Google Ads for ecommerce: the first 30 days step by step

The first thing I do when starting with a new Google Ads account for ecommerce is open Google Analytics. Not Google Ads. GA4. I want to see what the site is measuring before spending a single penny on paid traffic.

That order matters. The first 30 days of a Google Ads account aren’t about running advertising. They’re about building the foundations that advertising will rest on. If those foundations are wrong from the start, every pound you invest afterwards will work harder than it should have to.

This guide covers the process I follow with every new client: what I configure, what I review, what I don’t touch yet, and when I start optimising properly. If you already have active campaigns and want the full picture, the complete Google Ads guide for ecommerce gives you the global view of the whole strategy.

Key Takeaways

• Before launching, tracking must work without errors: no duplicate events, no misattributed conversions
• The first week is about structure and launch, not optimisation
• Don’t touch bids until you have at least 7-10 days of data
• The search terms report is the most valuable tool in weeks 2 and 3
• Smart Bidding only works well when you have enough conversion volume: wait until month 2

---

Contents

1. Before touching the platform
2. Week 1: structure and launch
3. Week 2: first data and quick fixes
4. Week 3: real optimisation begins
5. Week 4: review and next steps
6. Common mistakes in the first 30 days
7. Frequently asked questions

---

Before touching the platform

In my experience, 60-70% of the performance problems I see in ecommerce accounts have their root in tracking. Not in bids. Not in ads. In tracking. That’s why, before creating the first campaign, I spend time verifying that everything that matters is being measured correctly.

Setting up conversion events

The first step is defining what counts as a conversion. For an ecommerce, the answer seems obvious: a purchase. But there are details that make a real difference.

Set up the purchase conversion with the following parameters:

• Dynamic conversion value: the actual amount of each order, not a fixed value
• Deduplication: use the order ID as the identifier to prevent the same sale being counted twice
• Category: “Purchase”
• Conversion window: 30 days for clicks, 1 day for views

On deduplication, I say this from direct experience: without it, it’s common to see accounts counting double the actual conversions because the Google Ads pixel and the GA4 tag fire simultaneously on the same confirmation page. The Smart Bidding algorithm optimises against that inflated number. When you fix the error, the “real” ROAS drops by half. It’s an uncomfortable conversation with any client.

Linking GA4 and activating enhanced conversion tracking

Link GA4 to Google Ads from the GA4 admin panel. This lets you import conversions directly from GA4 rather than relying solely on the Google Ads tag. In well-configured accounts, GA4 is the source of truth for conversions - it’s more resilient to ad blockers and more consistent across devices.

Also activate enhanced conversion tracking, which improves attribution using hashed first-party data (email, name, phone). Google confirmed in its documentation that this can improve conversion measurement by between 5 and 15%, especially for users who block third-party cookies (Google Ads Help, 2025).

Setting up Google Merchant Center

If the ecommerce is going to use Shopping or Performance Max campaigns, you need Merchant Center linked to Google Ads and an approved product feed. Without a feed, there’s no Shopping. And with a poorly optimised feed, Shopping performs well below its potential.

The three fields with the most impact on feed performance:

1. Product title: include the main keyword, the brand and the differentiating attribute (size, colour, material)
2. Images: white background, high resolution, no overlaid text
3. GTIN: the product barcode. Without it, the feed is at a disadvantage compared to competitors who include it

Don’t launch without the feed showing zero errors in the Merchant Center diagnostics panel. You can leave warnings for later. Errors, you can’t.

---

Week 1: structure and launch

With clean tracking and an approved feed, the time comes to create campaigns. The temptation is to launch everything at once. Don’t.

What type of campaign should you start with?

I get this question on almost every new account. The answer depends on the available budget and the expected conversion volume.

Situation	Campaign type to start with
Budget < £500/month	Standard Shopping + brand Search
Budget £500-£2,000/month	Standard Shopping + product Search
Budget > £2,000/month	Performance Max with feed + brand Search
No product feed	Search only, until Merchant Center is active

Performance Max needs a minimum number of conversions to learn well. If volume is low, Standard Shopping gives more control and more predictable results in the first weeks. To understand the differences between both approaches in detail, the complete Google Ads guide for ecommerce covers the selection criteria more thoroughly.

Account structure in week 1

Less is more at the start. A new account with too many campaigns fragments the budget and slows the algorithm’s learning.

Recommended structure for a new account:

• 1 Shopping campaign (Standard or PMax, depending on budget)
• 1 brand Search campaign (only your brand name as keywords)
• Optionally: 1 product Search campaign if you have keywords with clear purchase intent

The brand campaign is cheap, protects your position on high-intent searches, and gives you quick conversion data that the algorithm can use. Don’t skip it.

Keywords and match types in Search

In Search campaigns, start with phrase and exact match. Broad match can work well once the algorithm has a conversion history, but in week 1 it doesn’t have one. With broad from the start, budget goes to irrelevant searches before you have enough data to add negatives.

My match types at launch:

• Exact for the keywords with highest purchase intent (e.g. “buy running shoes”, “running shoes price”)
• Phrase for variations of those same intents
• Modified broad, only if search volume is very low in niche categories

Negatives from day 1

There are terms I always add as negatives before launching, without waiting to see the search report:

• “free”, “freebie”, “no cost”
• “second hand”, “used”, “pre-owned”
• “PDF”, “download”, “tutorial”
• Competitor names (if you don’t want to pay for someone else’s brand traffic)
• Employment terms: “job”, “vacancy”, “career”

Add them as a negative list at account level so they apply to all current and future campaigns.

Initial budget

A practical rule: the daily budget should allow at least 10-20 clicks per day. Below that, the algorithm takes too long to accumulate data. Above that, you can scale gradually once the initial ROAS is positive.

---

Week 2: first data and quick fixes

Seven days have passed. You have data. What do you look at first?

The first things I check when opening the account in week 2

Before looking at ROAS or conversions, I check these three things:

1. Are conversions being recorded? If there are clicks but zero conversions in 7 days, there’s a tracking problem or the campaigns aren’t reaching the right audience.
2. Are there impressions in Shopping? If the feed isn’t configured properly, Shopping campaigns won’t serve. Check the Merchant Center diagnostics.
3. Is the cost per click consistent with the sector? If the CPC is 5x higher than the market reference for your keywords, something is wrong with the structure or the initial bid.

The search terms report: your most valuable tool

In week 2, the search terms report is what takes me the longest. And it’s worth it. That’s where you discover what people who click on your ads are actually searching for.

Open the report under Keywords > Search terms. Sort by spend, descending. Look at the first 20-30 terms.

You’ll find three categories:

• Relevant terms with conversions: these are the ones the algorithm has already found. Add them as exact match if they aren’t already.
• Relevant terms without conversions (yet): they may need more time, or the landing page isn’t the right one for that search.
• Irrelevant terms: add them as negatives immediately.

Don’t wait for perfect data to add negatives. With 7 days you can already start.

Quality Score: which signals to look at

Quality Score is an indicator, not a goal in itself. But in week 2 it gives you clues about where there’s friction.

What I do look at in this phase:

• Expected CTR: if it’s “below average”, the ads aren’t connecting with search intent. Review the headlines.
• Ad relevance: if it’s low, the ads and keywords are in the wrong ad groups.
• Landing page experience: if it’s low, the problem isn’t in Google Ads. It’s in the website.

What NOT to touch yet

Here’s the most common mistake in week 2: changing too much, too soon.

Don’t touch bids. Don’t reorganise the structure. Don’t pause campaigns that “aren’t working” yet. The algorithm needs time. With 7 days of data, the only things you can confidently do are add negatives and fix tracking errors. Everything else waits.

---

Week 3: real optimisation begins

With two weeks of data, you can start optimising. A little. With judgement. Based on data, not intuition.

Bid adjustments

If campaigns are on manual bidding or “Maximise clicks”, week 3 is when you start making adjustments based on real data.

Two types of adjustment that make sense at this stage:

Device adjustment: if mobile has a similar CPC to desktop but a ROAS that’s 3x lower (which is very common in ecommerce with a poorly optimised checkout), apply a negative adjustment of 20-30% on mobile.

Schedule adjustment: if the ad scheduling report shows that Tuesdays and Wednesdays between 8pm and 11pm concentrate 40% of conversions with the lowest CPA, you don’t need more data to increase bids during that window.

In the ecommerce accounts I manage, the most common pattern in schedule adjustments is that the highest-converting hours are weekday evenings (6pm-10pm) and weekend mornings (10am-1pm). But it varies considerably by sector and product type.

Audience layering

Add audiences as an observation layer to your Search campaigns. Not as targeting, just to watch. The audiences I typically add on observation from week 3:

• Site visitors (remarketing)
• Previous buyers (customer list uploaded to Google Ads)
• In-Market segments related to your category

With observation, you see how those audiences perform without changing the campaign’s reach. In week 4 or 5 you can decide whether it’s worth applying a positive bid adjustment for them.

First ad tests

If Search campaigns already have more than 500 impressions per ad group, start testing headline variations.

A small, focused test: change headline 1 in two ads within the same group. Everything identical except headline 1. Let it run 10-14 days before drawing conclusions. With RSAs (responsive search ads), Google already optimises between assets automatically, but you can force the comparison using the combinations report.

---

Week 4: review and next steps

At the end of the first month you have enough data for an honest review. Not perfect. Honest.

Real ROAS vs. target

Calculate the real ROAS from the past 4 weeks. Compare it with the target ROAS for the account.

If you’re above target: good. You can scale gradually (increase budget by 20-30% every 7-10 days, not all at once).

If you’re below target: it’s not time to panic. The first month is usually the worst in terms of ROAS because the algorithm is still learning. The relevant question is whether the trend is positive: is week 4 ROAS better than week 1 ROAS?

If ROAS doesn’t improve week on week, the problem probably isn’t the bids. It’s the feed, the tracking, the product page, or the price.

When to switch to Smart Bidding?

The question I get most often at the close of the first month. My personal criterion: I don’t activate target ROAS until there are at least 30 conversions in the past 30 days in that specific campaign.

Below that threshold, the algorithm doesn’t have enough data to model correctly. According to Google Ads Help, Smart Bidding strategies perform better when there’s sufficient signal volume. With fewer than 30 conversions per month, target ROAS tends to restrict bids excessively and delivery drops.

If at the end of month 1 you have 30+ conversions, activate Smart Bidding with a conservative target ROAS, set 10-15% below your actual ROAS from the past 4 weeks. Don’t set the target at what you “want” to achieve. Set it at what the algorithm can actually deliver.

What to scale and what not to

In the month 1 review, categorise your campaigns into three groups:

• Campaigns with positive ROAS and volume: scale budget gradually
• Campaigns with positive ROAS but low volume: review restrictions (budget, overly restrictive keywords, too-aggressive negatives)
• Campaigns with negative ROAS: pause if they’ve gone 30 days without conversions. No more, no less.

Don’t scale campaigns that aren’t working in the hope that “more budget will fix it”. It rarely works that way.

---

Common mistakes in the first 30 days

These are the mistakes I see most frequently when I audit accounts that have been active for a month and aren’t delivering results.

Changing things too quickly

The most common mistake and the most costly. You change bids on Monday, restructure on Wednesday, pause two campaigns on Friday. Then you wonder why the data isn’t consistent.

Every significant change in an account resets or interrupts the algorithm’s learning. Google’s algorithm needs time. Give it at least 7 days between substantial changes to evaluate the impact.

Ignoring the search terms report

Many advertisers set up campaigns, look at ROAS, and never open the search terms report. It’s the most revealing report in the first 30 days. If you don’t review it at least once a week, you’re flying blind.

There’s a pattern I see repeated in ecommerces with Shopping campaigns: the terms consuming the most spend without converting tend to be informational searches (how to use X, what is X, X review), not purchase searches. Adding them as negatives can free up between 15 and 30% of the budget towards searches with real transactional intent.

Using incorrect conversion events

Some accounts have the main conversion event configured as “confirmation page view” rather than “purchase with transaction value”. The result: the system optimises towards sessions that reach the thank-you page, not towards actual orders.

Before launching, verify in Google Tag Assistant that the conversion event includes the value parameter with the actual order amount and transaction_id for deduplication.

Daily budget set too low

A campaign with a £5/day budget in a market where the average CPC is £1.50 generates 3-4 clicks per day. With that volume, the algorithm can’t learn. The account stays in a permanent learning state that never stabilises.

If the available budget is very limited, it’s better to concentrate it on a single campaign that works with sufficient data than to spread it across five campaigns that have no volume.

Expecting results in the first week

Google Ads isn’t immediate-response advertising to the extent people sometimes assume. In the first month, the goal isn’t a perfect ROAS. It’s validating tracking, accumulating data, and building the foundation to scale in months 2 and 3.

If anyone promises spectacular results in the first week of a new account, be sceptical.

---

Frequently asked questions

How much budget do I need to start with Google Ads for ecommerce?

There’s no universal figure, but a functional minimum is between £500 and £1,000 per month. Below £500, the data volume is so low that the algorithm takes months to learn. What matters isn’t the absolute number, but that the budget allows at least 10-15 clicks per day on the main campaigns.

Can I launch Performance Max from day 1?

You can, but it isn’t always the best approach. Performance Max works well when there’s a conversion history in the account. If it’s a new account with no previous conversions, Standard Shopping gives more predictable results in the first month. Switch to PMax in month 2 or when you have 30+ accumulated conversions.

When should I activate target ROAS?

According to Google Ads Help, Smart Bidding strategies perform better when there’s sufficient data volume. The standard is to wait for 30-50 conversions per month in that campaign before activating target ROAS. With less volume, the algorithm bids too conservatively and delivery drops.

What do I do if I have no conversions after 2 weeks?

First, verify that tracking is working correctly. Check in Google Tag Assistant or GA4 debug mode that the purchase event fires on the confirmation page. If tracking is fine, review the search terms report: the campaigns may be reaching searches with no purchase intent. And review the product page and checkout: sometimes the problem isn’t in the ads.

How often should I review the account in the first month?

Once a week is enough for making changes. You can take a quick look every 2-3 days to spot anomalies (runaway spend, zero impressions, tracking errors), but don’t make optimisation decisions with fewer than 5-7 days of data. A weekly review with clear criteria is worth more than daily adjustments without sufficient data.

---

What comes next

The first month is just the beginning. If you reach week 4 with clean tracking, a reviewed search terms report, and a ROAS that’s at least pointing in the right direction, you’ve done the hardest work.

Month 2 is when real optimisation starts: activating Smart Bidding with sufficient data, scaling what works, going deeper into Shopping structure with custom labels by margin, and starting to test Performance Max if you haven’t done so yet.

To understand how to advance the overall strategy once the first month is behind you, the complete Google Ads guide for ecommerce covers account structure, campaign types, and scaling criteria in more detail.

If you’d like an external review of what you have set up at the close of that first month, I can carry out a free audit of your account with a concrete report on the areas to improve.

Want to know if your first-30-days setup is correct? Download my Google Ads ecommerce audit checklist: it covers tracking, structure, feed, bids, and the 10 most common mistakes in new accounts. Request the free checklist here.

---

Sources

1. Enhanced conversion tracking - Google Ads Help
2. Smart Bidding and conversion thresholds - Google Ads Help
3. Merchant Center - Feed diagnostics
4. Google Tag Assistant - Tag verification
5. Conversion deduplication - Google Ads Help
6. Negative keywords - Google Ads Help
7. Quality Score - Google Ads Help
8. Performance Max - Google Ads Help