Facebook Ads for Ecommerce: The Beginner’s Guide to Your First Campaign
Your first Facebook ad campaign for ecommerce should follow a specific structure. Use the wrong objective, the wrong budget, or the wrong audience and you’re paying for data that won’t help you. This guide covers the minimum viable setup that gives new advertisers reliable results.
Campaign Objective: Always Sales for Ecommerce
When you create a campaign, Meta asks for your objective. For ecommerce, select Sales (formerly Conversions). This tells Meta’s algorithm to find people most likely to purchase, not just click. Other objectives — Traffic, Reach, Engagement — optimise for different behaviours and will not drive efficient purchases. The single biggest mistake beginners make is using Traffic campaigns and wondering why clicks don’t convert.
Within the Sales objective, set the conversion event to Purchase at the ad set level. If your Pixel has fewer than 50 purchase events in the last 30 days, use Add to Cart or Initiate Checkout as a proxy event until purchase volume builds.
Campaign Structure for Beginners
Start simple: one campaign, one ad set, three to five ads. Don’t try to run multiple audiences simultaneously in week one. You need to generate purchase data before you can compare audiences.
- Campaign: Sales objective, Campaign Budget Optimization OFF for now, manual ad set budget
- Ad Set: Broad targeting (your country or region, age 18–65, no interest filters), 7-day click / 1-day view attribution window, Purchase as conversion event
- Ads: Three to five creative variations. At minimum: one video under 30 seconds, two static images. Test different hooks, not different products.
Why broad targeting? Meta’s algorithm is better at finding buyers than manual interest targeting in 2026. Broad targeting gives the system maximum flexibility to find purchasing intent. Narrow interest targeting restricts the algorithm’s options and typically results in higher CPMs and fewer conversions for most ecommerce categories.
Starting Budget: The Minimum That Works
The minimum daily budget that generates actionable data is €30–50/day. Below this, the ad set can’t exit the learning phase (which requires 50 purchase events) in a reasonable timeframe. Budget to spend €500–1,000 before making any significant changes. This covers the data cost to identify at least one working creative and confirm your cost-per-purchase economics.
If €1,000 for testing feels like too much risk, your margins or product pricing need review before scaling paid acquisition. Ads don’t fix unit economics — they amplify them.
Reading Your Results: What Matters in Week One
Don’t obsess over CTR in the first week. The metrics that matter for ecommerce: Cost Per Purchase (is it below your maximum CPA threshold?) and ROAS (is it above your break-even?). Everything else is secondary until you have 20+ purchases in the data.
Calculate your break-even ROAS before launching: if your product sells for €60 and your cost (COGS + shipping + overhead) is €35, you need a ROAS of at least 1.67 to break even on ad spend. Target 2.0+ to be profitable.
What NOT to Do in Your First Campaign
- Do not change the creative or audience within the first 7 days. Let the algorithm learn.
- Do not turn the campaign off and back on repeatedly. Each restart extends the learning phase.
- Do not add retargeting in week one. Focus 100% on prospecting until you understand your CPP baseline.
- Do not run ads without a Pixel installed and firing. You are paying for data that isn’t being recorded.
- Do not judge performance on days with under 5 purchases. Small samples produce meaningless variance.
Frequently Asked Questions
Use the Sales objective and set your conversion event to Purchase. This tells Meta’s algorithm to find people most likely to complete a purchase. Using Traffic or Engagement objectives for ecommerce optimises for clicks or interactions, not purchases, and results in poor ROAS.
Minimum $30-50/day to generate actionable data. Budget $500-1,000 for your initial test period before drawing conclusions. This covers enough data to identify at least one working creative and confirm your cost-per-purchase economics. Lower budgets extend the learning phase indefinitely.
Broad targeting (country/region, age 18-65, no interest filters) outperforms narrow interest targeting for most ecommerce categories in 2026. Meta’s algorithm is better at finding buyers than manual targeting. Broad targeting gives the system maximum flexibility to find purchasing intent.
Minimum 7 days before making decisions, and only after 20+ purchase events in the data. Small purchase samples produce meaningless ROAS variance. If you have fewer than 50 purchases per month, evaluate over 14-day windows instead of 7-day windows.
ROAS (Return on Ad Spend) is revenue divided by ad spend. Break-even ROAS = selling price divided by (selling price minus total cost). If you sell for $60 and your total cost is $35, break-even ROAS is 60/25 = 2.4. Target at least 20% above break-even to sustain profitability at scale.