PIING
Reading Piing: IntelligenceHow Intelligence works
Piing: Intelligence
- How Intelligence works
- The fields it writes
- Profile refreshes and legacy units
- Preview your profiles
- Custom fields and prompts
- The storefront pixel
- Run and schedule analyses
- Where your data goes
In this guide
- What it produces
- How an analysis runs
- What is sent where
- Try it before you commit
- Next steps
Search or browse documentation
Piing: Intelligence 1 / 8
How Intelligence works
Piing: Intelligence reads your Shopify order history, builds a structured customer profile with AI and rules, and writes those fields to metafields you own.
Piing: Intelligence is a Shopify app that turns the order history you already have into a structured customer profile. There is no loyalty programme to launch, no quiz to build, and no new tracking to add before it works. The app reads each customer's past orders, infers a set of fields about them, and writes those fields back to Shopify customer metafields that belong to your store.
Because the output lives in Shopify metafields, Shopify Flow, customer segments, Klaviyo, Attentive, POS extensions and other metafield-aware tools can use it. The merchant still configures the receiving segment, flow or personalisation rule.
What it produces
Every analysed customer gets a profile made of two kinds of field:
- AI-inferred fields, such as preferred category, preferred style, brand affinity, likely sizes, likely gender, customer persona, likely next category, and seasonal buyer. These are inferred by a language model from the products a customer has bought.
- Rule-based fields, such as AOV tier, purchase frequency, sale sensitivity, reactivation risk, lifetime value tier, average days between orders, and preferred price range. These are calculated directly from the order data with no AI involved.
For the full list, see The fields it writes.
How an analysis runs
- You install the app and grant read access to orders and customers. No code or developer is needed.
- The app reads each customer's order history. AI-inferred and rule-based fields are produced for every customer in the run.
- Each value is given a confidence score, and your output rules decide what actually gets written.
- The accepted values are written to that customer's Shopify metafields, along with a summary field that records confidence and source.
A run starts when you press Run analysis, on a schedule you switch on yourself, or for a single customer shortly after they place an order. Every run in the history says which. See Run and schedule analyses.
What is sent where
Order data is read from Shopify. Only the AI-inferred fields involve an outside service: for those, product titles, variants, product types, vendors, prices, and order dates are sent to Anthropic to produce the inference. Rule-based fields are calculated inside the app and never leave it.
The Intelligence database stores shop configuration, run metadata, optional pixel events, anonymous visitor identifiers and operational logs. It does not store customer names, emails, phone numbers or addresses in its own tables. The selected profile values live in Shopify metafields. For the full data path, see Where your data goes and the security page.
Note
Intelligence is a standalone Shopify app and needs no hardware. If the retailer also runs Piing, Intelligence fields can appear in the Piing staff view and Piing visit context can sit on the same Shopify customer record.
Try it before you commit
You can preview the proposed fields for up to five real customers with orders, with no allowance spent. See Preview your profiles.
Next steps
Related guides
Was this guide helpful?
Next guide in Piing: Intelligence
The fields it writes
The full set of AI-inferred and rule-based customer fields Piing: Intelligence writes to Shopify metafields, plus the storefront pixel fields and the summary field.
Still need a hand? Contact us