.d53b12e5.png&w=3840&q=75 "Trivas.ai")
by Trivas.aiOpen up most ecommerce dashboards and you'll find 40+ widgets fighting for attention. Pageviews next to impressions next to some engagement score nobody can define. Ask the founder a simple question, "are we profitable this month," and watch them scroll for five minutes before giving up and pinging their accountant.
That's the problem. Not too little data. Too much of the wrong data.
Vanity metrics feel good because they're big numbers that mostly go up. Pageviews, impressions, follower counts. None of them tell you whether to pause a campaign or restock a SKU. The metrics that actually drive decisions tend to be less flattering: contribution margin, blended CAC, repeat purchase rate. Numbers that can go down and force you to do something about it.
This post is a working list of ecommerce KPI dashboard metrics to include, grouped the way a business actually operates: revenue, customers, marketing, and operations. Build your dashboard around these four buckets and you stop drowning in widgets that look impressive but answer nothing.
Start with net revenue, not gross. Gross revenue is what you rang up at checkout. Net revenue subtracts returns and discounts, and it's the number that actually hits your bank account. Brands that lead with gross revenue on their dashboard tend to overstate how good a month really was, sometimes by double digits once returns settle a few weeks later.
Average order value (AOV) matters, but only if you segment it. Blended AOV across new and returning customers hides the real story. Returning customers usually spend more per order; if your blended AOV is flat but new-customer AOV is sliding, you've got a problem the single number won't show you.
Conversion rate by channel is the same issue in a different shape. A blended conversion rate across your site, Amazon, and any marketplaces smooths over exactly where shoppers are dropping off. If Amazon converts at 12% and your own site converts at 1.8%, that blended average tells you nothing actionable. Break it out.
Revenue per visitor (RPV) ties traffic quality to sales outcome in a single figure: total revenue divided by total visitors. It's useful because it catches the scenario where traffic is up but quality is down, something conversion rate alone can miss if order sizes shift at the same time.
Customer acquisition cost (CAC) gets calculated wrong at most companies. Ad spend divided by orders isn't CAC, it's media efficiency. Real CAC is fully loaded: ad spend plus agency fees, plus whatever you pay for the tools running your funnel. Leave those out and you'll think you're more profitable than you are.
Customer lifetime value (LTV) paired against CAC gives you the ratio that actually tells you if the business model works. A 3:1 LTV:CAC ratio is the general floor worth watching. Below 1:1, you're paying more to acquire a customer than they'll ever spend with you, which is a flashing red light no amount of top-line revenue growth will fix.
Early on, before LTV data has matured, watch repeat purchase rate and time-to-second-order instead. These show up faster than lifetime value curves and tend to be the earliest honest signal of product-market fit. If customers aren't coming back for a second order within a reasonable window, no amount of acquisition spend will save the unit economics.
If you run a subscription or replenishment model, churn rate deserves its own line, separate from generic repeat rate. It's the number that determines whether your recurring revenue is actually recurring or just a monthly guessing game.
Here's a gap that trips up a lot of teams: blended ROAS versus platform-reported ROAS. Meta and Google will both happily report ROAS numbers that double-count conversions through generous attribution windows. The gap between platform-reported and blended, real-revenue ROAS usually runs 20 to 40%. If you're scaling budget off the platform number alone, you're scaling off inflated math.
That's why marketing efficiency ratio (MER), total revenue divided by total ad spend, earns a permanent spot on the dashboard. It's the sanity check against whatever Meta or Google claims, and it's much harder to game.
Cost per acquisition (CPA) by channel should sit separate from CAC. CPA by channel is media-only and tells you where to shift budget. CAC is the fully-loaded number that tells you if the business works at all. Confusing the two is how teams end up "profitable" on paper while burning cash in reality. Our ROAS calculator is a quick way to sanity-check platform-reported numbers against your actual blended spend if you want a gut check before building this into a live dashboard.
Finally, track new customer ROAS separately from blended ROAS. Blended ROAS can look healthy while your new-customer pipeline quietly shrinks, because existing customers keep buying regardless of that week's ad performance. If you're only watching the blended number, you won't notice the acquisition engine stalling until it's already stalled for months.
Operations metrics are the ones most marketing-focused dashboards skip entirely, and it shows.
Inventory turnover rate and days of inventory on hand aren't just sales-velocity numbers, they're cash flow numbers. Too much inventory sitting on shelves means cash tied up that could be funding ad spend or new product development. Too little, and you're about to stock out of your bestseller.
Return and refund rate by SKU and category rarely makes it onto a dashboard built by a marketing team, but it should. A product with a 25% return rate is quietly destroying margin no matter how well it converts or how cheap the CAC looks.
Fulfillment cost per order deserves tracking on its own line too. 3PL rates and shipping costs creep up steadily, and revenue can grow while margin erodes in the background if nobody's watching this number month over month.
And stockout rate for top SKUs is the metric that silently caps revenue without ever showing up as a bad number anywhere else. Your sales report won't flag a stockout. It'll just show lower sales for that week, and most teams blame demand or ad performance instead of checking the warehouse.
Some metrics only matter if you actually sell meaningfully on that channel. Add them only when they're relevant, otherwise they're just more noise on the dashboard.
For Amazon sellers, add TACoS (total advertising cost of sales), Buy Box percentage, and the organic-versus-PPC sales split. TACoS in particular catches what ACoS alone misses: whether ad spend is actually growing total sales or just cannibalizing organic orders. If you sell through Amazon, our breakdown on Amazon-specific reporting covers how these numbers pull together into one view.
For Shopify or DTC brands, track the new-versus-returning customer revenue split, and subscription MRR if that applies to your model. For a deeper look at building this out natively, see our Shopify integration guide, or install Trivas AI on the Shopify App Store directly if you want the metrics pulling automatically without manual setup.
For paid social and search, watch frequency and CPM trends. These are leading indicators: rising frequency or CPM shows up weeks before it drags your ROAS down, giving you a chance to adjust creative or audience before the damage hits revenue.
Apply one filter before adding anything: does this number change a decision this week, or is it just interesting to look at? If it's the second one, it belongs in a report, not the main dashboard.
Cap the primary view at 8 to 12 metrics. That's it. Everything else can live one click away as a drill-down, available when you need it but not fighting for attention every time you open the dashboard.
Split metrics by cadence too. Spend and conversion rate are daily-check numbers. LTV and inventory turnover move slower and belong in a weekly or monthly review instead. Checking a slow-moving metric daily just trains you to ignore it.
And revisit the list quarterly. What matters at $1M a year in revenue isn't what matters at $10M. A brand scaling past seven figures starts caring a lot more about fulfillment cost per order and stockout rate than it did when the whole operation fit in a garage.
Knowing which 25 numbers matter is the easy part. The hard part is pulling net revenue from Shopify, ad spend from Meta and Google, conversion data from GA4, and sales velocity from Amazon into one consistent view, every week, without three hours of copy-pasting into a spreadsheet.
That's the gap Trivas closes. It pulls these metrics automatically from your connected platforms into dashboards built on Redshift, so the weekly reporting pull goes from hours to minutes, and the numbers stay consistent across channels instead of each platform reporting its own flattering version of events.
If you want to see what this list looks like populated with your own store's real data instead of a mockup, start a trial and find out.
explore more insights
3 min read
3 min read
3 min read