Ecommerce dashboard Data Studio builds that show profit, not just revenue

What data does an ecommerce dashboard Data Studio build need?

It needs orders, costs and spend from their real sources. Orders come from Shopify, WooCommerce or Amazon Seller Central through a partner connector or export. Ad spend comes from Google Ads natively and from Meta, TikTok and others through partner connectors. Product costs, shipping rates and payment fees usually live in a Google Sheet the finance team already keeps.

GA4 adds sessions, conversion rate and funnel steps, but we do not use it as the revenue source. Consent banners, ad blockers and cross-device purchases mean GA4 purchase revenue rarely matches the store, and a report that disagrees with the bank account loses trust in a week.

How do you show contribution margin in Data Studio?

Contribution margin is net revenue minus product cost, fulfillment, payment and marketplace fees, and ad spend. We calculate it at order line level in the data layer, then roll it up by day, channel, product and market.

Doing this in a calculated field inside a chart is where most ecommerce reports go wrong: margin ratios get averaged instead of re-calculated, and refunds land in a different month from the sale. We build the margin table in Sheets for smaller stores or BigQuery for larger ones, with refunds matched to the original order, so every chart sums from the same rows.

• Gross sales, discounts, refunds and net revenue.
• Cost of goods, shipping and fulfillment, payment and Amazon fees.
• Ad spend by channel and contribution margin after marketing.

Can Data Studio show customer cohorts and repeat purchase?

Yes, once each order carries the customer's first order month. We add that field in the data layer, then a pivot table with first-order month as rows and months since first order as columns shows retention and cumulative revenue per cohort.

This is the view that tells you whether a promotion brought in customers who come back or one-time discount buyers. Paired with acquisition cost by channel, it shows which channels bring customers worth paying for.

We also add a 90-day repeat rate by acquisition channel and by first product bought. Those two cuts usually say more about where to spend next quarter than any single ROAS figure, and they need nothing beyond the order history you already have.

How do you reconcile ad spend with actual orders?

We put platform-reported revenue next to store revenue for the same period, so the gap is visible instead of hidden. Meta, Google Ads and TikTok each claim conversions on their own attribution windows, and summed together they often exceed what the store actually sold.

The summary page uses store orders and total spend to give a blended ROAS and marketing efficiency ratio. Channel pages keep platform figures for day-to-day optimization, labeled as platform-reported. A daily reconciliation table flags days where spend is missing, usually a sign that a connector token has expired.

Does it work for Amazon and multi-store sellers?

Yes. Amazon Seller Central data arrives through a partner connector or scheduled report exports, and we map it to the same order schema as your own store. Fees and FBA charges are separated so marketplace margin is comparable with direct sales.

For brands running several Shopify stores or regions, each store becomes a value in a store field, currencies are converted at a documented rate, and a report-level filter lets each regional manager see their own market.

When does an ecommerce brand need BigQuery?

When order volume or history makes direct connectors slow or unreliable. Line-level order data grows fast, and blends of up to 5 sources re-query live each time a page loads.

Loading orders, spend and costs into BigQuery tables partitioned by date, with scheduled queries building the margin table each morning, keeps the dashboard fast and the logic in one place. Smaller stores often run well on Sheets and partner connectors for a long time, and we will say so in the review.

What does a founder see when they open the report each morning?

The first page answers one question: did yesterday make money? It shows net revenue, orders, average order value, ad spend, blended ROAS and contribution margin for yesterday, month to date and the same period last year, each with a small trend line.

Below that, a channel table splits new and returning customer revenue, because a month carried by repeat buyers looks healthy while acquisition quietly stalls. A product table ranks SKUs by margin rather than revenue, which often reorders the list: a bestseller with high returns and heavy discounting can sit near the bottom.

Pages for operations follow, covering refund rate by product, fulfillment cost per order and stock cover where inventory data is available. Every page shares the same date control, so a manager comparing Black Friday weeks across two years changes one setting, not twelve charts.