PIING
Reading Piing: IntelligenceRun and schedule analyses
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
- How a run starts
- The schedule is off until you turn it on
- The refresh after an order
- Manual runs
- Reading a finished run
- Each customer gets 90 seconds
- A run that shows Interrupted
- Re-sync metafields
- Keeping profiles fresh
- Next steps
Search or browse documentation
Piing: Intelligence 7 / 8
Run and schedule analyses
The ways an analysis can start, what each run reports about the fields it wrote to Shopify, and how to re-sync metafields if a store's customer records look empty.
Piing: Intelligence only analyses customers when something starts a run, and every run says where it came from. This article covers the ways a run can begin, what the numbers on a finished run mean, and how to put a store right if its Shopify customer records look empty.
How a run starts
Open Settings then the Schedule tab and the app spells out the only three ways a full analysis begins:
- Manual. You press Run analysis.
- Scheduled. On the schedule below it, and only while the schedule is switched on.
- Order. For a single customer, a couple of minutes after they place an order, if that setting is on.
Three more labels appear in the run history, for work the app does on your behalf:
- Onboarding. The first analysis, kicked off when you finish setting the app up.
- Storefront. A single customer refreshed when your storefront asks for an up-to-date profile.
- Re-sync. The metafield re-sync described below. It rewrites fields and calls no AI.
Every entry on Analysis runs carries its source, so a run never appears from nowhere.
The schedule is off until you turn it on
Scheduled runs are included on every plan, but they do not run until you switch them on under Settings then Schedule.
- Tick Run an incremental analysis on a schedule.
- Choose a cadence, daily or weekly on Sundays, and the hour it should fire in UTC.
- Save.
A scheduled run is incremental: it only processes customers with new orders since the last run, so allowance is used on customers whose order history has changed. The tab shows when the last scheduled run happened. A scheduled run is skipped while your subscription is inactive or your analysis allowance is spent for the cycle. See Profile refreshes and legacy units.
The refresh after an order
Refresh a customer's profile after they place an order is off by default for new stores. Turn it on under Settings then Schedule if you want automatic refreshes after orders. Existing stores keep their saved setting.
When enabled, roughly two minutes after an order, that one customer's profile is refreshed and the run shows in the history marked Order. Turn it off in the same place to stop future order-triggered refreshes.
Manual runs
Start a run by hand at any time, for example after you enable new fields, add a custom prompt, or import a batch of customers. A manual run analyses up to the customer limit for your plan:
- Starter: up to 500 customers per run
- Growth: up to 2,500 customers per run
- Scale: up to 7,500 customers per run
These are the limits for new profile-refresh plans. Each successful customer refresh counts once, regardless of enabled fields. The remaining cycle allowance limits the work available; existing field-unit subscribers keep their approved limits.
Reading a finished run
Open Analysis runs and click a run. Alongside customers processed, failed and skipped, each run reports:
- Fields written to Shopify. How many values actually reached your customer records. This is the number that matters. A run that processes customers and writes nothing is not a successful run, and the app says so rather than reporting a clean finish.
- Write failures, and, under Why some fields did not reach Shopify, the reasons behind them.
- Profile refreshes used on new plans, or units used on legacy plans, alongside fields per customer.
Rejections that Shopify returns inside an otherwise successful response, which is how it reports rate limiting and permission problems, count as failures rather than successes. Rate-limited writes are retried with a back-off. If the app's connection to your store has expired, the run stops and asks you to reconnect Piing to Shopify instead of writing nothing for every customer.
Background runs, scheduled runs, order refreshes and re-syncs, always use the current connection to your store, so a rotated access token does not turn them into silent failures.
Each customer gets 90 seconds
A single customer is given 90 seconds. If a request to Shopify hangs past that, the customer is recorded as failed with the reason and the run carries on to the next one. One slow customer never stalls the rest of the run.
A run that shows Interrupted
An app restart or provider failure can interrupt a run. Check its recorded status, write counts and error before starting more work.
- Fields already written remain on the Shopify customer record.
- Eligible failed full or incremental runs offer Recover saved work in Analysis runs. Saved checkpoints are retained for up to seven days; recovery availability depends on the recorded stage and retained data.
- On profile-refresh plans, successfully completed customer refreshes still count. Unfinished reservations are released when the run ends. Recovery does not charge a completed refresh twice.
- Starting a separate new run is a new refresh request. Fields are updated in place, but successful refreshes in the new run use the current allowance.
Re-sync metafields
If the app shows profiles but the same customers look empty in Shopify, use Re-sync metafields on the Customers page.
It re-creates and pins the metafield definitions, then rewrites every stored profile field onto your Shopify customer records using the current connection, reporting live progress and the exact reason for any failure. It uses no analysis allowance and makes no AI calls, so a store in that state can put itself right in one click.
Tip
Want to see the output before a full run? Use the free preview on a small sample first. It spends no analysis allowance.
Keeping profiles fresh
Because profiles are written to Shopify metafields, other tools can read the latest values after a run writes them. If you want automatic refreshes, enable the schedule, after-order refreshes, or both. Your saved settings determine which runs can start.
Next steps
Related guides
Was this guide helpful?
Next guide in Piing: Intelligence
Where your data goes
The Intelligence data path from Shopify orders and optional pixel signals to AI processing, output controls and app-owned Shopify metafields.
Still need a hand? Contact us