The KPIs an Executive Marketing Dashboard Should Show

An executive does not need fifty numbers. An executive needs to know, in a few minutes, whether to spend more, spend less or investigate. A dashboard that cannot support one of those three decisions is decoration.

This article gives you a way to pick the KPIs: start from the decision, then choose the smallest set of numbers that supports it.

Why most dashboards fail executives

They are built from what the platforms offer, not from what the owner decides. The result is a wall of impressions, clicks, reach and platform-reported ROAS. Each is accurate on its own terms and none answers "are we making money after ads?".

Two more faults are common. The numbers come from different sources with different definitions, so they never reconcile. And nothing shows how fresh or complete the data is, so a stale tile looks as trustworthy as a live one.

A framework: KPIs grouped by decision

Group the dashboard into four questions.

• Are we getting orders?

  Ad spend, incoming orders, cost per incoming order.

• Do the orders stay?

  Share of orders delivered or kept, and cost per kept order.

• Are we profitable?

  Net revenue on kept orders, contribution after ads, blended return.

• Can we trust this?

  Data freshness and a reconciliation flag against your store.

1. Volume
2. Quality
3. Profit
4. Trust

The eight KPIs follow from these groups.

1. 1

   1. Ad spend

   All platforms combined, same period

   Volume
2. 2

   2. Incoming orders

   From your store, not the ad platform

   Volume
3. 3

   3. Cost per incoming order

   Spend / incoming orders

   Volume
4. 4

   4. Kept-order rate

   Delivered or kept / incoming

   Quality
5. 5

   5. Cost per kept order

   Spend / kept orders

   Quality
6. 6

   6. Net revenue

   Revenue on kept orders only

   Profit
7. 7

   7. Contribution after ads

   Margin on kept orders minus spend

   Profit
8. 8

   8. Data freshness

   Last update and match with the store

   Trust

Blended return (net revenue divided by total spend) can sit beside KPI 6. Keep one version only, and say which.

Illustrative example, not real results

Assume 400 incoming orders at an average of 300 SAR, 300 of them kept, a 40% margin on kept revenue and 30,000 SAR of ad spend. Every figure is invented.

Revenue

Incoming view: 120,000

Kept view: 90,000

Return on spend

Incoming view: 4.0

Kept view: 3.0

Cost per order

Incoming view: 75

Kept view: 100

Contribution after ads

Incoming view:

Kept view: 6,000

The arithmetic: 400 x 300 = 120,000; 300 x 300 = 90,000; 120,000 / 30,000 = 4.0 and 90,000 / 30,000 = 3.0; 30,000 / 400 = 75 and 30,000 / 300 = 100. Contribution is 40% of 90,000 = 36,000, minus 30,000 of spend, which leaves 6,000. The incoming view looks comfortable. The kept view shows a thin margin. An executive needs the second one. The method is in Contribution Margin per Order.

Read the example as a method, not a benchmark. The point is that the same month can look comfortable or thin depending on which orders you count, and the dashboard should make the harder view the default.

Trade-offs and when this changes

• Not a COD business. If almost all orders are prepaid, the gap between incoming and kept orders is smaller. Returns and cancellations still matter.
• Long sales cycles. For lead-based businesses replace "kept orders" with qualified leads or closed deals, and accept a lag.
• Subscriptions. Add retention or repeat purchase; one order is not the whole value.
• Too many tiles. If the screen needs scrolling, remove KPIs. Move diagnostics to a second page.

Implementation steps

1. Write the definition of each KPI in one line and keep it on the dashboard.
2. Pull orders and delivery status from your store or order system, not from ad platforms.
3. Add costs: product, shipping, failed parcels and fees.
4. Show every KPI against the previous period, not alone.
5. Add a freshness stamp and a reconciliation check against the store's total.
6. Review the set quarterly and remove what no one used to decide.

If you track conversions in GA4, note that Google distinguishes key events in Analytics from conversions used in Google Ads; check the current help pages before mixing the two in one number.

Measurement limits

Platform-reported conversions overlap and attribution windows differ, so adding platform figures can count one order twice. Margin assumptions are estimates until the books close. And a dashboard shows what happened, not why; pair it with questions such as those in asking plain-language questions of your data.

Next step

An Executive Marketing Dashboard puts these KPIs on one screen from your own data. To build the inputs first, see How to Build a Dashboard That Connects Ads, Sales and Profitability and Automating a Weekly Ads Report.

Sources

• Conversions vs. key events in Google Analytics (Analytics Help)