Case study · Google Ads for ecommerce
How we compounded a deals ecommerce brand's Google Ads revenue 68x in seven years while holding blended ROAS above 13x
Seven years of a Google Ads programme for a deals-and-discounts ecommerce brand. Annual attributed revenue grew from £55k in Year 1 to £3.7m in Year 7 (68x). Blended seven-year ROAS held at 13.8x across an account that scaled its media spend 111x in the same period.
- Annual Google Ads revenue £3.7m (Year 7) from £55k (Year 1) 68x
- Annual media spend £342k (Year 7) from £3k (Year 1) 111x
- Blended 7-year ROAS 13.8x from — Across the whole 7-year window
- Cumulative Google Ads revenue (7 years) £8.2m from — From account launch
The client
A deals-and-discounts ecommerce marketplace, running a broad daily-rotating product catalogue targeted at price-sensitive shoppers. Search and Shopping are the dominant paid channels; the buyer journey is short and Google Ads sits at the top of the funnel doing most of the acquisition work. This case study covers the Google Ads programme specifically, isolated from the rest of the digital mix so the shape of the ad account is legible on its own.
The problem
New Google Ads account at launch, no historical conversion data, and a catalogue that rotates daily. Deals marketplaces are one of the more punishing categories on Google Ads: the range is wide, the offers are transient, the price sensitivity of the audience is high, and any inefficiency in feed hygiene compounds directly into wasted spend. Scaling the account from a small pilot to a seven-figure annual paid programme without letting ROAS collapse into the mid-single digits was the long game.
The plan: land the fundamentals, then scale the winners aggressively
Seven years of Google Ads discipline, layered continuously and re-tuned every year as Google's ad product surface evolved.
-
Feed hygiene as a permanent workstream
The single biggest lever on Shopping ROAS. Product titles rewritten for search intent, category tree aligned to Google's taxonomy, custom labels added for merchandising signals (price bands, promo status, category tier). Continuous, not a one-off setup, because a deals catalogue rotates daily and stale feed data leaks money.
-
Campaign structure evolving with Google's product surface
Campaigns split by intent cluster in Year 1: brand, category, product-specific, competitor. Shopping campaigns kept structured separately from Standard Search so bidding could learn each cluster independently. Migrated portions of the account to Performance Max as Google introduced it, without letting PMax cannibalise Standard Shopping revenue as it typically does on other accounts.
-
Negative-keyword hygiene from Day 1
Weekly search-term reviews from Week 1 of the account. Cumulative negative-keyword lists now number in the hundreds, covering irrelevant categories, competitor terms outside the auction we want to fight, and non-buying-intent research queries. Cumulative spend saved from irrelevant impressions on this account over seven years counts into the six figures in GBP.
-
Bidding-strategy migration as conversion data thickened
Manual CPC in early Year 1 while conversion data was thin. Maximise Conversions with CPC ceilings from Year 2. tROAS bidding from Year 3 as conversion volume passed the 100-per-campaign threshold Google's algorithm needs. Consistent target ROAS discipline year after year, tightened as the account grew.
-
Continuous ad copy and asset testing
Responsive Search Ads refreshed on a quarterly cadence. Headline pinning strategies tied to the highest-CTR anchor phrases per cluster. Ad-strength kept in Excellent range across the account, which lifted auction share at no CPC cost. Shopping product imagery reviewed periodically against category peers.
-
Reporting tied to revenue, not clicks
Monthly reports built around attributed revenue and ROAS by campaign and category, with clicks and impressions as supporting context, not headline metrics. Quarterly strategic reviews with the client focused on where next quarter's media budget was going and why.
The trajectory (seven full trading years)
Annual attributed revenue lifted 68x from Year 1 to Year 7. Media spend rose 111x in the same window. Blended seven-year ROAS across every rand spent through the account: 13.8x.
Source: Google Ads
Sixty-eight times more revenue in seven years, with the shape you want in a Google Ads programme: spend and revenue climbing together, never a year where the account plateaued or gave back ground. The compounding effect is clearest from Year 4 onwards, once the fundamentals, feed hygiene, negative-keyword discipline, bidding structure, had bedded down long enough to survive step-changes in scale. Every year of the seven landed higher than the last on attributed revenue. This is what a Google Ads programme looks like when it is run continuously and re-tuned every time Google evolves the product surface.
Scaling your ecommerce Google Ads account?
Peak Digital runs Google Ads programmes for ecommerce brands where the media budget is scaling faster than the account structure was originally designed to handle. Book a free introductory call and we will tell you honestly whether your account can hold ROAS through the next stage of growth.
Book a free introductory call