Reach people at the moment they are already searching for what you sell. I build Search, Performance Max and Display campaigns around genuine buying intent, keep the wasted queries out, and make sure the tracking underneath can tell you which clicks actually turned into business.
Google Ads is unusually easy to start and unusually easy to waste. The platform will happily spend a budget that is pointed at nothing.
Most accounts I am asked to look at fail in one of three ways. The first is structural: everything sits in one or two broad campaigns, so loosely related themes share a budget and nobody can tell which is paying for itself. The second is measurement: conversions are either untracked, or tracked so loosely that page views, duplicate submissions and internal traffic all count as wins — so automated bidding optimises hard toward something that is not a sale. The third is neglect: the account was set up competently, then left, and the search terms report has been quietly filling with queries nobody would have chosen to pay for.
The platform's defaults do not help. Left alone, campaigns opt into partner networks, expand into loosely matched queries, and spend across placements you never chose. None of it is hidden, but it is on by default, and defaults are how most budgets leak. The result is an owner who has spent real money, cannot say what it bought, and concludes that Google Ads does not work for their business when nobody was reading the reports the account was already producing.
Underneath the interface, paid search only asks four questions. Everything below exists to answer one of them, in language you can repeat to your team without needing me in the room.
Separating the queries typed by someone ready to buy from the ones typed by researchers, students and competitors — and only bidding on the first group.
Structuring campaigns, ad groups and match types so budget follows intent, and so a single underperforming theme cannot quietly drain the whole account.
Making sure the landing page answers the query the ad promised, because the strongest campaign in the account cannot rescue a page that does not convert.
Tracking that records the actions that matter to your business, so the account is judged on enquiries and sales rather than on clicks and impressions.
Intent first, structure second, automation last — and never automation handed a goal that has not been defined properly.
The starting point is always commercial intent rather than search volume. A term with enormous monthly numbers is usually dominated by people who will never buy: researchers, students, job hunters, competitors checking prices. A narrower phrase typed by someone comparing suppliers or looking for something they are ready to order is worth far more, even though the volume figure looks unimpressive on a slide. In paid search that distinction is not academic — you pay for every one of those clicks.
Structure is what makes that intent controllable. Grouping queries into tight, single-theme ad groups lets the ad echo the search almost word for word, which is better for the person reading it and better for what the auction rewards. It also gives each theme its own visible performance, so a service line that is not converting can be reduced without dragging down the ones that are. Match types are chosen per ad group rather than left broad across the account, and negative keywords are continuous hygiene rather than a setup task: the search terms report is read weekly, and whatever the account has learned it does not want goes in as an exclusion.
Automation gets used, but on terms. Performance Max reaches inventory that Search cannot and is often worth running, provided it is given structure instead of one undifferentiated asset group: groups separated by product or service line so the reporting can be read, audience signals supplied from your own data, assets refreshed rather than uploaded once, and the exclusions the campaign type offers actually applied. Display is used for what it is genuinely good at — re-engaging people who already visited. Smart bidding is only trusted once tracking has been verified, because a bid strategy pointed at a badly defined conversion becomes extremely efficient at buying the wrong thing.
Mapping the queries that precede a purchase, and separating them from informational and job-seeking searches that look relevant in a keyword tool and never convert.
Tight, single-theme ad groups so ad copy can match the query closely, with match types chosen deliberately rather than left broad by default.
Search-terms reports read regularly and negatives added at the right level, so the account stops paying for the queries it has already learned it does not want.
Asset groups split by product or service line, audience signals fed in deliberately, and the exclusions available on the campaign actually applied.
Display used for re-engaging people who already visited, with placement and content exclusions reviewed so ads do not run wherever the network fancies.
Conversions defined and verified before spend scales, then bidding, budgets, ad copy and landing pages revised against what the data shows week by week.
The measurement layer underneath all of this is the same one I use on every channel: which numbers define success, and where each one is sourced, is agreed before the work starts rather than afterwards. How an engagement actually runs — cadence, access, ownership of the accounts — is set out in about.
Paid search is the wrong answer for some businesses, and it is cheaper to establish that before a budget starts running than after one.
Six groups of work. Which of them leads in a given month depends on what the audit turns up and what the search terms say afterwards.
Five phases, with the timings I plan against. They are expectations rather than promises.
An account with years of history and clean tracking moves differently from one being built from nothing — but you will always know which phase the work is in and what comes next.
I start with your business rather than your account: what you sell, what a customer is worth, and what has already been tried. Then I go through whatever exists — campaign structure, search terms, budget history, tracking — and produce a written picture of where the money has been going and what is capping the account. If there is no account yet, this phase is market and query research instead. You finish it knowing what I found, whether or not you take the work further.
Measurement is fixed before spend scales, because everything afterwards depends on it. Conversion actions are agreed, implemented, then tested with real submissions rather than assumed to be working. Duplicate and inflated conversions are cleaned up, primary and secondary actions are separated so bidding optimises toward the ones carrying revenue. An account bidding toward a badly defined conversion becomes very efficient at buying the wrong thing.
Campaigns are built to the agreed structure: search organised into tight themes, ad copy written against the query rather than around it, extensions in place, and Performance Max asset groups split so their reporting can be read. Negatives go in from day one rather than after the first wasted month. Where landing pages work against the ads, I will say so and either fix them or brief whoever can, and launch is staged so nothing reaches full budget on an untested assumption.
The account needs data before it can be judged, and intervening daily in the first fortnight costs more than it saves. Search terms are reviewed and negatives added; ad copy variants are tested; budget shifts toward the themes earning conversions and away from those that are not; bidding moves onto automated strategies once there is enough conversion volume for them to work with. Changes are made one at a time so it stays possible to say which one caused what.
A steady rhythm: weekly search terms and performance review, continuous negative keyword work, seasonal and promotional adjustments, asset refreshes before creative fatigue sets in, and a monthly report plus a call where I explain what changed and what comes next. Auctions shift as competitors enter and leave, and platform defaults change without notice, so an account left alone drifts.
Management is quoted after the discovery conversation, and it is always separate from the money you give Google.
Two figures matter and they should never be blurred together. The first is your media budget — what the auction charges for clicks, billed by Google directly to your own payment method, in your own account. The second is my management fee for the strategy, the build and the ongoing work. Where that fee lands depends on how many campaigns and markets are in scope, how much rebuilding the account needs, whether tracking has to be built from scratch, and how much ad copy and creative work is involved. I do not price per keyword or per campaign, because both reward the wrong behaviour: one encourages padding an account with terms nobody searches, the other encourages splitting work into campaigns that never needed to exist.
I recommend a minimum of three months, because the first weeks are partly research: the account has to gather search terms and conversion data before it can be optimised rather than merely adjusted. Judging paid search on a fortnight rewards whoever tells the best story about the learning period. If at any point the honest read is that your budget would do more on another channel, or that the maths does not support paid search at all, I will tell you.
See packages and pricingPaid search rarely works hardest on its own. These are the three it most often runs beside — the full list is on services.
Google Ads is one channel inside a paid programme. Run together, budget can move to whichever platform is producing conversions rather than sitting where it started.
Explore paid advertisingPaid and organic compete for the same results page. Knowing which queries you already earn for free changes which ones are worth bidding on.
Explore SEOIf it is not yet clear that paid search is the right channel at all, the strategy work answers that question before a budget is committed to it.
Explore strategy and consulting