Thousands of people saw the ad. Did it work? That number alone can’t tell us. If the goal is purchases, we need to know what happened after the views and clicks, and what it cost to get the result. That’s why planning your first ad campaign starts with defining success before opening a report.
Choose a clear action you can track
“We want awareness, sales and messages” combines different outcomes. Pick one primary action for the test: an enquiry submitted, a booking completed or a purchase completed. Call it what it is. Someone who leaves a phone number is a lead, not necessarily a paying customer.
Preparing measurement means checking that the report records the action you actually care about. Tools such as a Pixel help record website events for ad measurement, but need appropriate setup and checks, while respecting privacy choices and applicable requirements. Before getting technical, answer one question: which event do I need to measure?
Understand the first numbers you’ll read
Impressions count ad displays; clicks count clicks on the ad. Click-through rate, or CTR, is clicks divided by impressions, multiplied by 100. Average cost per click, or CPC, is ad spend divided by clicks. Google Ads explains these definitions in its help documentation. [1] [2]
CPL means cost per lead. CPA means cost per defined action. State whether that action is a purchase, booking or something else, so you don’t compare different outcomes under the same label.
A worked example you can check yourself
These figures are hypothetical, not a recommended budget or expected result. A shop spends $100 and records 10,000 impressions, 200 clicks, 10 enquiries and two sales generating $160 in revenue. Assume the figures cover the same period and the sales are attributed to the campaign under the measurement setup in use.
| Metric | Calculation | Result |
|---|---|---|
| CTR | 200 ÷ 10,000 × 100 | 2% |
| Average CPC | 100 ÷ 200 | $0.50 |
| CPL | 100 ÷ 10 | $10 |
| Purchase CPA | 100 ÷ 2 | $50 |
| ROAS | 160 ÷ 100 | 1.6× |
ROAS is revenue attributed to advertising divided by ad spend. A value of 1.6 here doesn’t mean a $60 profit: we haven’t deducted products, shipping, fees, returns or other costs. Similarly, 200 clicks don’t necessarily mean 200 unique visitors or 200 completed page loads.
How should you set a first-test budget?
Start with an amount you can afford to test, a review period and a stopping rule written in advance. This guide doesn’t prescribe one budget for every market or product. Check the minimums and spending conditions shown in your account before launch.
Pay attention to what “daily budget” means. For example, some Google Ads campaigns use an average daily budget, and actual daily spend can vary under the platform’s rules. Don’t automatically treat that figure as a hard cash limit for every day. [3]
What should you do when results arrive?
Check the data first, then the journey. Do people reach the page? Can they understand the offer and price? Are the enquiries valid? If you change the creative, audience, page and budget at once, it becomes harder to understand what made the difference.
Your exercise: copy the table, calculate it using a different set of hypothetical figures, then write what you would investigate if clicks produced no enquiries. Don’t automatically choose “increase the budget”. Aim first for a report you understand and a test you can explain. That gives you a basis for decisions beyond the appeal of a big number.
Official sources
Documentation reviewed: 16 October 2026. Interfaces and availability vary by account, market and updates. Examples and exercises are hypothetical, not campaign results.
