eCommerce PPC for Fashion: Category, Seasonality, Return Rates

2. Category structure - women's, men's, kids', accessories

The biggest mistake we see in fashion accounts is "The Big Bucket" - throwing all products into a single Performance Max campaign. While Google's AI is powerful, it needs distinct signals to work effectively. If you give Google 10,000 SKUs across five categories and one ROAS target, it will naturally gravitate toward the "easy wins"-usually high-volume, generic items with low margins. Separate campaigns per gender and age category are non-negotiable for four reasons:

• Different Bid Landscapes: The CPCs for women's occasion wear are vastly different from men's casual essentials. Mixing them leads to budget being sucked into the high-volume/high-CPC categories at the expense of efficient ones. Women's footwear might have a CPC of £1.20, while accessories might be £0.40. Mixing them creates a "bid-averaging" effect that hurts performance.
• Distinct Seasonality: Men's shopping cycles are often "need-based" and more concentrated around traditional peak periods like Christmas and Father's Day. Women's cycles are more "trend-based" with frequent micro-peaks throughout the year driven by social media trends and event calendars. Mixing these signals dilutes the algorithm's ability to predict demand spikes.
• Varying Profit Margins: Accessories often carry much higher margins than apparel. By separating them, you can set aggressive ROAS targets for apparel to drive volume (market share), while setting higher targets for accessories to protect margin. This "portfolio approach" to bidding is much more effective than a flat account-level target.
• Return Rate Divergence: As mentioned, return rates vary wildly. You cannot apply a single ROAS target across the account if one category returns at 10% and another at 40%. Bidding on an unadjusted basis means you are effectively penalising your most profitable (low return) categories.

To execute this effectively, you should use **Custom Labels** in your product feed. Custom Label 0 for Category (Women's, Men's, Kids'), Custom Label 1 for Sub-category (Dresses, Knitwear), and Custom Label 2 for Profit Margin (High, Medium, Low). This allows you to build a campaign structure that reflects the business's P&L, not just the website's navigation.

Recommended Budget Split: For a typical UK fashion brand, Women's usually takes 55-65% of the budget, Men's 25-35%, and Kids'/Accessories make up the remainder. This should be adjusted dynamically based on return-adjusted ROAS, not historical spend. If Men's is achieving a higher RA-ROAS, don't be afraid to shift budget mid-season.

3. Seasonality - 5 fashion seasons + budget allocation

Fashion does not follow the standard retail calendar where everything ramps up for Black Friday and dies in January. Instead, it operates on a "lead-and-lag" cycle across five distinct seasons. In the UK, the weather is the primary driver of fashion search volume. A warm week in February can trigger a massive spike in "spring dresses" even if the items aren't physically in stock yet.

Rule of thumb: Always launch 4-6 weeks before the peak demand window. If you wait until it's cold to launch a winter coat campaign, you've already lost the research phase where consumers build their "wishlists." Google's algorithm needs this time to learn which users are most likely to convert for your new collection.

4. Return rate impact on ROAS reporting

This is the "silent killer" of fashion PPC profitability. Most Google Ads accounts are set up to track conversions at the point of sale (thank you page). However, in fashion, a conversion isn't a conversion until the 30-day return window has closed and the customer has kept the item. In the UK, the culture of "wardrobing"-buying an outfit for a single event and returning it-is a significant burden on eCommerce margins.

Industry data from *Fashion United* and *Business of Fashion (BoF)* consistently shows that online apparel return rates in the UK hover between 30% and 40%. For specific categories like high-end footwear or evening gowns, this can climb to 50%+. This means that for every £100,000 in revenue Google Ads claims to have generated, only £60,000 - £70,000 is actually "banked" by the business.

If you treat every pound of revenue equally, you are lying to yourself and to the Google Ads algorithm. High-return items effectively have a "tax" on their conversion value. By not accounting for this, you end up over-bidding for items that don't actually contribute to the bottom line. Furthermore, high return rates often correlate with higher shipping and processing costs, further eroding the true ROI.

We've seen accounts where the "Best Seller" in Google Ads (based on revenue) was actually the "Worst Performer" after accounting for returns. The item looked like a hero but was a logistical nightmare that cost the company money on every sale. Transitioning to a return-adjusted model is the first step toward true profitability.

5. Return-rate-adjusted ROAS calculation

To fix this, we implement "Return-Adjusted ROAS" (RA-ROAS). This simple framework ensures you are bidding based on kept revenue, not gross revenue. This isn't just about reporting; it's about shifting the algorithm's focus away from "high-returners" and toward "high-keepers."

By calculating this at a category level, you can set "Actual ROAS" targets that align with your business goals. If your target net ROAS is 4x, and you know dresses return at 50%, you need to set your Google Ads target to 8x for that category. Use our interactive calculator below to see the impact of return rates on your own account.

**Action Step:** You should apply these adjustments per category. Don't use a blended account-level return rate. Instead, use Google's conversion value rules to automatically discount the reported revenue based on the product category. This feeds "cleaner" data back to the bidding algorithm, allowing it to find customers who are more likely to *keep* what they buy. You can also import "offline conversions" once the 30-day window has closed for a perfectly accurate view, though value rules are a more agile starting point.

6. Fashion-specific feed attributes

In most verticals, a product title and a price are enough. In fashion, Google's "Apparel & Accessories" vertical is one of the most strictly policed. If your feed is missing key attributes, you aren't just losing visibility - you are often paying more for lower-quality traffic. Google uses these attributes to place your ads in the "filters" on the Shopping tab. If a user filters for "Blue Silk Dress" and your feed doesn't specify the material or colour, you're invisible.

A deep dive into feed health can be found in our product feed fundamentals guide. For fashion, we recommend using a feed management tool (like Channable or Feedonomics) to dynamically append these attributes from your Shopify or Magento product descriptions. Don't rely on the "default" feed export; it's almost always insufficient for a competitive fashion landscape.

7. Fashion PMax asset groups

Performance Max is the primary driver for most fashion accounts. However, the biggest lever isn't the bid-it's the **creative assets**. In fashion, the visual representation of the product is everything. Consumers aren't just buying a piece of fabric; they are buying an aesthetic and a feeling. If your assets look like a warehouse inventory sheet, your performance will reflect that.

The CTR Gap: Lifestyle vs. White Background

Our internal data shows that lifestyle or "lookbook" style imagery consistently outperforms flat-lay or white-background shots by **25-30% in CTR**. While white-background images are often required as the main image in the feed (to satisfy Google's strict policies), your PMax asset groups should be dominated by high-quality lifestyle photography that shows the product in its "natural habitat."

• Image Assets: Include shots of the garment "in motion" and on different body types if possible. Use "lifestyle_image_link" in the feed to provide Google with more context. High-contrast, well-lit shots work best for the small thumbnails on mobile Discovery.
• Video Assets: Short, 15-second vertical videos of models walking or showing the fabric's movement are highly effective for YouTube and Discovery placements. Don't use corporate-style videos; use "user-generated" (UGC) style content which feels more native to the platforms PMax serves.
• Audience Signals: Don't just target "People interested in fashion." That's too broad. Use your first-party data. Create audience signals for "Top 10% Lifetime Value Customers" and "High-Return Shoppers." Use the latter as an exclusion or a low-bid segment to avoid "serial returners" who drive up costs.

Remember: PMax is a "black box," but you control the inputs. If you give it average assets, it will find average customers. If you give it "lookbook-standard" assets, it will find the users who are willing to pay full price.

8. Fashion Shopping titles

Your product title is the single most important piece of text in your Shopping Ads. It's the primary factor Google uses to determine which searches your ad should appear for. In fashion, users search with specific intent. If your title is just "Silk Dress," you are competing with everyone from fast-fashion giants to luxury houses. If it's "Women's Navy Silk Midi Dress - Size 8-16 - Visionary Brand," you are capturing high-intent traffic that is much more likely to convert.

Testing shows that including the size range in the title can increase CTR by over 30% because it immediately answers the user's most critical question: "Will it fit me?" If a user is a size 14 and sees "Size 8-16" in the title, they click with confidence. If they don't see it, they might skip your ad for one that provides that clarity. This reduces wasted clicks and improves your "Quality Score" over time.

For multi-pack items (e.g., socks or basic tees), always include the pack count in the title. "Pack of 3" or "3-Pack" is a major driver of value perception in fashion basics.

9. Common fashion PPC mistakes

Avoid these five common pitfalls that plague even seasoned UK fashion brands. These are the "silent margin killers" we look for first during an audit.

1. Treating all categories the same: Blending high-return categories (dresses, shoes) with low-return ones (socks, bags) into a single ROAS target. This leads to the algorithm over-serving the high-revenue/high-return items and starving the truly profitable ones.
2. Ignoring the seasonal lead-time: Starting your summer campaigns in June when the "research" phase began in April. In fashion, if you aren't top-of-mind during the research phase, you're just a commodity during the purchase phase.
3. Weak lifestyle imagery: Relying on dry, manufacturer-provided white-background shots that fail to convey "style." Fashion is an emotional purchase; your creative must reflect that.
4. Over-reliance on Brand: Letting PMax claim credit for branded searches instead of using Standard Shopping or Search to protect those terms at a lower CPC. PMax will always take the "path of least resistance," which is usually your existing customers.
5. Short product titles: Failing to include size, colour, and material. This leads to broad, irrelevant clicks that don't convert because the user's specific requirements (e.g., "size 12") weren't met.

By fixing these, you can often see a 20-40% improvement in account efficiency within the first 60 days. It's not about working harder; it's about making the algorithm work smarter by providing it with better signals and better structural boundaries. Our eCommerce PPC attribution guide provides further detail on how to measure these improvements accurately beyond the basic Google Ads dashboard.