How we work
Audit first, then decide
Nobody can promise you a ranking or a conversion rate. What can be promised is how the work is scoped, who does it, what gets fixed first, and what you keep when it ends.
How does an engagement start?
With a paid diagnostic on the channel in question, scoped and priced before it begins, delivered as a written document you keep regardless.
The first conversation is not a pitch. We ask what you are running, what changed recently, and what the number is that made you go looking. Most of the time that conversation ends with us naming the two things we would check first, whether or not you hire us to check them.
If it goes further, the next step is a diagnostic on the specific channel in question, with the scope and the price agreed before any work starts. It produces a written findings document with the problems ranked and the reasoning shown, and that document is yours whatever you do next. Take it in-house, take it to another agency, or act on it yourself.
Only after that does anyone talk about ongoing work. A retainer decided before the account has been looked at is a guess dressed as a quote, and the person carrying the cost of that guess is always the client.
Why does every engagement start with an audit?
Because scoping ongoing work before seeing the account means guessing, and a monthly fee set by guesswork is wrong for somebody in every direction.
Two accounts with identical spend can need completely different amounts of work. One has clean conversion tracking, a sensible structure and a landing page that functions, and needs a few hours a week. The other has three conflicting tags, a conversion action counting page views, and six years of accumulated campaign sediment, and needs a month of unpicking before optimisation means anything.
An agency that quotes both the same way is either overcharging one or underserving the other, and usually both. Auditing first prices the unknowns out rather than padding for them.
It also gives you an exit that costs you nothing but the diagnostic. You find out what is actually wrong, and you find out how we think, before either side commits to a monthly relationship.
How do you decide what to fix first?
By expected impact divided by effort, with anything that corrupts measurement moved to the front, because decisions made on broken data stay wrong afterwards.
Measurement problems jump the queue regardless of their own size. If the conversion data is wrong, every optimisation made on top of it inherits the error, the automated bidding learns the wrong thing, and the reporting that would have caught it is the thing that is broken. Fixing a tracking fault is rarely the most exciting item on a findings list and it is almost always the first one.
After that, the ranking is expected impact against the effort to get it, stated openly so you can disagree. Where two items are close, the one that is reversible goes first.
Some of what comes out of this is not marketing work at all. If the answer is that the offer is wrong, or that the sales team is not calling the leads back, we say so, because spending more on media in front of either of those makes the problem more expensive rather than smaller.
Who actually does the work?
The person who audited your account. There is no account manager layer, no junior handoff after signature, and no pod you get rotated through.
The common agency failure is not incompetence, it is the handoff. Senior people win the work and more junior people do it, and the gap between what was described in the pitch and what happens in the account is where the relationship quietly fails.
This practice is small on purpose so that gap does not exist. The constraint that keeps it honest is capacity: we can only hold as many accounts as one senior specialist can genuinely work on, so at times the right answer is that we do not have room, and we say that rather than taking the work and thinning it out.
That is also why growth into a layer of account managers is not the plan. It would break the only thing that makes this model worth choosing over a larger agency with more resources.
How is the work reported?
Short, written, and tied to decisions. Every number carries the action it should trigger, and a metric nobody would act on gets removed.
A twenty-tile dashboard is a cost, not a deliverable. It takes time to build, more time to maintain, and it is read for about four seconds a month. Worse, it hides the two numbers that matter among eighteen that do not.
Reporting here is written prose with the numbers in it: what we changed, what happened, what we are doing next, and what we got wrong. If a metric appears, it is because a decision hangs off it. If nothing would change whatever the number said, it comes off the report.
You also get told when something did not work. An agency report that only contains wins is not a report, it is marketing, and you are already paying for marketing.
Who do you take on, and who do you turn away?
We take work where measurement can be trusted and the budget can move. We turn away guaranteed rankings, unmeasurable goals, and accounts we cannot improve.
Being specific about who this does not suit is more useful than a list of industries served. The criteria below are the actual ones, and they are the questions we ask ourselves before quoting.
Work we take on
The conversion you care about can be measured
Not necessarily measured well today, because fixing that is often the first job. But it has to be measurable in principle. If the real outcome happens six months later in a system nobody will connect, media optimisation has nothing honest to aim at.
There is enough volume for decisions to mean something
Enough spend or enough conversions that a change produces a signal rather than noise. Below that threshold the useful work is structural rather than iterative, and we will say so instead of billing for optimisation that cannot be evaluated.
Somebody on your side can make a decision
Work stalls when every recommendation needs four approvals and a committee. It does not need to be a large team. It needs one person who can say yes to a landing page change without a quarter of lead time.
You want to know what is wrong
The diagnostic finds things. Some of them will be decisions somebody at your company made and defends. If a findings document naming those is going to be a political problem, the engagement will be a political problem too.
Work we turn away
Anyone wanting guaranteed rankings or guaranteed lifts
Nobody controls a ranking system or an ad auction. An agency promising a specific position or a specific percentage improvement is either misunderstanding what it sells or knowingly misrepresenting it, and we will not compete on that basis.
Accounts where we cannot see a way to add value
Sometimes an account is already run well and the honest finding is that the ceiling is elsewhere. Taking a retainer to make small changes to something already working is easy money and it is not what this is for.
Work that needs a headcount we do not have
Large multi-market programmes needing several specialists in parallel are a real need and a bigger agency serves them better. Pretending otherwise would mean thinning out every other account to cover it.
Anyone who wants the reporting to look good
If the brief is a dashboard that keeps a board comfortable rather than an account that performs, there are agencies who do that well. The reporting here is designed to surface problems, which is the opposite requirement.
What do you commit to, if nobody can promise results?
Response times, who does the work, what you own when we stop, and a notice period short enough that staying is a choice.
Nobody controls a ranking, an auction or a conversion rate, so a guarantee about any of them is either a misunderstanding or a sales tactic. What can be committed to is everything around the outcome, and those commitments are worth more precisely because they can be checked.
You get a reply within one business day
Enquiries and client messages are read by the person who does the work, not by an inbox rota. If something is genuinely urgent, calling reaches the same person.
The person who audits your account is the person who runs it
No handoff after signature, no junior taking over once the relationship is signed, no rotation through a pod. If that ever has to change, you hear it from us before it happens.
The findings document is yours whatever you decide
The diagnostic is bought outright. It is not a sales artefact held back pending a retainer, and it contains the reasoning rather than just the conclusions so somebody else could act on it.
Every account and property stays in your name
Ad accounts, analytics properties and tag containers are yours, with your own admin access kept throughout. We work inside your assets rather than renting them back to you, so leaving costs you access to nothing.
The commercial terms are on the table before you sign
Scope, fee, what is included and the notice period are all agreed in writing in advance. There is no annual lock-in used as a retention mechanism, because a client who stays only because leaving is expensive is not a reference.
You are told what did not work
Tests that failed and calls that were wrong appear in the reporting alongside the ones that worked. An agency that only reports wins is filtering, and filtered reporting makes every future number less useful.
What certifications and partner status do you hold?
Google Partner, Google Ads individual certifications through Skillshop, Meta Blueprint, and Microsoft Advertising Partner. Each is a platform credential, not a performance guarantee.
Platform credentials prove that someone has passed the platform's own examinations and, in the case of partner status, that an account footprint and spend threshold have been met. They do not prove that anyone is good at this, and an agency leaning on them as evidence of results is telling you something about the strength of its other evidence.
They are listed here because they are verifiable facts and because the platforms require accuracy in how they are stated. They are not offered as a reason to hire anyone.
Google Partner
Issued by Google
Company-level status in the Google Partners programme, which requires certified staff, an active spend threshold across managed accounts, and a performance requirement measured by Google.
Google Ads certifications
Issued by Google Skillshop
Individual certifications taken through Skillshop, Google's own training platform, and revalidated annually. They cover the campaign types the platform examines directly.
Meta Blueprint
Issued by Meta
Meta's own certification programme for advertising on Facebook and Instagram, examined and issued by Meta rather than by a third party.
Microsoft Advertising Partner
Issued by Microsoft
Partner status in the Microsoft Advertising programme, covering the Bing, Yahoo and partner search network that Google-only agencies routinely leave unmanaged.
Where to go next
How pricing works covers what actually drives the cost of an engagement, with third-party benchmark data rather than a rate card. About Admaster covers who you would be contracting with, and the calculators let you run the economics yourself before speaking to anybody.
Tell us what is not working
The first conversation costs nothing and usually ends with us naming the two things we would look at first, whether or not you hire us to look at them.