by Trivas.aiSearch "commerce kpis dashboard" and you'll get the same article eleven times over: a bulleted list of twenty metrics, each with a one-line definition, no indication of how any of them connect. Revenue sits next to CAC sits next to inventory turnover, as if they're unrelated line items instead of parts of the same system.
A real commerce KPIs dashboard is one view that ties revenue, ad spend, margin, and retention data together, pulled from Shopify or Amazon, your ad platforms, and GA4, so you can see cause and effect instead of isolated numbers. Spend goes up, margin should move with it. Repeat rate drops, LTV:CAC follows a few weeks later. A glossary can't show you that. A dashboard built with the right structure can.
This piece covers both: the five KPI categories that actually belong on a dashboard, benchmark patterns pulled from real account data, and a free template you can build from in under an hour.
Revenue and growth. Net revenue, AOV, revenue per session. The mistake here is blending channels into one number. Shopify revenue per session and Amazon revenue per session tell you different things about different funnels. Report them separately, then roll them up.
Ad efficiency. Blended ROAS, MER, CAC by channel, and some flag for incrementality. Platform-reported ROAS alone isn't an efficiency metric anymore, it's a vanity number most platforms inflate through last-click attribution.
Margin and profitability. This is the category most dashboards get wrong. Gross margin isn't enough. You need contribution margin after COGS, shipping, and ad spend, which is the only number that tells you if a sale actually made you money.
Retention. Repeat purchase rate, LTV:CAC ratio, cohort-based retention curves. Most "dashboards" show a single LTV number with no cohort breakdown, which hides whether retention is improving or quietly decaying.
Operational KPIs. Inventory turnover, stockout rate, fulfillment cost per order. These rarely make it onto a dashboard, which is strange, because they're often the earliest warning signs of a margin problem before it shows up in the P&L.
A dashboard with just the first two categories looks complete. Clean charts, ROAS trending up, revenue climbing. It's also the kind of dashboard that misses a margin collapse until it's already happened, because nothing on it was built to catch it early.
We looked at aggregated, anonymized KPI tracking patterns across Trivas accounts on Redshift, spanning Shopify-only, Amazon-only, and multichannel sellers, to see what dashboards actually look like at different revenue stages, not what they should look like in theory.
The pattern is consistent across tiers: dashboard complexity scales with GMV, but not evenly.
The gap that shows up across nearly every tier, regardless of size: stores track ad-platform ROAS religiously, but don't have true blended MER or contribution margin sitting next to it. Meta says 3.5x ROAS. Google says 4x. Nobody's added up total spend against total revenue to see what the number actually is once both platforms are paying for the same customer.
That gap is the thing most "commerce KPIs dashboard" listicles can't tell you, because it only shows up when you look at how real accounts are actually built, not how a metric glossary says they should be.
Shopify layer. Shopify's native reports cover order count, AOV, and basic sales trends fine. What they don't give you cleanly: discount impact on margin, true repeat purchase rate across customer cohorts, or revenue per session segmented by traffic source. That's where a connector earns its keep. If you're running Trivas on Shopify specifically, the setup is covered in Shopify integration, and you can also find the app listed directly on Trivas AI on the Shopify App Store.
Amazon layer. This is the layer people underestimate. Amazon margin math is harder than DTC margin math because FBA fees, referral fees, and ad spend get pulled from Seller Central payouts, not reported alongside Amazon Ads spend in any single place by default. Reconciling Amazon Ads cost against what actually hit your payout requires joining two data sources that don't talk to each other natively. Amazon's solutions layer exists specifically because this reconciliation is where most sellers' margin numbers go wrong.
Meta and Google Ads layer. Pull platform-reported ROAS, then check it against GA4 and actual Shopify order data. If Meta and Google combined are claiming more revenue than your store actually processed, you've got over-attribution, and it's extremely common. Every platform wants credit for the same conversion.
GA4 layer. GA4 is useful for funnel diagnostics: session-to-cart rate, cart-to-checkout rate, where people actually drop off. It's a poor source of truth for revenue, though. GA4 undercounts due to cookie consent gaps, ad blockers, and attribution model mismatches with your actual order data. Use GA4 data for funnel shape, not for the top-line revenue number on your dashboard.
Blending platforms into one ROAS number. It feels cleaner. It also hides which channel is actually profitable and which one is riding on the other's coattails.
Manual spreadsheet refreshes. Daily or weekly manual pulls mean you're always making decisions on data that's already stale by the time it's in the sheet. A live warehouse feed removes that lag entirely.
Vanity metrics above the fold. Impressions and sessions look good on a slide. They don't tell you if the business is healthy. Margin and retention should sit at the top, not buried under a chart of traffic.
No historical baseline. Without last year's numbers next to this year's, you can't tell if a 15% dip is seasonal or the start of a real problem. Every KPI on the dashboard needs a comparison point, not just a current value.
The template is a pre-built tab structure covering all five categories from above: revenue and growth, ad efficiency, margin, retention, and operations. Each tab has formulas already set up, you just plug in your own numbers (or connect a live feed, more on that below).
If you want a starting point before downloading anything, here's the minimum 12 KPIs worth tracking, grouped by category:
If you're Shopify-only, drop the Amazon-specific margin fields (FBA fees, referral fees) and focus the operational tab on fulfillment cost and stockouts. If you're Amazon-only, you'll want a separate reconciliation row for ad spend versus payout, since that's where the numbers tend to drift. Multichannel sellers should keep blended revenue and margin on one master tab, but split operational metrics by channel, since Amazon and Shopify inventory and fulfillment costs behave differently.
Manually keeping any of this updated works fine at low order volume. It stops working once you're pulling from three or four data sources daily. If you'd rather have these KPIs auto-populated from live Shopify, Amazon, ad platform, and GA4 data instead of typing numbers into a spreadsheet every morning, Trivas's BI reporting layer does exactly that, and you can try it with a free trial.
What's the difference between a KPI dashboard and a reporting dashboard?
A reporting dashboard shows what happened, usually a wide set of metrics with no judgment attached. A KPI dashboard is narrower on purpose: a curated set of numbers tracked against targets and benchmarks, built to answer "are we on track" at a glance.
How many KPIs should be on a commerce dashboard?
Ten to fifteen, max, per audience. More than that becomes noise nobody actually checks. Split views by role instead of cramming everything into one screen, a founder view looking very different from a performance marketer's view, which is one reason marketing leaders and founders often need separate dashboard setups even inside the same company.
Should Amazon and Shopify KPIs live on the same dashboard?
Yes, for blended revenue and margin, since that's the real picture of the business. But keep channel-specific operational metrics, like FBA fees or Shopify app costs, on separate tabs. Mixing those in with the blended view just adds clutter without adding clarity.
How often should a commerce KPIs dashboard refresh?
Daily, at minimum, for ad spend and revenue. If you're adjusting ad budgets daily (most performance teams are), you want near-real-time data, not numbers that are a day or two stale by the time you act on them.
A metric list isn't a dashboard. A dashboard is the five categories above working together, measured against real benchmarks, so you can tell the difference between a seasonal dip and an actual problem before it costs you money.
Spreadsheets hold up fine at low order volume. Past a certain point, manually reconciling Shopify, Amazon, ad platforms, and GA4 every morning just isn't sustainable, and that's usually the moment teams start looking for something built to do it automatically.
If you want to see what this looks like running on live Redshift data instead of a manually updated template, take a look at Trivas's BI reporting product, or just talk to a founder directly about what it would take to get your accounts connected.
explore more insights
3 min read
3 min read
3 min read
.d53b12e5.png&w=3840&q=75 "Trivas.ai")