Marketing intelligence
Salla Store Analytics: The Weekly KPIs That Show What Your Ads Are Doing
Ten store-side numbers to review every week, where each one comes from in Salla, and how to read them next to your ad spend.
The best way to see what your ads are doing on a Salla store is to watch store-side numbers, not just platform dashboards. Each week, review net revenue, orders, average order value, new versus returning customers, cancellation rate, abandoned-cart recovery and blended ROAS.
Compare each one with the previous week and the same week last month. Ad platforms tell you what they claim, and your Salla orders tell you what actually happened.
Why store-side KPIs come first
Every ad platform reports its own conversions, and those numbers overlap. A shopper who saw a Snapchat ad, clicked a TikTok ad and searched on Google can appear as a purchase in all three. Salla records that order once.
Starting from the store means you begin with a number that is true, then use platform data to explain it.
The 10 weekly KPIs
| # | KPI | How to calculate it | Salla data you need | What it tells you about ads |
|---|---|---|---|---|
| 1 | Net revenue | Order value minus cancellations and refunds, with VAT treated consistently | Orders, order statuses | The business result every channel is judged against |
| 2 | Orders | Count of orders in a counted status | Orders, order statuses | Volume, separate from price changes |
| 3 | Average order value (AOV) | Net revenue ÷ orders | Orders | Whether discounts or bundles are changing basket size |
| 4 | New customers | Customers whose first order falls in the week | Customers, orders | Whether prospecting campaigns are reaching new people |
| 5 | Returning-customer revenue share | Revenue from repeat customers ÷ net revenue | Customers, orders | How much of the week came from people you already had |
| 6 | Cancellation rate | Cancelled orders ÷ all orders placed | Order statuses, order histories | Low-quality traffic or COD refusals hiding behind "sales" |
| 7 | Abandoned-cart recovery | Recovered carts ÷ abandoned carts | Abandoned carts | Whether retargeting and reminders are working |
| 8 | Coupon share | Orders with a coupon ÷ orders | Coupons, orders | How dependent sales are on discounts and influencer codes |
| 9 | Top products by net revenue | Rank products by net revenue | Order items, products | Whether ads are pushing the products you intended |
| 10 | Blended ROAS | Net revenue ÷ total ad spend | Orders + spend from each ad platform | Overall efficiency, without double counting |
The data groups in the fourth column match sections of the Salla Merchant API, which covers orders, order statuses, order histories, customers, abandoned carts and coupons. You can get most of these from Salla's dashboard reports too. The API, or a connector built on it, is what makes the review repeatable.
Define "counted" orders once
The most common mistake is letting each person decide which orders count. Write one rule and apply it to every KPI:
- Counted: orders that are paid, or delivered for cash on delivery.
- Excluded: cancelled, refused and fully refunded orders.
- Pending: orders still awaiting payment or delivery. Report these separately and don't add them to revenue until they settle.
If your team reports in SAR and includes VAT, say so on the report. Finance usually reports excluding VAT, and mixing the two is a quiet source of disagreement.
A 30-minute weekly review
- Check the store total first. Compare net revenue and orders with last week and with the same week last month.
- Split the change. Did revenue move because of more orders, or because of higher or lower AOV? Those two have different causes.
- Check quality. If orders rose but the cancellation rate also rose, part of the increase isn't real yet.
- Bring in spend. Calculate blended ROAS. If spend went up 30% and net revenue went up 5%, the extra spend mostly bought orders you would have had anyway, or didn't work.
- Only then open the platform dashboards, to find which campaigns explain the change.
An illustrative week
The numbers below are invented to show the method.
| KPI | Last week | This week | Change |
|---|---|---|---|
| Net revenue (SAR) | 84,000 | 91,500 | +8.9% |
| Orders | 420 | 488 | +16.2% |
| AOV (SAR) | 200 | 187.5 | −6.3% |
| Cancellation rate | 6% | 11% | +5 points |
| Coupon share | 22% | 41% | +19 points |
| Total ad spend (SAR) | 18,000 | 21,000 | +16.7% |
| Blended ROAS | 4.67 | 4.36 | −0.31 |
Read together, these say the extra orders came mainly from a discount code: coupon share almost doubled and AOV fell. Cancellations also rose, which often happens when a code spreads beyond its intended audience.
Platform dashboards might show this as a great week. The store data says efficiency actually fell, and part of the revenue may still disappear through cancellations.
Common pitfalls
- Counting pending COD orders as revenue. They inflate the week and disappear later.
- Mixing order date and delivery date. Pick one date basis for each KPI and keep it.
- Comparing a sale week with a normal week. Compare with the same type of week, or note the promotion clearly.
- Joining platform conversions to Salla orders without IDs. Without a shared order or click identifier, compare totals and trends, not individual orders.
Making it repeatable
You can rebuild this table by hand each week, but most teams stop after a month. Two practical options:
- Spreadsheet or Looker Studio: our guide to connecting Salla to Looker Studio and Google Sheets shows the setup.
- Warehouse: for larger stores, connecting Salla to BigQuery gives you full control over order statuses and history.
AKM's Salla connector tracks sales, orders, product performance and customer insights alongside your ad accounts. That makes blended ROAS a single number instead of a monthly spreadsheet task.
If White Friday is coming up, use this KPI set as the baseline in your White Friday measurement setup. For how to define the ratio itself, see how to track blended ROAS and MER.