TL;DR: Shopping CPCs rose 19% YoY in Q2 2026. Beauty is up 23%. Your platform-reported ROAS is likely 2-5x higher than your actual incremental return. If you are not running a blended CAC ceiling, you are subsidizing Google's revenue growth with your margin.

The CPC Reality in 2026

Shopping ads are still the cheapest paid acquisition channel. The average Shopping CPC sits at $0.66, 87% cheaper than Search. That number sounds comforting until you realize it went up 19% year over year in Q2 2026.

Beauty brands absorb the worst: 23% YoY CPC inflation. Apparel and home goods are not far behind. Platform-reported ROAS flatters you. Cassandra's analysis of 253 media mix models across $383M in spend found that platform-reported ROAS runs 2-5x higher than true incremental impact.

Category Benchmarks

| Category | Avg CPC (Q2 2026) | YoY Change | Avg ROAS | Avg CVR | |---|---|---|---|---| | Apparel | ~$0.61 | +19% | 6.1x | ~2.1% | | Beauty | ~$0.72 | +23% | ~4.8x | ~1.9% | | Home Goods | ~$0.63 | +19% | ~4.2x | ~1.7% | | Electronics | ~$0.88 | +15% | 3.8x | ~1.6% | | Overall Avg | $0.66 | +19% | varies | 1.91% |

Performance Max now accounts for 62% of Shopping spend and 61% of attributed sales. But Standard Shopping still has a lower CPA: $38.87 versus $43.91 for PMax. That $5 gap matters at scale.

Desktop outperforms mobile by 24% on conversion rate. Your blended 1.91% masks a significant device-level performance split.

Why Per-Channel ROAS Lies

Every platform measures attribution in its own favor. Google counts a conversion if the user clicked a Shopping ad within the attribution window, regardless of whether your email or brand loyalty drove the actual decision.

PMax bundles Search, Display, YouTube, Gmail, and Discover into a single campaign type. When the blended result looks good, most brands increase budget. They do not ask which placements within PMax are driving returns.

Logical Position and Tinuiti have published guidance recommending tROAS floors inside PMax. Without a floor, PMax optimizes toward volume at the expense of efficiency.

The Blended CAC Ceiling Method

Step 1: Pull your true blended CAC weekly. Tools like Triple Whale and Northbeam pull spend across channels and match against new customer orders using first-party data.

Step 2: Set the ceiling. If your AOV is $85, gross margin 55%, and you need a 6-month payback, your first-order CAC ceiling is roughly $23-28.

Step 3: Map channel-level spend to the ceiling. If Shopping claims $18 CPA but your blended CAC is $47, Shopping is not performing as well as it appears.

Step 4: Adjust bids based on blended reality, not platform reports. When Shopping CPCs rise 19% and your blended CAC starts approaching the ceiling, reduce target ROAS, cut low-performing product groups, or shift budget.

Q4 Prep

Shopping CPCs spike 25-30% during the Black Friday window. On a $0.66 base, peak CPCs hit $0.83-$0.86.

Audit your product feed now. Title structure, attribute completeness, and image quality determine your Quality Score equivalent.

Model your Q4 CAC ceiling explicitly. Higher CPCs, higher conversion rates, higher AOV from gifting, higher return rates in January. Build a Q4-specific ceiling.

Decide PMax vs Standard Shopping for Q4. Given the $5 CPA gap, there is a real argument for Standard Shopping on highest-margin product groups during peak spend periods.

Data's DNA Applied

The 19% CPC increase is a signal about market structure. Beauty CPCs up 23% tells you brands in that category are fighting for the same customers with nearly identical products and are resolving competition through bidding rather than differentiation.

The $38.87 Standard Shopping CPA versus $43.91 PMax CPA is a data point about control. Standard gives you more control. PMax gives Google more control. The $5 gap is the cost of that control transfer.

The Product Feed Audit That Pays for Q4

Most ecommerce brands treat their product feed as a technical requirement rather than a strategic asset. That is a mistake that costs real money in a CPC environment that is 19% more expensive than last year.

Your Shopping performance is increasingly feed-quality dependent. Google's Smart Bidding relies on product data signals to determine which products to show, to whom, and at what bid. A feed with incomplete attributes, generic titles, and missing GTINs will consistently lose auctions to competitors with richer data.

Here is the audit:

Titles. Your product title is the single highest-asset field in the feed. The format that performs best across most categories: [Brand] + [Product Type] + [Key Attribute] + [Size/Color]. Example: "Patagonia Better Sweater Fleece Jacket Men's Navy Large" beats "Men's Jacket" by 200-300% in impression share.

Product Type. Use the most specific Google product taxonomy category available. "Apparel > Outerwear > Fleece Jackets" outperforms "Apparel" by giving Smart Bidding more signal about user intent.

Images. Primary image on white background, minimum 800x800 pixels. Lifestyle images in additional image fields. Brands running multiple image variants see 5-15% higher CTR across most Shopping formats.

Custom Labels. Tag products by margin tier, seasonal relevance, and inventory status. Use custom labels to create product groups in your Shopping campaigns that align with your blended CAC ceiling. High-margin products can tolerate higher CPCs. Low-margin products need tighter controls.

Availability and Pricing. Out-of-stock products with active bids waste budget. Price competitiveness data from Google Merchant Center's pricing report shows where you are above or below market.

Run this audit in July. Test improvements through August. By September, your feed is optimized for the Q4 auction.

The brands that treat the product feed as a strategic asset in a rising-CPC environment are the ones that maintain profitability while competitors chase diminishing returns. The feed is not a technical checkbox. It is the foundation of your Shopping economics.

Doctrine Connection

> Verification beats optimism. The 19% CPC increase is not just a number to react to. It is a symptom of a market where differentiation collapsed and brands compete on bid rather than value. The blended CAC ceiling stops you from asking "what is my ROAS on Shopping?" and starts you asking "what is the actual cost of acquiring a new customer across my entire business?"

Q: Should I switch from PMax to Standard Shopping?

Not necessarily across the board. Run Standard Shopping for highest-margin, highest-conviction product groups. Use PMax for broader discovery. The $5 CPA gap is a real cost, but it reflects a tradeoff in control versus scale.

Q: How do I calculate a blended CAC ceiling without a mature LTV model?

Start with contribution margin on first order. If your average order delivers $40 in gross profit, your hard ceiling is $40. For sustainability, acquire at 50-70% of first-order contribution margin. Rough ceiling for a $40 contribution margin business: $20-28 per new customer.

Q: Will CPCs keep rising through 2026?

The structural factors are not going away. More brands entering paid, AI-generated creative reducing differentiation at the ad level, and Google's push toward automated bidding all point toward continued pressure.