Updated October 2026 · By Farrukh Moin Ahmed, digital marketing expert
Short answer: Run Google Ads for a new online store in three stages. Days 1 to 15: Merchant Center, conversion tracking and account structure, with no spend. Days 15 to 45: a search-ready feed, then Standard Shopping and narrow Performance Max. Days 45 to 90: high-intent Search, remarketing and prospecting, each with its own budget and success measure.
Key takeaways
- Work out gross margin, order value and affordable cost per purchase before spending.
- Foundations first: Merchant Center and tested purchase tracking before any campaign.
- Launch narrow with Shopping and focused Performance Max, then add Search, remarketing and prospecting.
- Woda Bags reached an 8.12 ROAS and SofaBay a 23.83 ROAS after launching this way.
Why does the launch order matter for a new store?
A new store’s first Google Ads campaigns are built on almost no data. There’s no conversion history, no proof of which products sell through ads, and no record of which searches turn into orders. Google’s bidding systems learn from conversions, so the first few weeks decide what they learn.
Launch in the wrong order and the account learns the wrong lessons. Campaigns running on broken tracking optimise toward the wrong events. Shopping campaigns built on a weak feed show products for the wrong searches. Broad campaigns launched before the basics are proven spend budget finding out what a focused campaign would have shown cheaply.
This is the sequence I use. It’s the same order of work behind Woda Bags and SofaBay, both launched from scratch. The day ranges are a planning guide, not a promise: accounts with more traffic and conversions move through them faster, smaller ones more slowly.
What should you decide before spending anything?
Three numbers shape every decision in the first 90 days, so work them out first:
- Gross margin. What’s left of an average order after product cost. This sets your breakeven ROAS: 1 divided by your gross margin. At a 40% margin you need a 2.5 ROAS just to cover product costs.
- Average order value. Higher order values can afford more per click, which changes which searches are worth bidding on.
- The cost per purchase you can afford. Your margin per order, minus anything else that has to come out of it. This is the number every campaign is held against.
Write these down. When a campaign reports a 3.0 ROAS, they tell you whether that’s a success or a slow loss.
Days 1 to 15: what foundations come first?
Nothing launches until the foundations are in place:
- Merchant Center. Verified and claimed, with shipping and returns matching the store, linked to Google Ads, and with free listings turned on.
- Conversion tracking. Purchases tracked with real order values, tested end to end. Add to cart and begin checkout can be tracked too, but as secondary actions so bidding optimises for purchases.
- GA4 and Google Tag Manager. Set up so you can see the full journey, not just what Google Ads reports. My analytics and tracking work covers this step.
- Account structure on paper. Campaigns planned around product lines and margins, so you know where every product will live before building anything.
Skipping this stage is the most expensive mistake a new account can make. Every week of spend on broken tracking is a week of the bidding system learning from bad data.
Days 15 to 45: how do you launch Shopping and Performance Max?
In Shopping and Performance Max, Google reads your product feed to decide which searches you appear for. So the feed comes first: titles that lead with product type and searched attributes, complete identifiers, accurate prices and stock, and custom labels for margin and best sellers. The detail is in Google Shopping feed optimization.
Then launch narrow:
- Standard Shopping for control and clear reporting on which products and searches convert.
- Performance Max with a narrow brief: one product line or margin band per campaign, not the whole catalogue. Why that matters is in Performance Max for ecommerce.
- Brand search, if people already search your brand name, so competitors can’t take those clicks. Reported separately from everything else.
During this stage, check search terms every week. They show which words to add to product titles, which irrelevant searches to exclude, and which products attract buyers rather than browsers.
Days 45 to 90: when do you add Search, prospecting and remarketing?
Once Shopping is producing orders and the tracking is trusted, widen the account. I add campaigns in this order:
- High-intent Search. Product, category and attribute searches that Shopping data has shown to convert, protected by negative keywords so broad matching doesn’t drift.
- Remarketing. Bringing back visitors who didn’t buy, with windows that match how long your customers take to decide. Days for impulse purchases, weeks for furniture.
- Prospecting. YouTube, Display or Demand Gen to reach people not yet searching. Judged on new customers and assisted sales, not last-click return.
Each new campaign gets its own budget and its own success measure, so the account never becomes one blended number nobody can act on.
What should you measure at each stage?
| Stage | Main question | What to measure |
|---|---|---|
| Days 1 to 15 | Is the data right? | Test purchases recorded with correct values; Merchant Center products approved |
| Days 15 to 45 | Which products and searches sell? | Conversion value by product; search terms; cost per purchase against breakeven |
| Days 45 to 90 | Where can we grow profitably? | Non-brand conversion value; new vs returning customers; remarketing frequency |
| Always | Does it match the store? | Google-reported orders compared with actual store orders |
How should you set the starting budget?
There’s no universal starting number, and anyone quoting one without knowing your margins is guessing. I set budgets from two constraints:
- The learning constraint. Each campaign needs enough conversions to learn from. Splitting a small budget across many campaigns leaves each one guessing, so new accounts start with fewer, focused campaigns.
- The profit constraint. Spend should stay within what your breakeven cost per purchase allows, with room to test. Scaling comes from campaigns that prove profitable, not from raising budgets and hoping.
A practical rule: start with as few campaigns as the catalogue allows, give them enough budget to generate conversions, and only split further when each new campaign could stand on its own data.
What are the most common launch mistakes?
| Mistake | Why it hurts | Instead |
|---|---|---|
| Launching before tracking is tested | Bidding learns from wrong or missing purchases | Test purchases end to end before any spend |
| One Performance Max for the whole catalogue | No visibility into what sells; best sellers take everything | Narrow campaigns by product line or margin |
| Store product names as feed titles | Products miss the searches people actually make | Lead titles with product type and attributes |
| Broad Search with no negatives | Spend leaks to irrelevant searches | Start with proven queries, review search terms weekly |
| Judging prospecting on last click | Demand creation looks unprofitable and gets cut | Judge it on new customers and assisted sales |
| Too many campaigns too early | Every campaign is starved of data | Split only when each part can learn on its own |
What did this look like for Woda Bags and SofaBay?
Woda Bags is a US bag brand that had been running Meta Ads but hadn’t established Google Ads as a performance channel. The launch followed the order above: Merchant Center setup, Google Ads account structure, conversion tracking, product feed optimization with SEO-focused product titles and descriptions, then Shopping and Performance Max. The account reached 395 orders and $76,211 in conversion value at an 8.12 ROAS.
SofaBay, a UK furniture retailer, had no Google Ads foundation. It got the same foundations first, then high-intent Search and Performance Max alongside prospecting and remarketing. It reached a 23.83 ROAS and £218K in conversion value. Furniture’s long buying cycle is why remarketing mattered so much there, which the furniture and home page covers in more detail.
A new account doesn’t need a long warm-up if it’s built properly from day one.
How do you know when a campaign is ready to scale?
Raising budget on a campaign that isn’t ready just buys more of the same problems at a higher price. Before scaling anything in a new account, I want to see four things hold at the same time:
- Cost per purchase sits below breakeven for several weeks, not just on a good day. Short spikes often come from a handful of large orders.
- Results hold up against store revenue. Google-reported orders match the orders in the store closely enough to trust.
- Non-brand campaigns are profitable on their own. If profit only appears once brand searches are included, the account isn’t finding new customers yet.
- Search terms are clean. The searches driving spend are ones you’d choose to pay for.
When those hold, scale gradually. Large, sudden budget jumps can disrupt bidding systems while they adjust, so I increase in steps and watch cost per purchase at each one.
A 90-day launch checklist
- Gross margin, average order value and breakeven ROAS written down
- Merchant Center verified, claimed and linked to Google Ads
- Shipping and returns in Merchant Center match the store
- Purchase tracking tested end to end with real order values
- GA4 and Google Tag Manager in place
- Feed titles lead with product type and searched attributes
- Identifiers and custom labels added to the feed
- Standard Shopping and narrow Performance Max launched
- Search terms reviewed weekly, negatives added
- High-intent Search built from converting queries
- Remarketing windows matched to the buying cycle
- Prospecting judged on new customers and assisted sales
- Google-reported orders reconciled with store orders every month
What comes after the first 90 days?
By the end of the first 90 days, the account should have trusted tracking, a feed that keeps improving from search term data, and a clear view of which products and campaigns are profitable. That’s when real scaling decisions become possible: more budget for proven product lines, new markets, and adding Meta Ads to create demand that Google can then capture.
If you’re planning that next step, Google Ads vs Meta Ads for ecommerce explains how the two channels divide the work. For the full approach, see Google Ads for ecommerce, or tell me about your store.
If you are launching and want the first 90 days run for you, I work as a Google Ads expert for ecommerce, including for Shopify stores.
Frequently asked questions
How long does it take for Google Ads to work for a new store?
Fixes and foundations take the first couple of weeks. Shopping and Performance Max then need conversion data to learn, so judge new campaigns over weeks rather than days. Accounts with more traffic learn faster.
What should a new online store set up before running Google Ads?
Merchant Center (verified, with accurate shipping and returns), conversion tracking with real purchase values, GA4 and Google Tag Manager, and a search-ready product feed.
Should a new store start with Shopping or Search?
Usually Shopping and narrow Performance Max first, plus brand search if people already search your name. High-intent Search comes next, built on the queries Shopping shows convert.
How do I know what ROAS my store needs?
Divide 1 by your gross margin. At a 40% margin, a 2.5 ROAS covers product costs. Anything below that loses money before other costs.
When should a new store add Meta Ads?
Once Google is producing profitable orders and tracking is trusted, Meta can create demand that Google captures. Stores with products people need to discover may start Meta earlier.
