Conversions that are real
De-duplicated and tested, so optimisation aims at outcomes.
Spend that is accounted for, on pages built to receive it.
Most underperforming ad accounts are not a bidding problem. They are a tracking problem, a match-type problem, or a landing page problem – and the agency reporting on them has no ability to fix the last one.
The first thing to check in any account is whether the conversions are real. Double-counted form submissions, a thank-you page firing on refresh, “conversions” defined as a page view – all common, and all mean the optimisation has been steering towards the wrong thing. Fix the measurement and the account often improves without touching a bid.
The second is search terms. Broad match plus automated bidding will spend your budget on queries you would never choose, and the only way to know is to read the search terms report regularly and build negatives from it. Accounts that are never pruned drift towards irrelevance, quietly.
The third is the landing page, which is where most agencies stop. Sending qualified, expensive traffic to a slow page with a vague call to action wastes the click after you have paid for it. The page is usually a bigger lever than the bid, and it is the part a media-only agency cannot touch.
None of this is advanced. It is the basic hygiene that gets skipped when an account is managed as a dashboard rather than as a business expense.
Structure, tracking, search terms, negatives and wasted spend, reviewed against what you actually sell.
An honest picture before any money moves.
Set up properly, de-duplicated, and tested end to end so the numbers mean something.
Optimisation aimed at real outcomes.
Built around intent, with match types and negatives chosen deliberately.
Spend on searches you would have chosen yourself.
Built and tested by the person running the campaign, not briefed to someone else.
The click converts instead of bouncing.
Feed quality first, because the feed is most of the performance.
Products that show for the right searches.
Search terms read, negatives added, tests run, with monthly reporting in plain language.
An account that is pruned rather than left running.
Six differences, most of them about measurement.
De-duplicated and tested, so optimisation aims at outcomes.
Search terms read and negatives built every month.
Built by the person running the ads, and tested.
Under your billing, your data, your history.
Flat fee, so advice to spend less is not against my interest.
Cost per real enquiry, not impression share.
And when ads are the wrong answer.
Where people are already searching for what you sell.
Where one enquiry is worth enough to justify the click price.
Where the unit economics survive a paid click.
Running for years, never pruned, nobody reading search terms.
Where SEO is right but too slow for the current problem.
If you cannot measure a sale or the margin cannot carry a click, ads will lose money faster than anything else.
Click prices here are lower than in Western markets, which cuts both ways: it is cheaper to experiment and cheaper to waste money quietly for a year. The accounts I see are rarely overspending on a daily basis – they are spending a little, badly, for a very long time.
Location targeting deserves more care than it gets. The default settings include people merely interested in your area, which for a local service business means paying for clicks from people nowhere near you. Tightening targeting to a realistic radius is usually the fastest saving available.
The other local reality is the lead-quality one. An enquiry form that any passer-by can fill in will fill with people who were never going to buy, and the account will faithfully optimise towards producing more of them. Qualifying on the page – budget, location, requirement – costs you some volume and buys back the sales time.
Paying for your actual catchment, not everyone interested in it.
Whether the margin can carry a click, before spending.
Qualifying on the page so the account optimises for buyers.
Tracked properly, since it is how many enquiries actually arrive.
Serving Chennai, Bangalore, India, Canada, UK, USA, Dubai, Singapore.
High intent and enough value per enquiry.
Typically needs: Urgent, local, high-intent searches.
How I help: Tight location targeting, call tracking, fast booking pages.
Typically needs: Emergency and scheduled demand.
How I help: Call-focused campaigns with hours and area handled properly.
Typically needs: Seasonal admission demand.
How I help: Campaigns aligned to the cycle, enquiries qualified on the page.
Typically needs: Product demand that already exists.
How I help: Shopping campaigns driven by feed quality.
Typically needs: Few, valuable enquiries.
How I help: Narrow intent targeting and qualification before the sales call.
Typically needs: High value, long consideration.
How I help: Project-level campaigns with site-visit tracking.
Six steps, and money does not move until the tracking is trusted.
I audit the account and the tracking: what is being counted, whether it is real, and where spend is going that you would not choose.
You get: An audit naming the wasted spend and the tracking faults, with a fixed quote.
Why it matters: Optimising towards a broken conversion makes the account worse, efficiently.
Targets are agreed: what an enquiry is worth, what you can afford to pay for one, and what the account should aim at.
You get: Written targets, with the unit economics behind them.
Why it matters: Without a cost-per-enquiry target, every result can be presented as a success.
Tracking is rebuilt and tested end to end – forms, calls, WhatsApp – with duplicates removed.
You get: Conversion tracking you can trust, verified by real test submissions.
Why it matters: Everything downstream depends on this being right.
Campaigns are built or restructured: intent-based groups, deliberate match types, negatives, and landing pages built to receive the traffic.
You get: A restructured account and the pages it points at.
Why it matters: Campaign and page are one system. Splitting them across two suppliers is why so many accounts stall.
Testing: search terms read weekly at first, negatives added, ads and pages tested against each other.
You get: A pruned account and a record of what was tested.
Why it matters: The search terms report is where waste is found, and it is the report nobody opens.
Ongoing management with monthly reporting on cost per real enquiry, not on impressions.
You get: Monthly management and a report that leads to decisions.
Why it matters: Accounts drift. Pruning is the work, not the admin around it.
Platforms, measurement, and the pages the traffic lands on.
And the commercial terms, which matter as much.
Ads buy visibility now; SEO earns it over time. The useful connection is that the search terms report from a live campaign is the most honest keyword research available – real queries from real buyers, with conversion data attached.
I use that to inform the SEO work, so the pages that earn rankings are the ones already proven to convert rather than the ones a tool suggested.
Google’s automated bidding is genuinely good at its job, which is optimising towards the signal you give it. That is exactly why the conversion setup matters more than the bidding strategy: point it at a bad signal and it will pursue the wrong outcome with great efficiency.
Four things, two of them commercial.
Charging a percentage means advising you to spend less costs me money. A flat fee removes that, and it is worth asking anyone managing your account how they are paid.
The page is usually a bigger lever than the bid, and it is the part a media-only agency cannot touch. Here the campaign and the page are built by the same person.
I will not change bids on an account whose conversions are not trustworthy. It is the least impressive first month and the one that makes everything after it work.
Your billing, your history, your data. If we stop working together you keep everything, which is not true of every agency arrangement.
Tell me what you need. I reply within one working day, and the first conversation costs nothing.