Google Ads management for
UK D2C brands.
Run on the P&L, not ROAS.
D2C · From £3,000/month
I've led strategy and execution for paid search across agency and in-house, including at Loaf - answering to finance directors, CEOs and growth leads rather than to a dashboard.
Now I do the same thing for a small number of high consideration D2C brands. One specialist in your account, measured on contribution margin and new customer growth rather than on what the platform reports back.
Five years of the same argument
Five years, agency side and in-house. Most of it spent in rooms with CEOs, CFOs and marketing teams having a version of the same conversation.
Someone points at a dashboard. ROAS is up, the account looks healthy. And the person holding the P&L knows the business doesn't feel like it's growing, but can't say why from the numbers in front of them.
At Loaf I ran the search budget in-house, seven figures a year in a premium furniture category. That's where the argument stopped being theoretical. Sitting on the client side you find out quickly that platform return isn't the number anyone upstairs cares about. Contribution margin is. Cash flow is. Whether the customer base is actually growing is.
ROAS was never wrong, exactly. It was answering a question nobody in the room had asked.
The hidden conflict of interest your agency runs off.
Most Google Ads agencies charge a percentage of ad spend. Others take a revenue share on top of it, or quietly mark up the media before it reaches the auction. However it's dressed up, the arithmetic is identical: growing your budget grows their fee, whether or not it grows your profit.
Then there's the handoff. You meet a senior strategist at the pitch. Someone else manages the account - juggling 20 clients, optimising for ROAS because that's what gets reported, with no idea what your margins look like.
I work flat fee only. No percentage of spend. No revenue share. No markup on media. No commission from any platform, CSS or vendor. The fee is the fee, and it does not move when your budget does.
ROAS tells you what the platform thinks happened. These four tell you what's actually going on.
ROAS flatters. Repeat buyers, brand searches, the users who were going to buy anyway - all of it inflates the number and makes an account look healthier than it is. The business could be standing still while the dashboard looks green.
The north stars I work to:
What you keep after cost of goods and variable costs - the only profit number that actually matters.
What it cost to win a customer who hasn't bought from you before. ROAS tells you nothing about this.
Whether the business is actually growing its customer base or recycling existing demand.
What the spend does to working capital. Stock, lead times and payment terms decide how fast you can actually scale.
Together those four tell you something ROAS never can - whether you're building a business or recycling existing demand at an increasingly expensive rate.
What the ROAS number hides
Two accounts, with illustrative figures. Neither problem is visible in the metric the platform puts in front of you, and both change what you should do next.
A brand runs at 40% contribution margin before media, which sets breakeven ROAS at 2.5. Every £1 of ad spend needs £2.50 back before the advertising has paid for itself.
At 40% contribution, £1.90 of revenue produces 76p against £1 of media. Every pound of non-brand advertising is roughly 24p underwater. Brand demand was always going to convert. It's carrying the blended number and hiding the fact that acquisition is losing money.
A different account reports 9.0 ROAS. New customer acquisition cost sits at £80, against £240 of contribution on a first order.
That business could pay three times as much to win a customer and still break even before a single repeat purchase. The high ROAS isn't efficiency, it's evidence of under-investment - targets set so high the account only ever bids on demand that was already going to arrive.
This is the work. Not making the dashboard look better, but finding out what the dashboard is concealing and deciding what the business should do about it.
Which of those two is your account? A fixed-price audit answers it with your own numbers.
The method, published in full
The models I run inside client accounts are the same ones sitting on this site as free calculators - margin waterfalls, breakeven thresholds, feed diagnostics. Anyone can use them on their own numbers.
If you want to know how I'd think about your account, the fastest route is to run your own figures through them first. No sign-up, no email gate, nothing held back for paying clients.
Map every cost that touches a basket, end to end, then read breakeven ROAS and breakeven NCAC off the result. Built per vertical, because a furniture basket and a skincare basket break down nothing alike.
Every common GMC error and warning, what actually causes it, and the fix - rather than the Google help-centre version that tells you the error exists.
Responsive search ad structures and shopping title formats front-loaded with search language instead of collection names nobody types.
Where finance and instalment options change the economics on a high AOV basket, and what that does to the most you can afford to pay for a customer.
All free, in the resources section of this site.
Starting with a full audit. Strategic from day one.
Every engagement starts with the same account and GMC audit I sell as a standalone service - margins, SKU roles, feed health, account structure, conversion tracking integrity. Most people start there, and it's the lower-risk way in: a fixed-price diagnosis with no retainer attached to it.
The difference under a retainer is that the diagnostic lens never switches off. As your performance marketing consultant, I'm in the account spotting not just when metrics shift, but why - and whether the reason is an account problem, a feed problem, a margin problem, or something structural that no amount of optimisation will fix.
The longer I'm in the account, the more context I accumulate. Seasonal patterns, product launches, what held up and what didn't. An agency that churns clients resets that knowledge to zero. I don't.
What's actually included
Not a channel list. This is the work that happens in an account run against a P&L rather than against a platform target.
- Breakeven ROAS derived from your contribution margin
- SKU-level CM2 and CM3 profitability modelling
- Brand and generic separation, and true non-brand return
- Attribution model and conversion window audit
- Offline and in-store sales imported as conversions
- Incrementality assessment on local and store campaigns
- Shopping, Performance Max, Search, Demand Gen and remarketing
- Campaign restructure for catalogue coverage
- Custom label systems driven by margin, velocity and spend
- Impression share and target headroom analysis
- Conversion volume thresholds and portfolio bidding
- Negative keyword frameworks and brand exclusions
- Scripts and automation
- Account hygiene: targeting, inventory type, content exclusions, AI asset settings, ad schedules
- Title restructuring, front-loaded for search language
- Attribute enrichment: colour, size, material, pattern, product type depth
- AI-surface attributes: Q&A, document link, product highlights, video link
- Merchant Centre settings, diagnostics and disapprovals
- Feed management platform configuration
- CSS assessment
- Daily Slack access, direct to me
- Weekly calls against contribution margin, NCAC, cash flow and new customer targets
- QBRs in person if you want them
- 30/60/90 review checkpoints against a fixed baseline
- You own the account and all the data throughout
That's the scope. Whether it's the right level for your account is a twenty-minute conversation.
Direct access. No agency middle layer.
Daily Slack comms. No account manager fielding your questions. No waiting until the next scheduled call to flag something urgent.
Weekly calls to go through what actually happened - against your contribution margin, NCAC, cash flow and new customer targets, not a report of platform metrics. Quarterly reviews in person if that's useful to you.
Why 3 months, not monthly rolling from day one
Bidding changes, feed restructures, campaign rebuilds - none of them show their full impact in 4 weeks. The 3-month initial term exists because meaningful work takes time to prove out.
That's even more true in high consideration verticals where buying cycles can run 90 days or longer. Cutting an engagement short before a full cycle completes doesn't tell you anything useful. Here's what the three months actually contain.
Business deep-dive on margins and SKU roles. Account and feed audited together. Measurement corrected - attribution, brand and generic separation, conversion integrity. Account hygiene fixed. Structural work begins.
Campaign restructure for catalogue coverage. Custom label system wired up against margin and velocity. Feed enrichment across titles, attributes and AI surfaces. Targets held steady so the changes can be read.
First full buying cycle runs through the new structure. Exposure scaled against proven margin thresholds. 30/60/90 checkpoint against the baseline set in month one - the earliest honest read on whether the work has landed.
If we both want to continue, we re-sign on a rolling monthly basis. Seasonal context accumulates, and decisions get faster because the account history sits in one head rather than being handed between people.
Flat fee. No percentage-of-spend conflicts.
No percentage of ad spend. No markup on budget. No kickbacks off growing your spend. The only thing that changes month to month is whether your business is growing - which is exactly how it should be.
- 3-month initial term, then re-signed monthly by agreement
- Account and Merchant Centre audit included from day one
- You own the account and all the data throughout
Brands spending £25k+ per month on Google Ads. Below that the fee stops making commercial sense as a proportion of what's being managed, and I'd rather say so than take the work.
If you're in the £15k to £25k range, a one-off account and feed audit gets you more per pound than a retainer would.
And if you have a capable team already, consulting is a separate offer entirely - strategic oversight and training rather than execution. That one isn't tied to a spend threshold at all.
Questions you should be asking
What's the minimum spend?
£25k per month on Google Ads for the retainer. Between £15k and £25k, a one-off audit or a bespoke consulting engagement is the better use of your money and I'll point you there instead. Below £15k, neither of us would be getting value from a monthly fee at this level.
We don't need execution, just strategic oversight. Do you do that?
Yes, and it's a separate engagement from the retainer. Consulting is for teams who can already execute but want the strategy directed properly - account architecture reviews, a second opinion on structure or bidding, margin and SKU-role modelling handed over for your team to run, or training on profit-first Google Ads and feed management.
It works as a one-off diagnosis, a standing advisory slot, or a fractional arrangement where handing over the whole account isn't the right fit. Priced per engagement rather than on the retainer scale, and there's no spend threshold attached - the question is whether you have the team, not the budget.
Are you a Google Partner?
No, and I'd be cautious about what that badge tells you. Partner status is awarded on three things: a 90-day spend threshold, certification exams, and maintaining a 70% optimisation score.
That last one is the problem. Optimisation score rises when you adopt Google's own recommendations - broad match, auto-apply, budget increases, target changes. Several of those are the first things I turn off in an account. The badge measures alignment with Google's commercial interests. It doesn't measure whether anyone is protecting yours.
Why a flat fee instead of a percentage of spend?
Because a percentage means I get paid more for spending more of your money, regardless of whether that spend made you anything. Plenty of the highest-value decisions I make involve cutting spend, pausing campaigns, or telling you not to scale yet. None of those should cost me money.
Why a 3-month initial term?
Feed restructures, campaign rebuilds and bidding changes don't show their full impact in four weeks, and in high consideration verticals a single buying cycle can run 90 days or longer. Ending before a full cycle completes tells neither of us anything useful.
It doesn't auto-renew at the end. We sit down, look at what the three months actually produced against the baseline, and decide together whether to re-sign on a rolling monthly basis.
What actually happens in the first month?
Business deep-dive on margins and SKU roles first, then the account and feed audited together. Measurement gets corrected before anything else - attribution model, brand and generic separation, conversion tracking integrity - because every decision after that depends on the numbers being honest. Account hygiene is fixed in the same window and structural work begins.
How many clients do you take on?
Three to five. That's the number where I can be in every account properly rather than checking in on them. It's also the honest reason I sometimes can't start immediately.
You're one person. What happens if you're unavailable?
The client cap is the main protection - I'm not spread across twenty accounts, so nothing is running unattended by default.
Beyond that: you own the Google Ads account, the Merchant Centre and all the data throughout. Everything I build lives in your systems rather than mine - the label logic, the modelling sheets, the campaign structure. If the engagement ends, you get a documented handover another specialist can pick up. There's no version of leaving that costs you the work.
Can you work alongside our in-house team or another agency?
Yes, and often that's the right structure. If you have someone internal running other channels, or an agency handling Meta, I'll work to the same commercial targets rather than competing for credit on a blended number. I'm not precious about owning the whole thing.
What if it isn't working?
Nothing rolls over automatically. At the end of the initial term we both decide whether to continue, and if we do it's a rolling monthly arrangement from there. Before that point, the 30/60/90 checkpoints exist precisely so neither of us is guessing - we agree the baseline in month one and measure against it, rather than debating it later.
If I get three months in and think the constraint is somewhere I can't fix - your margins, your stock position, your site - I'll tell you that rather than keep billing.
Find out if your account is leaving profit on the table.
Most accounts I review are optimising to the wrong metric. Twenty minutes is usually enough to tell whether there's a structural problem and roughly what it's costing you.
- No obligation, no pitch deck. A direct conversation.
- An honest assessment, including if we're not the right fit.
- If a retainer isn't the right answer, I'll say so and tell you what is.