This guide covers the full launch process: pre-launch checklist, campaign structure, keyword research for new campaigns, budget and bidding, ad copy that converts from day one, the first 30-day plan, and common launch mistakes.
1. Pre-Launch Checklist
Before you spend a dollar, make sure these are in place. Skipping any of them almost guarantees wasted spend in the first few weeks.
- Conversion tracking is verified. Run a test conversion through your entire funnel. Check that the conversion shows up in Google Ads within 24 hours. Verify the value is correct. Double-check there is only one primary conversion action per conversion type (one purchase action, one lead action).
- Landing pages are ready. The page you send traffic to should load in under 3 seconds on mobile. It should have a clear call to action above the fold. And it should match what your ad promises. Sending people to your homepage when your ad talks about a specific product is a common waste of money.
- Negative keyword lists are prepped. Build a starter negative keyword list before launch. Include obvious non-buyers: "free," "jobs," "salary," "DIY," "tutorial," and any terms that attract researchers instead of purchasers. You will add more after launch, but starting with a baseline saves 10-15% of your budget in the first week.
- Google Merchant Center is approved (for Shopping). All products need to be active and approved. Feed errors will silently prevent your products from showing. Check the Diagnostics tab before launching.
2. Campaign Structure for New Launches
When launching a new campaign type, start simple. One of the biggest mistakes is building a complex multi-campaign structure before you have any data. You do not know which keywords, audiences, or products will work best yet. Let the data tell you where to segment.
For a new Search campaign, start with one campaign containing 3-5 tightly themed ad groups. Each ad group should have 10-20 closely related keywords. This is enough granularity for Google to learn what works without spreading your budget too thin.
For a new Shopping or Performance Max campaign, start with one campaign and 2-3 asset groups segmented by your top product categories. Do not create 15 asset groups on day one. You need each group to accumulate enough data to be useful, and spreading across too many groups delays learning.
The time to add complexity is after week 4, when you have real conversion data. At that point, you can split out high-performing segments into their own campaigns with dedicated budgets and tROAS targets. See our scaling Google Ads guide for the full structure at each spend level.
3. Keyword Research That Saves Money
Most keyword lists for new campaigns are too broad. You have 200 keywords, half of which are informational queries that will never convert. Here is how to build a list that does not waste your first month of budget.
Start with high-intent keywords only. These are terms where the searcher is close to buying:
- Product-specific terms: "buy [product name]," "[product name] price," "[product name] sale"
- Category + qualifier: "best [category] for [use case]," "[category] under $100"
- Competitor alternatives: "[competitor name] alternative," "[competitor] vs [your brand]"
Avoid these on a new campaign: broad informational queries ("what is [category]"), how-to terms, and single-word category keywords. These have volume, but they do not convert at a rate that will produce useful learning data in your first month.
Use exact match and phrase match for the first 2-4 weeks. Broad match is great for discovery once your campaign has conversion data, but on a new campaign it lets Google experiment too widely with your limited budget. Start tight, then broaden as data comes in.
4. Budget and Bidding for New Campaigns
Your launch budget needs to be enough to generate 30+ conversions in the first month. That is the minimum for the algorithm to start making decent predictions. For most ecommerce stores, plan on $1,500-$3,000 per new campaign per month.
If that sounds like a lot, consider this: a campaign with 8 conversions in its first month has learned almost nothing. The data is too thin to distinguish real patterns from random noise. You will not know which keywords work, which ads resonate, or what time of day converts best. You have basically wasted that $500-$800 on inconclusive data.
For bidding, start with Maximize Conversions (no target). This tells Google to find conversions as efficiently as it can without constraining it to a specific CPA or ROAS target. Adding a target too early limits the algorithm's ability to explore and learn.
After 2-4 weeks, once you have 30-50 conversions, switch to Target ROAS or Target CPA. Set the target at 80% of your actual trailing performance. This gives the algorithm room to maintain volume while gradually improving efficiency.
5. Ad Copy That Converts From Day One
Your first set of ads needs to do double duty: attract clicks AND teach the algorithm what kind of clicks convert. Here is the practical approach:
- Write 3-4 responsive search ads per ad group. Each should have 10-15 headlines and 4 descriptions. Google will test combinations and learn which ones drive conversions.
- Include your primary keyword in at least 3 headlines. This directly impacts quality score, which affects your CPC from day one.
- Include price or price range in at least one headline. This pre-qualifies clicks. People who click after seeing the price are more likely to buy. People who would have bounced after seeing the price never click and you save that CPC.
- Use a clear call-to-action. "Shop Now," "Get 20% Off," "Free Shipping on Orders Over $50." Vague CTAs like "Learn More" attract browsers, not buyers.
Pin your strongest headline to position 1 and your strongest CTA to position 2. Let Google rotate the rest. This ensures every ad impression leads with your best messaging while still testing variations.
6. The First 30-Day Plan
Here is what to do and when during the first month of a new campaign:
Days 1-3: Check that ads are approved, keywords are showing impressions, and conversion tracking is firing correctly. Do not judge performance yet. Fix any disapprovals or tracking issues immediately.
Days 4-7: Review your search terms report for the first time. Add negative keywords for any clearly irrelevant queries. This is usually the biggest budget-saver in the first week. Check your quality scores. Anything below 5 needs ad copy or landing page attention.
Days 8-14: Review early conversion data. Which ad groups are spending but not converting? Which keywords have high CTR but no conversions? Pause the worst performers, but do not make drastic changes. The data set is still small.
Days 15-21: You should have 15-25 conversions by now if your budget is right. Start identifying patterns: which keywords convert, which audiences respond, which time slots perform. Make your first ad copy optimizations based on asset performance reports.
Days 22-30: Evaluate overall performance. If you are at or near your target ROAS/CPA, start planning to scale (see our scaling signals guide). If performance is significantly below target, diagnose the issue: is it the keywords, the landing page, the ad copy, or the audience?
7. Common Launch Mistakes That Burn Budget
- Launching with automated bidding AND a target. Target ROAS on a new campaign with zero conversion history means Google has no data to work with. It will either underspend (cannot find traffic that meets the target) or overspend wildly. Start with Maximize Conversions, add a target later.
- Too many campaigns at once. Launching 5 campaigns simultaneously with a $5K/month total budget means each campaign gets $1K. That is probably not enough for any of them to collect meaningful data. Launch 1-2 campaigns, prove they work, then expand.
- Not checking search terms in the first week. The first 48-72 hours of a new campaign are when you discover all the irrelevant queries you did not anticipate. Check daily for the first week and add negatives aggressively.
- Changing too much too fast. Every significant edit resets the learning phase. Batch your changes. Make optimizations once a week, not daily. The exception is adding negative keywords, which you should do as soon as you spot wasted queries.
- No baseline expectations. If you do not know your target CPA or break-even ROAS before launch, you can not evaluate whether the campaign is working. Set benchmarks before spending. If this is your first campaign on Google Ads, use industry averages as a starting point and adjust as you collect data.
Frequently Asked Questions
Enough to generate at least 30 conversions in the first month. For most ecommerce stores, that means $1,500-$3,000 per campaign. Starting with less means the algorithm does not get enough data to learn, and you can not distinguish real patterns from noise.
Most new campaigns need 2-4 weeks to exit the learning phase and stabilize. Performance in the first week is unreliable. Make your first real evaluation at day 14, and your scaling decision at day 30.
Start with Maximize Conversions (no target) for the first 2-4 weeks to collect data. Once you have 30-50 conversions, switch to Target ROAS or Target CPA set at 80% of your actual trailing performance. Manual CPC can work for small accounts where you want full control during the data collection phase.
Audit Your Account Before Launching New Campaigns
COREPPC's free audit checks your Google Ads and Meta accounts in 60 seconds. Find issues in your existing campaigns before adding new ones.
Start Free Audit
