Marketing
Google Ads for an Online Store – Where to Start and What to Watch in the First 90 Days
A practical plan for launching Google Ads for e-commerce: the technical foundation, the order to roll out campaigns, and what to watch in the first week, month and up to 90 days.

Article
Short answer
Start by getting the foundation right: your Google Ads account, a Google Merchant Center feed set up correctly, and conversion tracking in Google Analytics 4. Only on that basis should you launch product (Shopping) campaigns and search campaigns targeting high-intent purchase queries, adding Performance Max once you have historical data. In the first 90 days, the priority is collecting reliable conversion data and reviewing the search terms report regularly – without that, no optimisation makes sense.
The foundation, before you hit “launch”
Before any campaign starts spending budget, it's worth checking a few technical elements. Their absence is the most common reason campaigns “don't work”, even though the advertising settings themselves look correct.
- A Google Ads account linked to Google Merchant Center – without this connection, product campaigns and Performance Max with a product component simply won't launch.
- A product feed that meets Google's requirements – correct titles, categories, availability, prices matching what's on the site, and GTIN/MPN identifiers where required.
- Conversion tracking set up in Google Ads or imported from Google Analytics 4 – without it, the platform optimises campaigns “blind”, based only on clicks.
- Enhanced conversions – worth turning on from the start, as they improve the quality of the conversion data used by bidding algorithms.
- A domain and returns/shipping policy consistent with Google Merchant Center's rules – mismatches are a common reason for account disapproval or suspension.
Which campaign types make sense at the start
Not every ad format performs well from day one. The order in which you roll campaigns out matters, because some formats – especially those built on machine learning – need historical data to work effectively.
- Product (Shopping) campaigns – built directly on the Merchant Center feed, showing a product image, price and name. This is usually the natural starting point for an online store.
- Search campaigns – worth launching for queries with clear purchase intent, including your own brand name, so you don't hand that traffic to competitors.
- Performance Max – combines product, display, video and other formats into one campaign run by automation. It tends to perform better once the account already has conversion data – running it as the only campaign from day one makes it harder to see what's actually driving results.
- Remarketing campaigns – targeted at people who already visited the store, and worth setting up once there's enough traffic to build meaningful audience lists.
The first week
The first week is mostly about technical checks, not results – it's too early to judge performance.
- Check in Merchant Center whether products show as approved, not disapproved or pending.
- Verify that conversions are being recorded correctly – ideally through a test purchase or by checking events in GA4.
- Set a daily budget high enough to let the algorithm gather data, but not so high that it exposes the account to uncontrolled spend.
- Add your first exclusions – queries clearly unrelated to your offer, and locations outside your delivery area if the store doesn't ship there.
- Set up a reporting schedule so you have one consistent view of the data from the start, rather than only after a month.
The first month
How much data you have after two to three weeks depends on the budget and the number of clicks – for some accounts it is already material to analyse, for others still too little to draw conclusions from. This is the point for tidying up, not for radical strategy changes.
- Review the search terms report in search campaigns and add exclusions for queries generating clicks without conversions.
- Check the feed for low-visibility products – sometimes fixing a title or category is enough for a product to start showing more often.
- Start looking at ROAS (return on ad spend) broken down by campaign and product group, not just as a single aggregate figure for the whole account.
- Don't change the daily budget too often – significant changes to settings or bidding can push a campaign back into a learning phase; check the current rules in Google Ads Help, as they change along with the bidding system.
- Cross-check Google Ads data against actual product margins – a high ROAS on a low-margin product may not be profitable in practice.
Up to 90 days
After the first month, campaigns should have enough data to show what's working and what needs to change.
- Split product campaigns into product groups by profitability, and manage budget separately for the best- and worst-performing categories.
- Assess whether Performance Max is adding traffic on top of product and search campaigns, or mostly cannibalising traffic that would have shown up anyway.
- Check the quality of traffic from each campaign in terms of bounce rate and time on site – a low ROAS is sometimes a site issue rather than a campaign issue.
- Decide which campaigns to scale up with budget and which to cut back or pause – that decision should be based on data from at least several weeks, not a few days.
Common mistakes at the start
- Launching too many campaigns and ad types at once, making it impossible to tell which one is driving the result.
- Setting the daily budget too low, which prevents the algorithm from gathering the data it needs to optimise.
- Skipping enhanced conversions and relying only on standard tracking – in some cases it records fewer conversions, for example where cookie consent is withheld.
- Judging campaign performance after a few days instead of after a full data-collection cycle.
- Treating the product feed as “set and forget”, even though prices, availability and categories in the store change constantly.
Frequently asked questions
Which campaign type should an online store start with? Usually shopping campaigns based on the Merchant Center feed, because they show the product photo, price and name. Search campaigns on terms with clear purchase intent are worth running alongside them.
Is Performance Max a good choice on day one? It works better once the account has conversion data. Running it as the only campaign from the start makes it hard to tell what actually drives sales.
How long before campaign results can be judged? It depends on the budget and the number of clicks – decisions about scaling or switching campaigns off are worth basing on several weeks of data, not several days.
What next
If you'd rather have someone set up and run these campaigns for you – the Google Ads for online stores service page describes the scope, and the pricing page has the price ranges. For details, get in touch.
