Example Outdoor Co.

The curve bends but doesn't break. Your four lowest-spend weeks (~£12.1k/wk) averaged 5.02x ROAS; your four highest (~£17.4k/wk) averaged 4.18x, a 17% drop across a 1.4x spend range.

• Shopping | Waterproofs is capped while winning. It came in above its 4.0x target in 9 of 12 weeks and was limited by budget in 7 of those, losing 31% of impressions to budget. That's the clean "efficient and capped" headroom. Lifting its daily budget from £352 to £510 is the growth move.
• Generic Search has a ceiling. Under £3,200 a week it returns 3.82x; above it, 2.90x. Hold it there until ad and landing page relevance improve.
• PMax is the stronger home for marginal non-brand budget at 4.26x against non-brand Search at 3.61x.
• Treat the last week or two gently: conversions attribute to click date, so recent ROAS will rise as the window closes.

Demand for your keyword set is up 6.1% year on year and the trend reads GROWING. 94 of your top 100 keywords by spend returned Planner data.

• Waterproof jackets are the strongest segment (+11% YoY); fleeces and mid layers are softer (-4%).
• Clear seasonality: volume lifts from late September, peaks in November, and dips in May and June.
• Competition is moderate: 31 HIGH, 44 MEDIUM, 19 LOW.
• Indicative market context: outdoor apparel online has been growing at low-to-mid single digits. Cross-check against a current sector report before building a plan on it.

Short answer: you have room to scale, and it's in one place.

• The headroom is real. Demand is growing, and Shopping | Waterproofs has been capped while beating target for most of the period.
• Fix measurement before you scale. With purchases double counted, the account looks on target when it isn't. Scaling on a flattered number buys returns you think you're getting.
• Hold Generic Search at its ceiling and push new budget into Waterproofs and PMax first.

Spend has grown steadily and efficiency has mostly held. Last 12 months: £712,400 spend against £598,000 the year before, ROAS 4.62x against 4.95x.

• Build (Oct 2023 to Mar 2024): Search and Shopping only, ~£38k a month at 5.2x.
• PMax launch (from 14 Apr 2024): spend up 64%, ROAS settled lower at ~4.7x.
• Rising costs (Jun 2026 onwards): CPC up 19% year on year; ROAS holding at ~4.7x reported.
• 2 October 2026: reported conversions rose 8% overnight with no change in clicks or orders on site. That's the date the GA4 purchase import was set to primary (see section 4.2).

Spend-weighted average QS is 6.4, and 18.2% of non-brand spend sits on keywords scoring 4 or below.

• Generic | Walking Boots is the one to fix: £6,210 at a spend-weighted QS of 4.8, with 70% of its keywords landing on the all-footwear page.
• Landing page experience is the weakest component here; ad relevance is the weakest account-wide.
• Brand runs at 9.6. Nothing to do there.

No over-pinning, but the copy repeats itself and six ads rate POOR.

• 11 of 42 ads reuse the same eight headlines. Free returns and your 365-day guarantee appear in only 40% of ads.
• Three PMax asset groups have no video, so Google is generating one for you.
• Rebuild the six POOR ads first, add the returns and guarantee lines everywhere, and upload one short video per asset group.

300 unactioned search terms, with £4,120 on terms that produced zero conversions.

• Repairs and second-hand intent: "jacket zip repair" £740, "used walking boots" £880. Negate "repair", "used" and "second hand".
• Rental: "ski jacket hire" £520. Negate "hire" and "rental".
• Jobs: "outdoor shop jobs" £610.
• Promote the winners: "packable waterproof jacket" and "wide fit walking boots" convert as unmatched queries. Add them as exact keywords.

Two shared lists (310 terms) are in place and well kept.

• Brand terms are excluded from Generic Search but not from PMax Core, where 6.8% of search-term spend is on your own brand name. Apply a brand exclusion to both PMax campaigns.
• Apart from "kids" (covered in the keyword blocker check), nothing is over-reaching.

• Walking Boots (£6,210) lands on /footwear rather than /footwear/walking-boots.
• Women's Waterproofs (£4,880) lands on /womens rather than the waterproof jackets category.
• Fixing both cuts a navigation step for people who've told you exactly what they want.

• Five live campaigns: Brand, Generic, Shopping | Waterproofs, PMax Core and PMax Accessories.
• Waterproof jackets sit in both Shopping | Waterproofs and PMax Core, and PMax wins most of those auctions at a lower return (4.72x against 5.17x). Exclude waterproofs from PMax Core, or merge them into one campaign with one target.

• 6 Oct: PMax Core tROAS moved from 4.5x to 4.2x. Spend rose 14% and ROAS fell, as expected.
• 13 Oct: 22 negatives added to Generic. Neutral.
• Shopping | Waterproofs: budget unchanged despite being limited by budget for most of the period. That's the missed opportunity.
• PMax Accessories has been under target for 12 weeks with no changes made.

• 640 products took £14,800 of feed spend with zero conversions, concentrated in accessories and end-of-line colours.
• Custom labels are empty on 100% of products, so you can't bid by margin band, bestseller status or season.
• Titles average 52 characters; 1,240 products are missing gender and 410 are missing colour.
• Fix the price sync, add custom labels, then rebuild titles as Brand + Product type + Gender + Key feature + Colour.

GTM is live (container GTM-XXXXXXX), running client-side. That works, but it's where conversions most often go missing: ad blockers, Safari's cookie limits and consent choices typically cost 10% or more. Moving to server-side GTM on a first-party subdomain is the priority measurement project, ideally after peak.

Two purchase actions are in the Conversions column, and they count the same orders.

• Purchase (Google Ads tag): 7,733 conversions, £822,020 value over 90 days.
• Purchase (GA4 import): 612 conversions, £61,290 value, set to primary on 2 October.
• From 2 October both have recorded the same orders, so reported ROAS of 4.74x falls to 4.41x when one is removed: under your 4.5x target.
• Set the GA4 import to secondary. Smart Bidding has been learning from inflated results for three weeks.

• Of 5 campaigns, 3 pass cleanly.
• PMax Accessories: 2.41x against a 4.0x target, 12 weeks running. Its product mix can't reach the target; restrict it to products over £40 or pause it.
• Shopping | Waterproofs: above target and budget-limited. Raise the budget, not the target.

Enhanced Conversions for web is on and customer data terms are accepted. Diagnostics show a healthy match rate on the Google Ads purchase tag. Nothing to fix.

No value rules are active. Accessories carry about half the margin of jackets, so revenue-based bidding over-values them. A value rule reducing accessories is a quick interim fix until margin-based values are in place.

• Fix the double count first. Everything else is judged on a flattered number until it's done.
• Enhanced Conversions is working. Server-side GTM is the next measurement project, after peak.
• Bid strategies are set sensibly apart from PMax Accessories, which can't hit its target.

ROAS is 4.41x against your 4.5x target once double counting is removed.

• CVR (3.67%) is the biggest lever: walking boots and women's waterproofs land on the wrong pages, and 21% of products never show.
• CPC (£0.82) is second: up 19% year on year, driven by Competitor A and PMax's Display share.
• CTR (2.50%) is healthy; copy fixes are second order.

The account doesn't need rebuilding. It needs its measurement fixed and its best campaign let off the lead.

• Remove the duplicate purchase action from the Conversions column today.
• Lift the Shopping | Waterproofs budget from £352 to £510 a day.
• Restrict PMax Accessories to products over £40, or pause it.
• Get peak in place: promotions, sale prices and a seasonality adjustment for 27 to 30 November.
• Hold Generic Search at about £3,200 a week until relevance improves.