Reading the Data (CTR, CPC, CPM, CPA, ROAS)

10 min · interactive · Intermediate

Your dashboard is screaming at you with 47 metrics. 4 of them matter. 43 are noise. Today we draw the line - and show you the exact Majorka dashboard that surfaces the ones that matter.

The metric hierarchy: what to measure

Dropshipping metrics stack in order of relevance to profit:

ROAS (Return on Ad Spend) - total revenue ÷ ad spend. The only metric that is directly tied to profit margin.
CPA (Cost Per Acquisition) - total ad spend ÷ total purchases. The second-most important; if CPA exceeds your contribution margin, you cannot scale.
CPC (Cost Per Click) - total ad spend ÷ total clicks. A diagnostic only; rising CPC + stable ROAS is noise; rising CPC + falling ROAS is a signal to pause and rotate creative.
CTR (Click-Through Rate) - clicks ÷ impressions. Useful only as a creative fatigue detector (CTR dropping 40%+ = creative exhausted, rotate).

Everything else is supporting narrative, not decision drivers:

CPM (Cost Per Thousand Impressions) - only watch for audience fatigue signals (CPM rising 20%+ = time to rotate creative)
Conversions - just the numerator of CPA; useless without context
Impressions - only useful divided into CTR
Frequency (avg times user sees ad) - >4 frequency with rising CPM = fatigue signal, not a metric to optimize

Reading ROAS in real time

ROAS in the testing phase (days 1–14) is noisy. Meta pixels take 24–48 hours to attribute conversions. Your day-3 ROAS is usually wrong by 30%.

How to interpret daily ROAS by day:

Day	ROAS reading	Action
Day 1–2	Ignore (pixel learning, <5 conversions)	Do not pause
Day 3–7	Watch but do not act (Meta learning phase)	Do not pause
Day 7–14	First reliable signal (20+ conversions accumulated)	Pause if ROAS < 1.0; scale if > 1.5
Day 14–30	True signal (100+ conversions)	Scale if > 1.5; kill if < 1.0
Day 30+	Stabilized (trust it)	Optimize by creative rotation

A campaign that shows 0.9 ROAS on day 5 might land 1.8 ROAS by day 14. Do not kill on day 5.

CPA ceiling: when to kill

Your contribution margin (from m1-l1) is your CPA ceiling. If your margin is $9 per unit, you cannot afford a 2 CPA in the long run (loses $3/order).

Realistic kill thresholds (7-day rolling average):

Product type	Contribution margin	CPA kill threshold
Impulse gadget (kitchen, pet)	$8–12	0–13
Higher-ticket ($50+ retail)	5–22	8–25
Apparel / soft goods	$6–10	$7–11

If your 7-day average CPA exceeds this threshold, kill the campaign (or pause creative and rotate). Scaling a campaign with CPA > margin is just accelerating a loss.

Diagnosing what is actually wrong

Rising CPA + stable CTR = algorithm learning, not creative exhaustion. Wait.

Rising CPA + falling CTR = creative fatigue. Rotate immediately.

Rising CPM + stable CPA = broad audience expansion (Advantage+ finding new user segments). Often a good sign in week 2–3. Monitor for 5 more days.

Rising CPM + rising CPA = audience saturation. Rotate creative or pause campaign.

Stable metrics but ROAS < 1.5 = product is wrong, not ads. New product test, do not scale this one.

The Majorka revenue dashboard integration

Majorka's Revenue Dashboard surfaces ROAS, CPA, and profit-per-order in real time, pulling from your Shopify or store pixels. Instead of hunting Meta Ads Manager for the raw numbers, the dashboard tells you:

Which products hit CPA ceiling today (kill candidate)
Which campaigns are scaling and ready for creative refresh (scale signal)
Which products have dipped below ROAS 1.5 for 5+ days (pause signal)

The dashboard also surfaces kill/scale decision in plain language: "This product has 60 days of data, ROAS 2.3, CPA $8.50 below margin threshold. Ready to scale to $500/day."

Why this matters

Operators who trust the wrong metrics overscale losers and kill winners by accident. The four metrics above (ROAS, CPA, CPC as diagnostic, CTR as fatigue detector) are sufficient to run a profitable business. Everything else is window dressing.