Forward deployed engineers for e-commerce and marketplace ops. Pricing, marketplace offboarding, CX, paid media, CDP — built, shipped, and operated. See the proof →
This is one capability within our forward deployed model. See our live engagements →
Marketing Analytics

We found £2.9M hiding in one client's marketing data

We dig into your channels, find the hidden ROAS, plug the coupon leakage, and fix the attribution gaps. Then we build the automation to keep it running. This is a capability we deploy, not a report we hand over.

190 Tools Unified | 3.2x ROAS Improvement | Real-Time Attribution | £340K MarTech Savings

The Problem

Marketing teams optimize what they can measure — but most measurement is wrong. Brand search inflates ROAS. Email open rates are 86% fake (Apple Mail Privacy Protection). SMS attribution is over-counted. And nobody is measuring how much margin they are giving away through coupon leakage to customers who would have bought anyway.

The real opportunities are invisible without deep cross-channel analysis. A headline ROAS of 13.9x tells you nothing about where the incremental revenue is actually coming from, which categories are starved of budget, or how much of your discount spend is pure waste.

We dig in, find the gaps, and fix them. Then we build the automation to keep it running.

Paid Media

What We Found in Google Ads

£874K spend £12.1M revenue 13.9x blended ROAS

We audited a 90-day Google Ads programme: £874K spend generating £12.1M in revenue — a headline 13.9x ROAS. But the headline number is misleading. Breaking it down reveals where the real value sits and where budget is being wasted.

Channel Spend Share ROAS What We Found
PMax Campaigns 95.2% 11.9x Workhorse — consistent, scalable
Brand Search ~3% 74.5x Inflated — these customers were coming anyway
Non-Brand Search ~1.8% 7.4x True incremental demand generation
Blended 100% 13.9x True non-brand: 11.8x

The real measure is non-brand ROAS at 11.8x — still 2.9x above the 4.0x breakeven threshold. But the headline 13.9x was masking the real picture and preventing informed budget allocation.

Category Analysis

The Category Nobody Was Investing In

22.7x ROAS 2.6% of spend Budget capped at £150/day

The biggest finding was buried in category-level data. One product category representing just 2.6% of ad spend was generating a 22.7x ROAS — more than double the account average. Subcategories within it were hitting 25.4x. Budget was capped at £150/day.

Meanwhile, premium brand campaigns were running at 5.7-6.9x ROAS, consuming disproportionate budget. Simple reallocation unlocked significant incremental revenue. This is the kind of insight that only surfaces when you move beyond headline metrics and analyze at the category level.

Omnichannel

What We Found in Email & SMS

32M
Emails Delivered
90-day period
£196K/mo
Journey Revenue
From automated email journeys we built
39x
SMS vs Email
SMS attributed revenue vs email: £7.8M vs £133K
3.4x
Conversion Lift
Dual-channel subscribers vs email-only

The headline email open rate was 48.4% — but 86-88% were Apple Mail Privacy Protection fakes. The true reliable metric was click rate: 2.77%. Saturday sends outperformed Wednesday by 2.4x on click rate. Automated journeys (welcome, abandoned cart/browse, coupon nudge) that we built generated £196K/month — more reliably attributable than campaign blasts.

SMS delivered 39x more attributed revenue than email (£7.8M vs £133K in 90 days), but click-through rates were declining year-on-year (9.8% in 2023 to 3.7% in early 2026) — clear channel fatigue. We negotiated SMS costs from 5p to 2.4p per message, saving £392K/year while the channel still performed.

Dual-channel subscribers (email + SMS) converted at 22.4% vs 6.7% for email-only — a 3.4x lift. Peak conversion window was 6-12 hours after click, not immediate.

Margin Protection

£2.9M in Unnecessary Discounts We Stopped

£727K per quarter 17.6% suppress segment leakage

The CDP we built revealed that 17.6% of customers classified as "full-price buyers" were still using coupon codes. These customers buy regardless of discounts — the coupons were pure margin giveaway.

At £727K per quarter, this single insight pays for the entire analytics infrastructure many times over. The fix is CDP-driven discount suppression — identifying price-insensitive customers through RFM scoring and purchase history, then excluding them from promotional campaigns. The margin recaptured drops straight to EBITDA.

What We Built

190-Tool Marketing Automation

190 tools 15+ platforms

For a DTC health and wellness client, we built a 190-tool marketing automation system spanning 15+ platforms — Google Ads, Analytics, Search Console, Meta, Shopify, email, SMS, push notifications, affiliate networks, subscription management, CRM, and more. Every platform connected through a unified interface.

Daily performance monitoring runs automatically. Weekly creative refresh cycles generate new ad variants based on performance data. This is not a dashboard — it is an operator. The marketing automation agent can query any platform, trigger any action, and surface anomalies before they become problems. We built it and we run it.

Results

What We Delivered

Cross-channel analysis Data-verified
13.9x
ROAS
Blended Google Ads, 11.8x non-brand
£2.9M/yr
Leakage Stopped
Coupon leakage identified and addressed
190
Tools Unified
Marketing automation across 15+ platforms
22.7x
Hidden ROAS
Underinvested category we discovered

These results come from deep analysis across two engagements — a major UK e-commerce retailer and a DTC health and wellness brand. The approach is the same every time: go beyond headline metrics, break down performance by channel and category, identify the gaps between attributed and incremental value, and find the margin being given away unnecessarily. Then build the automation to keep it running.

Frequently Asked Questions

What is the real ROAS of our Google Ads?

Almost certainly different from your headline number. Brand search inflates ROAS by 3-10x because those customers were coming to you anyway. True incremental ROAS requires separating brand from non-brand, analyzing category-level performance, and calculating your breakeven threshold. Most businesses are over-investing in premium segments and under-investing in high-performing niches.

What is coupon leakage?

Coupon leakage is when discount codes are given to customers who would have purchased at full price. If 17% of your full-price segment is using coupons, that is pure margin loss. The fix is CDP-driven discount suppression — identifying price-insensitive customers and excluding them from promotional campaigns. This typically recovers 2-5% of revenue as recaptured margin.

Are email open rates reliable?

No. Since Apple Mail Privacy Protection launched, 86-88% of email "opens" are machine-generated. Click rate is now the only reliable engagement metric. If your marketing team is optimizing for open rates, they are optimizing for noise. Focus on click rate, revenue per email, and automated journey performance.

How did you build the 190-tool marketing system?

Platform by platform, API by API. Each integration connects to a unified orchestration layer that can query any platform, trigger any action, and monitor any metric. We build it incrementally — start with the highest-value platforms (ads + analytics + CRM), then expand. Total build time: 3-6 months for comprehensive coverage. Then we stay and run it.

Is SMS marketing still worth the cost?

Yes, but it is declining. SMS attributed revenue is typically 10-40x higher than email, but click-through rates decline year-on-year as the channel saturates. Negotiate your per-message costs aggressively (we cut a client's from 5p to 2.4p — £392K annual saving), and focus SMS on high-intent triggers (abandoned cart, subscription renewal) rather than broadcast campaigns.

Your marketing data is hiding seven-figure opportunities.

Every marketing programme has hidden ROAS, wasted discount spend, and misattributed channels. We find the gaps, fix them, and build the automation to keep them fixed.