Qia is a high-precision logic engine. While it can handle simple queries, you unlock its true power when you provide Context, Constraints, and Comparison. Think of yourself as a CEO giving a brief to a senior analyst: the clearer the brief, the sharper the strategy.
Context (The 'Who' and 'What'): Don't just ask for "sales." Ask for "sales from repeat customers" or "sales attributed to Instagram."
Constraints (The 'When'): Qia is a historian. Always define your window—"last 30 days," "Q4 2025," or "since the Spring Sale started."
Comparison (The 'So What?'): Data is silent in isolation. Ask Qia to compare results against the "previous month" or "same period last year" to reveal the trend.
Good: "Show me my AOV."
Result: A single number.
Better: "Calculate AOV broken down by utm_medium so I can see which channel brings the big spenders."
Result: A segmented table revealing channel quality.
Best: "Calculate AOV by utm_medium for last month, compare it with the month prior, and explain which channel is losing steam."
Result: A full strategic audit with a direct diagnosis of your marketing health.
Remember, Qia knows your Persona Handbook. You don’t need to explain what a "VIP" is every time. If you’ve defined it in your Persona Editor, you can simply ask: "How many VIPs bought or Type A items this week?"
If you can say your question in one breath, it's usually a great prompt. If it takes three breaths, try breaking it into a Follow-up Question. Qia remembers the conversation, so you can always say: "Now filter that just for orders over $100."
Qia doesn't just display data; it listens to your store. The interaction follows a two-stage "Logic & Narrative" cycle designed to move you from raw numbers to strategic action.
When you ask a question, Qia’s engine performs a surgical strike on your database. It translates your English request into code, retrieving the specific rows and columns that hold your answer. The result is a Report—a clean, tabular view of your store’s reality. At this stage, the data is "silent," waiting for interpretation.
Once the report is generated, you invoke the "Whisperer" by asking Qia to Explain the Numbers. This is where the heavy lifting happens. Qia ingests the report data and provides a narrative diagnosis—identifying trends, spotting anomalies, and suggesting your next move. This two-step process ensures that the math is always validated before the strategy is formed.
You will notice that different plans have different Row Limits for narratives (ranging from 1,000 to 10,000 rows). While Qia can process vast amounts of data to create a report, we intentionally cap the amount of data passed to the "Narrative Engine" for one fundamental reason: Cognitive Precision.
AI models, like human consultants, can suffer from "data fatigue." When a model is fed too many data points at once, its ability to find the "signal" within the "noise" diminishes. By focusing the Narrative Engine on the most relevant rows—whether it’s the top 1,000 for a Basic store or the top 10,000 for a Super Qia enterprise—we ensure the analysis remains sharp, accurate, and free from the "hallucinations" or vagueness that occur when models are overwhelmed by raw numbers.
We don’t just give you a wall of data; we give you a high-resolution interpretation of what matters most.
If the Logic Engine is the "brain," the Persona is the "memory." It is a centralized configuration file that houses your shop’s unique vocabulary, business rules, and strategic priorities. Instead of explaining your business model every time you ask a question, the Persona ensures Qia already knows your "VIP" customer thresholds, your fiscal year start, and your specific definition of "Gross Margin."
To get you started, Qia performs an Initial Compilation. Upon installation, the engine scans your public storefront and metadata to build a "Draft Persona." It identifies your brand voice, your primary currency, and your public-facing product categories.
While the initial draft is a strong starting point, the true power of the Persona lies in your edits. You are invited to refine this handbook to include internal jargon that no outsider would know:
Custom Segments: Define exactly what "High Churn" means for your specific niche.
Key Milestones: Save your specific sale dates (e.g., "The 2025 Spring Refresh") so you can simply ask, "How did the Refresh compare to last year?"
Logic Guardrails: Set your preferred calculation methods for metrics like Net Revenue or Marketing Contribution.
Eliminate Ambiguity: Prevents the AI from guessing. When you say "Profit," Qia uses your formula, not a generic one.
Conversational Velocity: You can use shorthand. "Show me my Heroes" works because you’ve already defined what a "Hero" product is in the Persona.
Universal Consistency: Every report and every narrative explanation will align with the rules set in the Handbook, ensuring a single version of the truth for your entire team.
Qia operates on a Prorated Fairness model. Your monthly credit grant is calculated based on when you join the billing cycle.
The 1st of the Month: You receive your full monthly credit allowance.
Mid-Month Sign-up: We calculate the "Time Remaining" in the month (Days / 30) and grant you a proportional amount of credits. You are only billed for the portion of the month you actually use.
Monthly Reset: Monthly subscription credits do not carry over; they refresh on the 1st of each month to keep your analyst ready for a new month of data.
During your Initial Trial Period, we apply a Safety Phase (25%). This allows you to explore the engine's full capabilities while we sync your store’s history.
As soon as the trial ends and the first billing cycle begins, our Just-In-Time (JIT) Sync automatically unlocks the remaining 75% of your balance.
If you are having a high-intensity month, you can purchase Manual Top-up Packs.
No Expiration: Unlike subscription credits, manual top-ups never expire. They stay in your "Wallet" until you use them, even across different billing cycles.
Priority Use: Our system smartly uses your monthly subscription credits first, only dipping into your "Wallet" once the monthly allowance is exhausted.
Qia is a high-precision Order and Marketing logic engine. It is built to reason across your transactional history to find the "thread" between sales, attribution, and profit.
Order Logic: Qia has full visibility into your order data. This includes transaction totals, line items, fulfillment status, and financial breakdowns (taxes, shipping, discounts).
Marketing Attribution: The engine analyzes UTM parameters and referral sources tied to your orders to calculate channel-specific ROI.
Historical Depth: The "memory" of the engine is determined by your plan tier. While Qia can process your most recent performance on any plan, higher tiers unlock deeper historical data for multi-year trend analysis.
Yes. Qia is built on a Security-First Architecture designed for high-precision AI auditing and absolute merchant isolation.
The Specialized Schema: To ensure the AI can "reason" with surgical precision across your entire history, we sync your data into a specialized, AI-optimized wide-table schema. This architectural "Flat View" allows the Logic Engine to perform deep-pattern analysis that is impossible within the constraints of standard relational databases.
Isolated Storage: Unlike apps that pool merchant data into a single database, we keep your data strictly separate. Your information is stored in a dedicated environment, gated by your unique personal hash, ensuring your store's logic never crosses paths with another.
PII Logic Guardrails: We treat customer identity with extreme care. As a deliberate security measure, Qia provides deep links to Orders and Products for immediate action, but it cannot deep-link to Customer Profiles. We deliver the strategic trends; your sensitive customer CRM remains in the primary Shopify vault.
No Model Training: We strictly adhere to a Zero-Leakage policy. Your private business data, order details, and trends are never used to train or refine global AI models. Your competitive advantage remains entirely yours.
To give you a "Live Pulse" instantly, Qia immediately indexes your 100 to 200 most recent orders during initial the handshake. This allows you to start running reports on current activity within seconds.
However, Qia is more than a dashboard; it’s a Query Algebra Engine. For deep historical audits (YoY trends, LTV cohorts, and long-term attribution), we must process every single data point in your shop's history. Depending on your volume, this "Deep-Sync" takes a few minutes per month of history.
Qia doesn't just "fetch" a saved chart. When you ask a question, our Query Algebra Engine performs Autonomous Data Engineering in real-time:
Live Synthesis: It translates your question into complex query (a task that takes a human analyst 20 minutes).
The Data Scan: It performs a high-intensity scan across your raw order and attribution tables.
Self-Correction: If the engine detects a logic error or a schema mismatch during the scan, it automatically rewrites and re-runs the query to ensure 100% accuracy before you see it.
Once a report is successfully generated, Qia caches the result. If you revisit that report or ask the same question again later that hour, it will load instantly. You only pay the "time tax" (and the credit cost) when you ask Qia to perform a brand-new engineering task on your data.