Data Studio retail dashboards for stores and online channels

What does our Data Studio retail dashboard show?

It shows trading performance across every channel on one set of pages. The retail dashboard we built in Data Studio opens with revenue growth month on month and quarter on quarter, then breaks performance into the measures a head of retail reviews every Monday.

Each view is filterable by store, region, channel and category, and every chart reads from the same blended model, so a number on the overview always matches the number on the detail page.

• MoM and QoQ revenue growth, with the prior period shown alongside.
• Conversion rate for the web store and, using footfall, for physical stores.
• Sell-through rate by category and season.
• Cart abandonment tracked against the forecast you set for it.
• Online versus in-store sales, side by side and as a share of total.
• Store foot traffic by day and hour.
• Inventory shrinkage by store and category.

Where does retail data come from, and how do we join it?

Most retailers run three to five systems that were never meant to talk to each other. Online orders sit in Shopify or WooCommerce, marketplace sales in Amazon Seller Central, till transactions in a POS such as Square or Lightspeed, and door counts in a footfall counter vendor's portal or a CSV it emails each morning.

None of those is a native Data Studio connector, so we bring them in through a partner connector such as Windsor.ai or Supermetrics, or load them into BigQuery on a schedule. Store and SKU mapping tables live in a Google Sheet your team can edit, which matters the day a new store opens or a product is renamed on Amazon but not on Shopify.

We avoid stacking everything into one blend. Data Studio blends take at most five sources and only equality joins, so a retail model with a product master, store master, three sales feeds and footfall needs to be joined upstream, usually in BigQuery. That upstream modeling is most of the work a data studio agency does for a retailer, and it is what keeps the report fast once a second season of SKUs arrives.

How do you compare online and in-store sales fairly?

You agree the rules first, then encode them once. Online and store revenue look comparable until you ask where a click-and-collect order belongs, whether Amazon fees come off before or after the comparison, and which date a return lands on.

We write those decisions into calculated fields, usually a CASE statement that assigns every order to a channel, and document them on a definitions page inside the report. When finance asks why the channel split differs from last year's deck, the answer is on the page rather than in someone's inbox.

Tax and currency get the same treatment. A retailer selling in several countries needs net sales in one reporting currency, and the conversion rate used for each day should be visible, not buried in a formula.

Why do sell-through and shrinkage disagree between teams?

Usually because each team is dividing by a different stock figure. Merchandising may calculate sell-through as units sold over units received, while stores use units sold over opening stock, and both are reasonable until they share a slide.

Shrinkage has a timing problem too. It comes from the gap between book stock and counted stock, so it only moves when a stocktake or cycle count is posted. We show the count date next to every shrinkage figure so nobody reads a quiet month as a good one.

Our job is to pick one definition with you, label it clearly and keep the alternative available as a parameter where a team genuinely needs it.

What makes a retail dashboard slow, and how do we fix it?

Row volume and live API calls. A model at SKU, store and day level grows fast, and Data Studio returns too many rows errors on large queries. We aggregate to the grain each page needs and partition BigQuery tables by date so a date filter scans only the days it shows.

If cart abandonment comes from GA4 events, the GA4 Data API quota of 14,000 tokens per project per property per hour can run out when many people open a busy report. An extract or a daily GA4 export to BigQuery avoids those quota errors.

We also set data source credentials deliberately. Owner's credentials let store managers see the report without access to Shopify or the POS back office, and we keep the owning account on a shared service login so the report does not break when someone leaves.

Who on a retail team uses which page?

Each role gets a starting view rather than a separate report. Store managers open the store page already filtered to their location, merchandisers start on sell-through by category, the ecommerce lead checks conversion and cart abandonment, and finance takes the channel split.

Scheduled delivery sends a PDF of the trading summary to leadership every Monday. On Data Studio Pro, schedules can run hourly and deliver to Slack or Google Chat, which suits peak trading weeks. Reports built under the old name, Looker Studio, carry over unchanged, so an existing store report can be extended rather than rebuilt.