Fashion · Ecommerce
Revenue was falling. Q4 was six weeks away.
A fashion brand heading into peak trading on a year on year decline, with little room to increase budget. Rebuilding Shopping and Search before peak turned that into £884,000 of extra revenue.
The situation
A fashion ecommerce brand was in year on year revenue decline, heading into the quarter that decides its year. Budget flexibility was limited, so there was no option to simply buy the way out of it. Whatever was going to change had to come from structure, and it had to be in place before demand arrived.
The timing is the hard part. Q4 is the worst possible moment to rebuild an account. Do it in October and you are learning during the only weeks that matter. The work started in September so the structure was settled before peak began.
You cannot rebuild an account during peak. Whatever structure you go into October with is the structure you are stuck with until January.
What I did
Four changes, all completed before peak trading opened.
Rebuild Shopping and Search before peak
Campaign structure was reworked in September specifically for peak trading conditions, so October opened with an account already pointed at the right products rather than one still learning.
Bid to live stock levels
Bidding adjusted in real time against stock. In fashion, sizes sell out unevenly, and paying peak CPCs to advertise a line you cannot fulfil is the fastest way to waste a Q4 budget. This keeps spend behind what can actually ship.
Time promotions instead of running them constantly
Promotional cadence was planned around the peak calendar rather than always on discounting, so offers landed when demand was highest and margin was not given away in the quiet weeks.
Put budget behind the rebuilt structure
Spend rose about 58% across the quarter. Because it went into a structure already proven, the extra £196,000 of spend returned £880,000 of extra revenue, a marginal return of roughly 4.5x.
Worth being upfront about
Return on ad spend only moved from 4.29x to 4.36x, a gain of about 2%. This is not an efficiency story and it would be dishonest to sell it as one. Spend rose roughly 58% and revenue rose 61%. The achievement is reversing a year on year decline in the quarter that decides the year, and holding a 4.3x return while adding nearly £200,000 of spend into it. Efficiency usually drops when you scale that hard into peak. Here it held.
Year on year, Q4
What made the difference
The work happened in September. Every change was live before peak opened, so October to December was spent trading, not testing.
Want a read like this on your own account.
Book a 30‑minute call. Bring your current numbers and leave with a clear view of where things stand.
Message sent.
Thanks, that's landed with Jay directly. Expect a reply within one business day.
Prefer email? jay@mediaking.ai