Silent customer** — a customer who has stopped engaging across all channels. Not necessarily churning, but worth attention before the silence becomes a decision.

***

#### You are in control

The assessment is powerful out of the box, but it is designed to work the way your business works — not the other way around.

Through AI instructions in Settings → AI you control:

**Your instructions** — give the AI specific guidance about your business, your customers, and what matters to your CS motion. The AI applies this context to every assessment it runs.

**Team access** — choose which team members can view and use assessments.

**Customer scope** — select which customers get assessed. Useful if you are rolling out AI features progressively or have customer segments that require different treatment.

**Interaction settings** — fine-tune which interactions are included in the context, and create custom interaction types specific to your business.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.startdeliver.io/health-and-metrics/ai-health-assessment.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.