Paid advertising, with proof it’s actually creating demand
Ad platforms grade their own homework. They report conversions from people who were already going to buy from you, and the number looks like performance. We run paid campaigns across search, social and shopping, and we test whether they are generating customers or taking credit for customers you already had. Most agencies do not do the second part.
Campaign performance, tested properly
Paid campaigns
Search, social and shopping activity measured together.
Attribution
Separating platform claims from the customer journey.
Demand test
Testing whether advertising created customers or only claimed them.
Reported conversions
Customers actually created
The objective is not to make the ad platform’s report look good. It is to know whether paid media created additional demand.
The problem this solves
You will recognise the situation if:
Spend keeps rising and results do not.
The platform reports a strong return but revenue has not moved.
You cannot tell which campaigns produce customers rather than clicks.
Automated campaign types now control most of your spend and you have lost visibility into where it goes.
Leads arrive but the sales team says they are poor quality.
Your last agency reported cost per lead and nothing beyond it.
You suspect you are paying to reach people who would have found you anyway.
What we manage
Paid search
Google Ads and Microsoft Ads. Capturing demand from people already looking for what you sell.
Paid social
Meta, LinkedIn, and other platforms depending on where your buyers actually are. Creating demand rather than capturing it, which requires a different approach and different expectations.
Shopping and product advertising
For businesses selling products online: feed quality, product structure, and the campaign types that depend on both.
Video advertising
YouTube and in-feed video, where the work is mostly about creative and audience rather than bidding.
Retargeting
Reaching people who already engaged. Useful, frequently overrated, and one of the biggest sources of inflated performance reporting.
Automated campaign management
Performance Max, Advantage+ and their equivalents now control the majority of spend on the major platforms. They work, but they remove visibility. Managing them well means controlling the inputs — feeds, creative, audience signals, conversion definitions — because the levers that used to exist are gone.
Conversion tracking and attribution
Making sure what the platform counts as a conversion is something your business actually values. This is where most accounts go wrong, and it is why a campaign can report success while the business sees nothing.
Incrementality testing
Structured tests that hold campaigns back from a portion of your audience to see whether total conversions actually fall. If they do not, the campaign was claiming credit rather than creating demand. This is the part almost nobody sells, and it is the only way to answer the question every finance director asks.
Why incrementality matters more than ROAS
Worth understanding before you buy paid media from anyone.
When someone searches for your brand name, sees your ad, clicks it and buys, the platform records a conversion. It will show an excellent return. But that customer was already coming to you. The ad did not create the sale, it charged you for it.
Multiply that across brand terms, retargeting audiences and automated campaigns that optimise toward easy conversions, and you get accounts that report strong performance while the business sees no change in revenue.
Incrementality testing settles it. You withhold a campaign from a defined portion of the audience for a defined period, then compare total conversions between the groups. If the held-back group converts just as well, the campaign was not adding anything.
The results are sometimes uncomfortable. We would rather tell you a campaign is not working than keep billing you to manage it.
How we work
Diagnose
Account audit: what is being tracked, what the platform counts as a conversion, where spend is going, and which campaigns are likely claiming credit rather than creating demand.
Instrument
Conversion tracking corrected and baselines recorded before any spend changes. Without this, every later result is unverifiable.
Fix the foundations
Landing pages, tracking, feeds and campaign structure. Sending more traffic into a broken path is the most common way to waste a budget.
Build and scale
Campaign build, creative testing, audience development and bid strategy, prioritized by what the audit found.
Prove and compound
Ongoing optimization with incrementality testing built into the calendar rather than promised and never run.
How we measure it
Cost per acquisition
Cost per acquisition, and where possible cost per qualified acquisition rather than per lead.
Incremental conversions
Incremental conversions, tested rather than platform-reported.
Revenue or pipeline influenced
Revenue or pipeline influenced, where your systems allow the trace.
Lead quality
Lead quality feedback from your sales team, treated as data rather than anecdote.
Blended acquisition cost
Blended acquisition cost across all channels, not paid in isolation.
Wasted spend
Wasted spend identified and removed.
Common questions
01 How much should we spend?
It depends on your market, your margins and what a customer is worth to you. We would rather work that out from your numbers than quote a figure. Below that, the account cannot gather enough data to optimize.
02 How quickly will we see results?
Paid media moves faster than anything else we do. Early data within days, meaningful optimization within four to six weeks. Incrementality testing needs longer, usually a full cycle, because it requires a controlled comparison rather than a snapshot.
03 Do you charge a percentage of ad spend?
Percentage-of-spend models create an incentive to spend more.
If you use one, explain how you manage that conflict.04 Who owns the ad accounts?
You do.
[Confirm that accounts are created under the client’s ownership and that access and historical data remain with them if the relationship ends. Few agencies state this publicly and buyers care about it.]05 Can you manage campaigns in other countries?
Yes. Advertising rules, platform availability and buying behaviour differ by market, so we research each one rather than copying a structure across borders.
06 Our last agency reported great numbers but we saw no growth. What went wrong?
Usually the tracking, the conversion definition, or credit being claimed for demand that already existed. The audit will identify which. It is the most common thing we find.
07 Can you guarantee a return?
No. Anyone guaranteeing a return on advertising is either excluding the cases where it fails or does not understand the variables. What we commit to is correct measurement and an honest answer about whether it is working.