Skip to content

P&L dashboard for shop owners: 8 metrics to track weekly

Eight metrics that tell a multi-channel seller how much real profit they make, where, and whether to trust the numbers, with a 5-minute read.

By The DataxLand team· · 5 min read

Sales software and marketplace dashboards give you dozens of charts: views, conversion rate, revenue by hour, top products. Very few of them answer the question a shop owner actually needs answered: how much profit did I make this week, where did I make it, where did I lose it, and can I trust this number?

Guides to executive dashboards usually recommend keeping around 6–8 KPIs, each with an owner and an action threshold. This post proposes exactly eight metrics for a multi-channel seller, in three groups, with formulas and how to read them. These eight metrics are also the core of the "P&L overview" page in DataxLand.

Illustrative numbers come from the DataxLand sample data: a simulated scented-candle shop selling on Shopee, TikTok Shop and its own website, 1 Aug – 29 Oct 2026.

Group 1 — Money: how much real profit

1. Net revenue

Net revenue = GMV − discounts − allocated order-level vouchers − refunds

This is the revenue you actually keep before costs. Put it in the first tile, instead of GMV. Sample data: GMV 831,421,000 ₫, net revenue 755,554,000 ₫. If your dashboard only shows GMV, you are looking at a figure about 10% too high.

2. Contribution margin

Contribution margin = Net revenue − COGS − marketplace fees − payment fees − shipping fees − return fees

What is left to pay for ads, salaries, rent and profit. Sample data: 224,371,200 ₫ (29.7% of net revenue). This metric tells you whether your sales model is healthy, independently of how much you spend on ads.

3. Profit after fees and ads

Profit after fees = Contribution margin − allocated ad spend

The "how much did I make this week" number. Sample data: 89,301,200 ₫. Read it against the previous period.

4. Profit margin (%)

Profit margin = Profit after fees ÷ Net revenue

Sample data: 11.8%. Profit margin matters more than absolute profit when comparing weeks of different scale, or when you run a sale: if revenue doubles and profit margin halves, profit stays the same and you just have more work.

Group 2 — Where you make money, where you lose it

5. Profit after fees by channel

Metric 3, split by channel. This is the most valuable chart on the dashboard because it often turns your gut feeling upside down:

ChannelNet revenueProfit after fees
Shopee397,961,000 ₫89,927,600 ₫
TikTok Shop228,021,000 ₫−13,802,100 ₫
Website129,572,000 ₫13,175,700 ₫

The second-largest channel by revenue is losing money. For why, see How to calculate true profit on three marketplaces.

6. True ROAS by channel (and the gap with platform ROAS)

True ROAS = Net revenue ÷ ad spend (same channel, same period)

Compare it with the channel's break-even threshold 1 ÷ contribution margin %. Next to it, put the reported-vs-actual revenue gap metric: in the sample data, ads for the website report 92,828,000 ₫ more than reality. For setting thresholds, see What ROAS is profitable?

7. Return rate by channel

Return rate = Returned orders ÷ valid orders

Sample data: Shopee 2.7%, website 3.2%, TikTok Shop 12.4%. Return rate is a leading indicator: it rises before profit falls, because refunds and return fees arrive later. See The true cost of returns.

Group 3 — Can the numbers be trusted?

8. Data completeness: cost coverage and provisional fee rate

This is the metric most dashboards leave out, and the reason many shop owners do not trust their own dashboard.

Cost coverage         = Revenue of lines with a cost of goods ÷ total line revenue
Provisional fee rate  = Fees still provisional ÷ (marketplace fees + payment fees + shipping fees)

Sample data: 92.35% of revenue has a cost of goods — the remaining 7.65% is counted as if cost were zero, so profit is inflated; and 12.36% of fees are still provisional. These two numbers are not there to look good, but to tell you when you can draw conclusions: with cost coverage below roughly 95%, fill in costs before discussing profit by SKU; with a high provisional fee rate, do not close the most recent week's profit.

In DataxLand, every metric below 100% completeness shows an "Incomplete" label with the reason, and the Data health page lists which checks are failing (for example: order lines with a SKU not in the product catalog). See Verified numbers.

Eight metrics on one page

#MetricQuestion it answersThreshold to set yourself
1Net revenueHow much did we really sell?Versus previous period
2Contribution marginIs the sales model healthy?% of revenue, watch the trend
3Profit after feesHow much profit?Positive, versus previous period
4Profit marginDoes profit grow with scale?Your target profit margin
5Profit by channelWhere do we make or lose money?Any channel negative 2 weeks in a row
6True ROAS by channelDo ads pay for themselves?≥ 1 ÷ contribution margin %
7Return rate by channelDo orders stick?Below the break-even return rate
8Data completenessCan we trust the numbers?Cost coverage ≥ 95%, low provisional fees

Reading the dashboard in 5 minutes every Monday morning

  1. Tile 8 first. Is the data complete? If costs are missing or many fees are provisional, read the rest with a question mark.
  2. Tiles 3 and 4. Last week's profit and profit margin compared with the week before.
  3. Tile 5. Which channel moved most? Is any channel negative?
  4. Tiles 6 and 7 for that channel. Is it ads or returns?
  5. One action. Write down one thing for this week: switch off a campaign, fix the listing of a SKU with many returns, fill in missing costs.

If you would rather not open the dashboard, ask Si: "How much profit did we make last week, and which channel dragged it down?" — the answer uses exactly these metrics, with sources. Or get an automatic summary every Monday morning on Telegram (see Automated reports on Telegram).

What not to put on the front page

  • GMV on its own. Put it in a waterfall from GMV down to profit, not as the biggest tile.
  • Views and likes. Useful for content creators, not for profit decisions.
  • Twenty charts. Every chart you add makes the eight tiles above harder to see.
  • Numbers without definitions. If nobody can say what "profit" on the dashboard subtracts, it is a decorative number. Every metric in DataxLand has a definition, formula and version in Metrics and the semantic layer.

Getting started

You can build these eight metrics in Excel if you are willing to merge files every week. Or look at the retail finance dashboard template — DataxLand builds it from your files once you approve Si's mapping proposal — and start with sample data to try it before uploading real data. Business owners who do not sell on marketplaces can see solutions for finance teams.

Related posts

Verified dashboards from your own data, this afternoon

Free plan with no time limit · No card required · Sample workspace ready in minutes