Introduction

Launching a Google Ads campaign is not the end of the work. It is the beginning of a learning process. In the first weeks, real searches, clicks and conversions start to show which keywords, ads, audiences and landing pages work, and which waste money. Over the following months, steady optimisation turns those insights into better results.

Many businesses are disappointed by Google Ads because campaigns are set up and then left alone. Search behaviour changes, competitors adjust their bids, seasons come and go, and Google’s automated systems need guidance and good data. Without regular attention, costs drift upwards and opportunities are missed.

This article describes the third step of our Google Ads process: launch, optimisation and reporting. It explains how we launch campaigns safely, which metrics matter and which can mislead, the weekly and monthly optimisation routines we follow, how we test improvements, how to treat Google’s automated recommendations, and what a useful monthly report should contain. It completes the series that began with keyword research (Insights Article 2) and ads, landing pages and tracking (Insights Article 3).

Launching campaigns safely

A careful launch avoids early waste:

• Final checks: tracking tested, ads approved, landing pages working, settings and budgets confirmed (see Insights Article 3).
• Sensible starting budgets: enough to gather data, but not so much that early mistakes are costly.
• Close monitoring in the first days: checking that ads are showing, clicks are relevant, conversions are recording and spend is in line with plans.
• Patience with learning periods: automated bidding strategies need time and data to learn. Frequent large changes reset this learning, so early adjustments should focus on clear problems, such as irrelevant searches, rather than constant bid changes.

The metrics that matter

Google Ads reports dozens of metrics. These are the most important, and how to interpret them:

The most dangerous metrics are the ones that look good without reflecting business results. A campaign with cheap clicks and a high CTR can still lose money if the clicks do not convert. Our reports always start with conversions, cost per conversion and, where available, revenue and ROAS.

Profit, not just ROAS

ROAS shows revenue per unit of ad spend, but not profit. A ROAS of 4 (four rupees of revenue for each rupee spent) may be very profitable for a product with high margins and unprofitable for one with thin margins. We work with clients to define a target cost per lead or target ROAS based on margins and customer lifetime value. Insights Article 5 explains how to calculate these targets.

Our weekly optimisation routine

Each week, especially in the first months, we review:

1. Search terms: adding negative keywords for irrelevant searches and new keywords for valuable ones.
2. Spend and pacing: checking that each campaign is spending as planned and not limited unexpectedly.
3. Conversions: confirming tracking is working and investigating sudden drops or spikes.
4. Campaign and ad group performance: identifying areas with high cost per conversion or strong results.
5. Ads and assets: reviewing performance and disapprovals, and replacing weak assets.
6. Bids and targets: adjusting targets for automated bidding gradually, where justified by data.
7. Placements and channels: for Performance Max, Demand Gen and Display, reviewing channel and placement reports and excluding poor-quality placements where possible.
8. Policy and account notices: resolving warnings promptly.

Our monthly optimisation routine

Each month, we take a broader view:

• Performance against goals: comparing conversions, cost per conversion and ROAS with targets and previous periods.
• Budget allocation: moving budget from weaker campaigns to stronger ones.
• Keyword strategy: reviewing match types, expanding into new themes and pausing persistent underperformers.
• Audience insights: reviewing which audiences, locations, devices and times of day perform best.
• Landing pages: reviewing conversion rates and recommending improvements to the website team.
• Creative refresh: planning new ads, images and videos, especially for visual campaigns where creative fatigue reduces results over time.
• Testing plan: deciding which experiments to run next.
• Competitor and market changes: noting new competitors, offers and seasonal patterns.
• Reporting: preparing a clear summary with recommendations.

Testing and experiments

Good optimisation relies on evidence rather than opinion. Google Ads offers experiments, which split traffic between the original campaign and a variation, so changes can be tested fairly. Common tests include:

• A new bidding strategy, such as moving from Maximise Conversions to Target CPA.
• Broad match or AI Max for Search versus standard keyword matching.
• A new landing page against the current one.
• New ad messages, offers or calls to action.
• Different campaign structures.

Tests need enough data to reach a meaningful conclusion. In smaller accounts, this may take several weeks. We avoid running too many tests at once, so results stay clear.

Bidding strategies and their targets

Google’s Smart Bidding strategies use machine learning to set bids in each auction. The main options are:

• Maximise conversions: gets as many conversions as possible within the budget.
• Maximise conversions with a target CPA: aims for an average cost per conversion.
• Maximise conversion value: gets as much conversion value as possible within the budget.
• Maximise conversion value with a target ROAS: aims for an average return on ad spend.
• Maximise clicks and target impression share, used for specific situations such as new campaigns without conversion history or brand visibility.

Choosing and adjusting targets is an important part of optimisation. Targets that are too strict can stop campaigns from spending; targets that are too loose waste money. Blog Article 2 explains Smart Bidding in more detail.

Handling Google’s recommendations

Google Ads shows automated recommendations and an “optimisation score”. Some recommendations are helpful, such as fixing disapproved ads or adding missing assets. Others may increase spend without matching your goals, such as raising budgets or broadening targeting.

We review every recommendation against the client’s goals and data, apply those that make sense and dismiss those that do not. A high optimisation score is not the aim; profitable results are. We also review auto-apply settings so that changes are not made to the account without human review.

Budget management through the year

Budgets should follow demand. We plan increases before seasonal peaks, protect budget for the most profitable campaigns and reduce spending where results weaken. Google allows daily spending to vary, up to twice the average daily budget on some days, while keeping monthly spending within limits based on the average daily budget. Monitoring prevents surprises.

For businesses with capacity limits, such as clinics with fixed appointment slots or factories with production limits, budgets and schedules can be adjusted so that ads generate only the enquiries the business can handle.

What a good monthly report includes

Reports should be easy for business owners to understand, not lists of technical metrics. Our monthly reports include:

1. Summary: results against goals in a few sentences.
2. Key figures: spend, conversions, cost per conversion, conversion value and ROAS, compared with the previous period and, where available, the same period last year.
3. What worked: the campaigns, keywords, ads and audiences that performed best.
4. What did not: areas that underperformed and what we did about them.
5. Actions taken: the main optimisations made during the month.
6. Recommendations: proposed next steps, including any decisions needed from the client.
7. Lead quality notes: feedback from the client’s sales team on enquiry quality, which is essential for improving targeting.

We can also provide live dashboards using Looker Studio, which combine Google Ads, Google Analytics 4 and other data sources, for clients who want to follow results between reports.

Measuring beyond the last click

Customers often interact with several ads and channels before converting: a YouTube video, a search ad, a remarketing ad and a direct visit. Google Ads uses data-driven attribution by default for most conversion actions, which assigns credit to different ad interactions based on their contribution.

Reports should also consider effects that are harder to measure, such as people who see a YouTube ad and later search for your brand, or customers who call after visiting your shop. Comparing brand search volumes, direct enquiries and total sales over time gives a fuller picture of advertising impact.

Lead quality: the missing link

For lead-generation campaigns, not all enquiries are equal. Some are ready to buy; others are looking for information, prices for comparison or services you do not offer. Regular feedback from the sales team, and ideally importing qualified leads and sales as offline conversions, lets campaigns optimise for quality rather than quantity.

Simple steps help: ask the sales team to tag each enquiry as qualified or not, record the reason for unqualified leads and share totals monthly. Patterns, such as many enquiries from outside the service area, can usually be fixed quickly in campaign settings.

The first 90 days: what to expect

New campaigns follow a fairly predictable path. Knowing it helps set realistic expectations.

Days 1 to 14: launch and learning. Ads begin to show, and the first search terms, clicks and conversions arrive. Automated bidding strategies are in a learning period. Our focus is on removing clearly irrelevant searches, fixing any tracking or policy issues and confirming that spend is on track. Results can vary widely from day to day.

Days 15 to 45: first optimisations. Patterns emerge. We expand keywords that bring conversions, add more negative keywords, improve or replace weaker ads and assets, and adjust budgets between campaigns. Landing page improvements are often recommended at this stage, based on conversion rates.

Days 46 to 90: refinement and scaling. With more conversion data, bidding strategies and targets can be refined. Successful campaigns may receive more budget, and new campaign types, such as remarketing, Demand Gen or YouTube, can be tested. By the end of three months, most accounts have a clear picture of cost per lead or return on ad spend, and a plan for the next quarter.

Some businesses see good results within weeks; others, especially those in competitive sectors or with long sales cycles, need longer. What matters is steady improvement based on reliable data.

Account housekeeping

Well-organised accounts are easier to manage and less likely to hide problems. Our regular housekeeping includes:

• Removing or pausing old, unused campaigns, ad groups and assets.
• Keeping naming conventions consistent.
• Reviewing shared negative keyword lists and audience lists.
• Checking that user access is up to date, removing people who no longer need it.
• Confirming billing details, payment methods and invoices.
• Reviewing auto-apply recommendation settings.
• Checking the linked accounts, such as Google Analytics 4, Merchant Center, YouTube and Google Business Profile.

Signs a campaign needs a deeper review

Weekly checks catch most issues, but some signs call for a fuller investigation:

• Conversions fall sharply without a change in clicks, which often points to tracking or landing page problems.
• Cost per conversion rises steadily for several weeks.
• Lead quality complaints increase from the sales team.
• Search impression share drops because of rank, suggesting stronger competition or weaker quality.
• A Performance Max campaign shifts spend heavily towards one channel without better results.
• Sales slow in a season that is usually strong.

In these cases, we review tracking, landing pages, competitors, search terms, assets and settings together, rather than changing bids in isolation.

What we need from you

Optimisation works best as a partnership. Clients help most by sharing feedback on lead quality, telling us about stock shortages, price changes, new products and promotions in advance, letting us know about capacity limits such as fully booked periods, and making decisions promptly on recommended landing page changes and budget adjustments. Even a short monthly conversation with the person who handles enquiries often reveals improvements that data alone cannot show.

Remarketing: a common next step

Most visitors do not convert on their first visit. Remarketing shows ads to people who have already visited your website or app, engaged with your YouTube channel or are on your customer lists, reminding them of your offer. It is often one of the most cost-effective campaigns once search campaigns are running well. Remarketing should be used respectfully: limit how often people see ads, exclude existing customers when appropriate, and follow Google’s personalised advertising policies and privacy laws.

Illustrative example

The following example is illustrative and does not describe a real client.

An online store selling home appliances runs Performance Max and Search campaigns in Sri Lanka. After two months, sales are steady, but return on ad spend is lower than the owner needs, because many sales are low-margin items.

The monthly review shows that a large share of spend goes to inexpensive accessories. The team separates products into groups by margin, assigns more realistic values to accessory sales and focuses budget on higher-margin appliances. Search term reviews add negative keywords for repair and spare-parts searches, and new product videos are added to Performance Max. The team also tests a new landing page for the store’s top category. Over the following months, the store achieves a better balance between sales volume and profit.

Frequently asked questions

How often should Google Ads be optimised?

Accounts benefit from weekly checks and monthly strategic reviews. High-spend accounts may need daily monitoring.

Why do results fluctuate from week to week?

Seasonality, competition, budget changes, learning periods and normal statistical variation all affect results. We judge trends over meaningful periods rather than reacting to single days.

Should we follow all of Google’s recommendations?

No. We review each recommendation against your goals and data. Some are valuable, others are not.

What is a good cost per lead?

It depends on your profit margins, sales conversion rate and customer value. We calculate a target that keeps campaigns profitable for your business.

Can we see results in real time?

Yes. You keep full access to your Google Ads account, and we can provide a live dashboard in addition to monthly reports.

Conclusion

Successful Google Ads campaigns are built through steady, evidence-based improvement. Careful launches, the right metrics, weekly and monthly routines, fair testing, thoughtful use of automation and honest reporting turn early data into lasting results.

Our team manages Google Ads with this routine for businesses in Sri Lanka, the UAE and beyond, and reports in plain language so you always know what your budget achieves.

Want your campaigns to improve every month? Ask about our Google Ads management plans.

See Management Plans

Sources

• Google Ads Help, About Smart Bidding: https://support.google.com/google-ads/answer/7065882
• Google Ads Help, About campaign experiments: https://support.google.com/google-ads/answer/6261395
• Google Ads Help, About optimisation score: https://support.google.com/google-ads/answer/9061546
• Google Ads Help, About average daily budgets: https://support.google.com/google-ads/answer/6385083
• Google Ads Help, About data-driven attribution: https://support.google.com/google-ads/answer/6394265
• Google Ads Help, About impression share: https://support.google.com/google-ads/answer/2497703
• Google, “Channel performance reporting coming to Performance Max” (2025): https://blog.google/products/ads-commerce/channel-performance-reporting-coming-to-performance-max/
• Looker Studio: https://lookerstudio.google.com