Measuring campaigns: the numbers that really matter
Photo: Avila Ops Tecnologia
Six numbers show whether a campaign paid off: margin left after everything, cost per paid order, number of paid orders, average order value, return rate, and new customers acquired. Reach, impressions, likes, and website visits are steps along the way, not results. If final margin fell below an ordinary month’s, the campaign cost money, however attractive the reach chart looks.
The calculation that counts is the final report, prepared two weeks after Black Friday once refunds and returns are included. The campaign report guide covers the daily version; the final version determines next year’s decisions.
Which numbers mislead?
Reach and impressions: how many people saw the ad. Looking costs nothing. Clicks and visits: better, but still not money. Created orders: include unpaid boletos and expired Pix payments. Gross revenue: the number everyone celebrates without deducting discounts, fees, absorbed shipping, returns, and product costs.
A home decor store ended Black Friday with revenue 180% above October and posted about it. Two weeks later, after returns and ad costs, final margin was 15% below October. The celebrated number was real. It just was not the one that mattered.
The six numbers and how to calculate each
| Number | Calculation | What it answers |
|---|---|---|
| Final margin | Net revenue minus product costs, fees, absorbed shipping, advertising, and returns | Was more or less left than in an ordinary month? |
| Cost per paid order | Advertising plus discounts given, divided by paid orders | What did each sale cost? |
| Paid orders | Count excluding canceled and refunded orders | How many real sales? |
| Average order value | Net revenue divided by paid orders | Did each customer buy more or less? |
| Return rate | Returned or refunded orders divided by all orders that were paid, including those | Did discounts attract people who did not want the product? |
| New customers | Buyers who had never purchased before | Did the campaign bring new people or merely bring purchases forward? |
Net revenue is after discounts and before fees. Product cost is what you paid for goods sold, not everything purchased; stock remaining is a separate number, covered in the product margin guide.
Compare each row with October, not “market averages” mixing electronics stores with soap shops. The question is whether your campaign earned more than your store earns without a campaign.
What does each combination tell you?
Final margin above October and many new customers: the campaign worked; repeat it. Lower margin and many new customers: you bought customers; worthwhile if you have a repurchase plan. Lower margin and few new customers: you discounted purchases people would make anyway; next year, target only cold audiences or skip it. Returns far above an ordinary month: discounts or photos attracted people who did not want the product.
Cost per order above the margin on each order, calculated at full price and including its share of fixed costs (above R$ 45 for a R$ 150 full-price order with 30% margin): each order cost more than it earned. Small stores often discover this too late because ad managers show “cost per purchase” without including discounts.
What should you do with the result?
Write half a page with the six numbers, the October comparison, and one decision for next year: repeat unchanged, change the featured product, cut discounts, cut advertising, or skip the campaign. Store that page somewhere you will find it next September.
Many wrong Black Friday decisions come from memory: “we sold a lot last year.” Selling a lot is revenue. The half-page reminds you what remained.
What to do now
Mark a date two weeks after Black Friday to finalize the six numbers. Until then, record every return and refund with its order number so calculations do not rely on memory. To see all six ready on one screen, combining store, payment, and advertising data, Avila Ops builds the dashboard: avilaops.com/sistema-para-pequenas-empresas or contact us on WhatsApp: wa.me/5517997811471.
Frequently asked questions
How do I measure new customers without CRM?
Compare campaign buyers with previous buyers in a spreadsheet using email or phone. Anyone absent from the second list is new. Laborious, but possible.
Does the cost of my time enter the calculation?
Not margin, but it enters the decision. If the campaign earned R$ 1,500 more than October and consumed 60 hours of your time, each hour earned R$ 25. Compare that with what those hours earn during the rest of the year.
Should I measure advertising separately from organic traffic?
Yes, if possible. Ad orders, organic Instagram orders, and returning-customer orders have different costs. Without separation, ads receive credit for sales that would have happened anyway.
Turn what you learned into your next step.
Tell us about your company and what you want to improve. We can explore the right tools, integrations or custom project together.