Stop Guessing: Use Shopify Analytics to Drive Every Merchandising Call

Most Shopify merchants have access to better data than they realise. The problem is rarely a shortage of numbers. It's the gap between reading a report and actually changing what sits on your collection pages, how you price a bundle, or which SKUs you push in email. This post closes that gap.

What Shopify's Analytics Stack Looks Like Right Now

The platform has moved fast. The October 2024 analytics framework overhaul gave every store cleaner session metrics, a revised checkout-funnel methodology, and product-title accuracy improvements that make historical comparisons more reliable. Building on that foundation, a series of 2026 updates have made the toolset meaningfully more useful for day-to-day merchandising.

A few highlights worth knowing:

• Automated insight cards on your Home feed. As of April 20, 2026, Shopify started surfacing automated data summaries directly on the admin Home feed for merchants averaging 10 or more orders per week. The system now monitors over 80 data combinations and flags session trends broken down by geography, referrer, device type, and landing page, plus fulfillment efficiency broken down by carrier and sales channel. You spot the pattern without digging through reports.
• Event annotation markers. Analytics reports now overlay markers for store events like product changes, theme deploys, and app installs. When your conversion rate dips on a Tuesday, you can see immediately whether a theme change or a new app coincided with the drop.
• Scatter plots and radar charts in the custom report builder. The Summer 2026 Editions (released June 2) added these chart types, so you can compare metrics like conversion rate versus average order value across product categories without exporting to a spreadsheet or paying for a separate BI tool.
• Inventory audit trails. Inventory adjustments now carry a complete audit trail showing who changed what and when. For merchandising, this removes the guesswork when sell-through rates look odd after a stock correction.
• "Sales reversals" labelling. What used to show as Returns in analytics is now labelled Sales reversals to capture all order adjustments, not just physical returns. That matters for apparel and other high-return categories where the old label understated the true reversal rate.

Sidekick as Your Merchandising Analyst

The feature that has genuinely changed how smaller teams can work with data is Sidekick's natural-language query capability. The analytics query editor now integrates with Sidekick and understands natural language, making advanced reporting accessible to everyone. You type a question, Sidekick translates it into ShopifyQL, and you get a report with a plain-English explanation of what it measures.

Practical examples that map directly to merchandising decisions:

• "Show me revenue by collection for the last 60 days, split by new versus returning customers."
• "Which products in my Summer collection have a sell-through rate below 20%?"
• "What's the average order value for customers who came from Instagram versus organic search?"

Sidekick can now write ShopifyQL queries for web performance and payments data, and it has expanded its ability to write fulfillments and payouts queries. That means you can ask about fulfillment speed by carrier in the same session you ask about product margin, without switching tools.

The caveat: Sidekick works best when you ask precise, scoped questions. Vague prompts produce vague answers. Treat it like briefing a junior analyst, not wishing on a magic lamp.

The Native AI Merchandising Tools in Summer 2026 Editions

The Summer 2026 Editions, which shipped on June 2, bundled over 150 updates. For merchandisers specifically, the headline items are three tools now built directly into the admin: AI Collection Sort, Predictive Cross-Sell Blocks, and a Merchandising Insights panel. These replace functions that most stores previously handled with paid third-party apps.

Shopify released AI Collection Sort, Predictive Cross-Sell Blocks, and a Merchandising Insights panel, all embedded in the admin to replace third-party merchandising apps. The collection sort uses live store signals (conversions, revenue per visitor, inventory levels) to re-rank products automatically. You set the rules; the algorithm executes them on a schedule.

The practical upside: automated merchandising can reshuffle collections to highlight high-converting items or clear stagnant stock without manual intervention. That is especially valuable for stores with large catalogues where manually auditing every collection page once a week is not realistic.

Four Merchandising Decisions You Should Be Making from Data

1. Collection sort order

Stop sorting by "best selling" globally and start sorting by what converts within each traffic segment. A product that ranks first for email traffic may rank fifth for paid social. The new AI Collection Sort, combined with Sidekick queries breaking down conversion rate by referrer and landing page, gives you the inputs to set differentiated rules.

2. Bundle construction

Bundle performance metrics were added to Shopify Analytics in the Winter 2025 Editions update, and they are underused. Pull your bundle report, compare revenue-per-impression against individual SKU performance, and look for bundles where the attach rate is high but the bundle itself is not prominently placed. Shopify's predicted spend tier feature can help you anticipate high-value customers ready for cross-sell so you can target bundle promotions at the right segment.

3. Markdown timing

A missed markdown window costs margin; a premature discount trains customers to wait. The sell-through rate report, combined with inventory audit trail data, tells you exactly how fast a SKU is moving. Set a threshold in Shopify Flow: when sell-through on a variant drops below a defined percentage in a rolling 14-day window, trigger an internal alert or auto-apply a clearance tag. No spreadsheet needed.

4. Channel-specific assortment

You can now publish or unpublish individual product variants per sales channel and catalog, which means your wholesale catalog and your DTC storefront can carry different variant sets. Pair that with the sales-by-channel analytics to identify which variants drive disproportionate margin on each channel, then trim the assortment accordingly.

Shopify as Your Revenue Source of Truth

One thing worth cementing before you layer on any third-party attribution tool: use Shopify as your revenue source of truth for financial reporting and P&L work. Privacy restrictions in 2026, including iOS ATT and the Chrome Privacy Sandbox, limit GA4 tracking, while Shopify's server-side approach remains unaffected. That means Shopify's transaction data is the most complete count of what actually sold. GA4 remains useful for pre-purchase behaviour and user journey mapping, but for decisions about what to stock, what to mark down, and what to promote, start from Shopify's numbers.

Where Native Analytics Hits Its Limits

Being honest: for stores above roughly half a million in revenue running multi-channel operations, native Shopify Analytics will show you what happened but not the full profit picture. Ad spend, COGS from your accounting system, and blended channel margin require stitching together data from outside Shopify. Tools like Triple Whale, Northbeam, or a managed analytics layer become relevant at that stage. The new scatter plot and radar chart visualisations in the custom report builder close some of that gap for mid-market stores, but if your finance team is still building margin analysis in spreadsheets, the numbers are probably incomplete.

The Practical Starting Point

If you take nothing else from this post, start here:

1. Check your Home feed daily. The automated insight cards now flag meaningful deviations across 80+ data combinations. This takes 90 seconds and surfaces things you would otherwise miss for days.
2. Use Sidekick to build the three reports you actually need. Stop scrolling through default reports. Ask for the specific cuts of data that map to your current decisions.
3. Set one Flow automation tied to a sell-through threshold. Pick your slowest-moving category. Define a threshold. Automate the alert. That is data-driven merchandising in practice, not in theory.

The tools are there. The gap is almost always in creating the habit of acting on what the data says.