Habit 5

Be Precise About Dates, Metrics, And Sources

Avoid ambiguous Amazon sales numbers by specifying dates, metrics, grouping, and freshness.

Sales Is Not One Number

Amazon has multiple valid numbers for "sales." Ordered sales, ad-attributed sales, Business Reports sales, near-real-time order totals, and finalized reporting data can all answer different questions. A precise prompt prevents the model from choosing the wrong definition for your workflow.

What To Specify

SpecifyExamples
Date range"Yesterday," "last 7 completed days," or exact dates.
Sales metricOrdered sales, ad-attributed sales, or Business Reports sales.
GroupingBy SKU, account total, campaign, keyword, or marketplace.
FreshnessNear-real-time, latest snapshot, or finalized reporting data.

Copyable Prompt

Use Agent Central to show yesterday's ordered sales by SKU and product name. Include total revenue, units, orders, exact date range, data source, and freshness caveats.

Precision Template

Use Agent Central to answer this using precise metric definitions:

Question:

Date range:
Metric definition:
Grouping:
Data source:
Freshness expectation:

If there are multiple valid Amazon numbers for this question, explain which one you used and which ones you did not use.

Date Behavior

Use exact dates for saved reports and postmortems. Use relative dates like yesterday or last 7 completed days for recurring workflows that should adapt each time they run.

Ask for source and freshness caveats when the answer may mix near-real-time orders, daily snapshots, attribution windows, or finalized reporting data.

Be Precise About Dates, Metrics, And Sources — agentcentral