Google Performance Max for D2C Brands: How to Run It Without Wasting Budget

Google Performance Max is the most powerful campaign type available to D2C brands right now. It's also the most misunderstood. In the two years since PMax replaced Smart Shopping, the pattern has been consistent: brands either avoid it entirely ("it's a black box") or launch it with no creative assets, no audience signals, and a ₹500/day budget — then declare it doesn't work when the algorithm spends everything on brand searches and low-intent Display placements.

PMax works. But it requires a fundamentally different approach from traditional campaign management. The algorithm doesn't respond to micro-management — it responds to signals, creative variety, and enough data to learn. This guide breaks down exactly what you need to make PMax profitable for a D2C brand in India.

1. What Performance Max Actually Is

Performance Max is a single campaign type that runs across all seven Google inventory channels simultaneously: Search, Shopping, Display, YouTube, Gmail, Maps, and Discover. You provide the creative assets, audience signals, and conversion goals — Google's algorithm determines where and when to show your ads to maximise conversions within your budget.

PMax replaced Smart Shopping in September 2022. The key differences:

The "black box" criticism is partially valid — you don't get the granular search term reports you get from Standard Search campaigns. But PMax gives you more placement breadth than any other campaign type. The brands winning with PMax have learned to feed the algorithm correctly rather than fighting its structure.

2. Why Most D2C Brands Run It Wrong

After auditing dozens of D2C accounts, the same mistakes appear repeatedly:

• No asset groups, or one generic asset group for all products. PMax performs best when asset groups are tightly themed — one asset group per product category, each with tailored headlines, descriptions, and images. Running a single asset group for "all products" means Google is mixing signals from skincare, apparel, and accessories — the algorithm can't distinguish what works for which buyer.
• No audience signals. Without audience signals, PMax starts cold. It has no idea who your customers are. It wastes 2–3 weeks in a learning period that could have been cut in half by providing your customer list, website visitors, and custom intent audiences upfront.
• Budget too low to learn. PMax needs a minimum of 50 conversions per month to learn effectively. If your daily budget generates fewer than 2 conversions per day, PMax can't optimise. Budget below ₹1,000–1,500/day for most D2C categories is often insufficient for meaningful learning.
• tROAS set too high too early. Setting a target ROAS of 800% on day one is asking the algorithm to be conservative before it knows your business. Start at 300–400% and raise it gradually as the campaign stabilises.
• No brand exclusions. PMax will spend on people searching your brand name — capturing conversions that would have happened organically. This inflates ROAS artificially and eats into budget that should be working for new customer acquisition.
• No creative assets beyond the product feed. A PMax campaign running only a product feed is essentially Smart Shopping. The power of PMax comes from video, lifestyle images, and text assets that run on YouTube and Display. Without them, you're leaving the most efficient placements empty.

3. Asset Groups — The Creative Foundation

Asset groups are the building blocks of PMax. Each asset group is a collection of creative assets (headlines, descriptions, images, videos, logos) combined with a product group from your feed. Google assembles ads from these assets dynamically based on the user, placement, and context.

How to Structure Asset Groups for D2C

Don't create one asset group for your entire catalogue. Structure by product category or audience intent:

Asset Checklist Per Asset Group

• Headlines: Minimum 5, ideally 15. Mix benefit-led, feature-led, and CTA-led. "Free shipping on orders over ₹499", "Rated 4.8★ by 12,000 buyers", "New arrivals — shop now".
• Descriptions: Minimum 2, ideally 5. 60–90 characters. Expand on the headline — explain the benefit or overcome an objection.
• Images (landscape 1.91:1): Minimum 3. Lifestyle images, not just product-on-white. People using the product. Real environments. High emotion.
• Images (square 1:1): Minimum 3. Product close-ups, lifestyle crops, before/after where relevant.
• Logo: Your brand logo in square format. Non-negotiable.
• Video: At minimum a 15–30 second product or brand video. If you don't have video, Google auto-generates one from your images — which is almost always poor quality. Always provide your own video.

4. Audience Signals — The Most Underused Lever

Audience signals are the most powerful and most neglected feature in PMax. They tell Google's algorithm who your customers are — so it can find more people like them faster. Without signals, PMax starts from zero. With strong signals, the learning period compresses and ROAS improves significantly in the first 3–4 weeks.

Audience signals are suggestions, not restrictions. Google uses them as starting points but will expand beyond them. The stronger your signal quality, the tighter the initial targeting and the faster the algorithm learns.

Signal Priority Stack for D2C Brands

1. Customer match list (highest priority) — Upload your customer purchase history. Even 500–1,000 emails is valuable. Google uses this to find similar buyers. Update it monthly.
2. Website visitors — purchasers (last 180 days) — People who completed a purchase. The algorithm models from these.
3. Website visitors — cart abandoners — High-intent non-converters. Important signal for the retargeting asset group.
4. Custom intent audiences — Build audiences around search terms your buyers use: "[product category] buy online India", "[brand competitors]", "[product use case]". These are strong signals for new customer acquisition.
5. In-market audiences — Google's built-in audiences: Online Shoppers, Apparel, Beauty, Home Decor — depending on your category.
6. Similar segments — Google automatically creates "similar to" segments from your remarketing lists. Add these as signals.

5. PMax + Standard Shopping Together

The best-performing Google Ads setups for D2C brands in India run PMax and Standard Shopping simultaneously — not one or the other. Here's why:

PMax cannibalises Standard Shopping when both exist in the same account — PMax wins the auction for most queries because Google prioritises it. But Standard Shopping gives you control that PMax doesn't: you can set exact bids per product, see exact search terms, and prioritise specific SKUs.

The Recommended Structure

• PMax campaign: Full catalogue, broad new customer acquisition across all channels. tROAS 350–450%. Let it run across all placements.
• Standard Shopping — Priority campaign: Your top 10–15 highest-margin SKUs with manual CPC or tROAS bidding. Set campaign priority to "High" so it wins auctions for these specific products over PMax.
• Standard Shopping — Brand campaign: Brand search terms only. Exclude from PMax at account level. This captures high-intent branded searches with full control.

The Priority setting in Standard Shopping (Low/Medium/High) determines which campaign wins when both are eligible. Set your Standard Shopping campaign to High priority for your best products — Google will run it over PMax for those specific items, giving you control where it matters most.

6. Budget, Bidding & tROAS Strategy

PMax uses smart bidding — either Maximise Conversions or Maximise Conversion Value (with or without a tROAS target). For D2C brands focused on revenue, Maximise Conversion Value with a tROAS target is the right starting point.

Bidding Progression for New PMax Campaigns

Budget minimum: For PMax to learn effectively, it needs at least 50 conversions per month. If your average CPC is ₹20 and your conversion rate is 2%, you need roughly 2,500 clicks — approximately ₹50,000/month or ₹1,600/day minimum to generate learning-level conversion volume. Below this, PMax underperforms.

Don't make major changes during the learning period. Changing bid strategy, tROAS target, budget by more than 20%, or adding/removing asset groups resets the learning period. Make one change at a time and wait 7–10 days before evaluating.

7. Exclusions — Brand, Competitor & Low-Margin SKUs

Exclusions are where D2C brands reclaim wasted spend. PMax without proper exclusions will spend on your brand name, cannibalise organic traffic, and show ads for products where you're losing money on ad cost.

Critical Exclusions to Set Up

• Brand keyword exclusions: Add all variations of your brand name, product names, and misspellings as negative keywords at the account level. Request campaign-level brand exclusions through your Google rep — this is now available as a self-serve feature in Google Ads.
• Competitor brand exclusions: Exclude competitor brand names you don't want to appear for. PMax will otherwise bid on competitor queries where conversion intent is low and CPC is high.
• Low-margin product exclusions: In the asset group product feed, exclude any SKUs where the margin is below your minimum — typically products under ₹299 or items with less than 30% margin. Use product labels in your feed to segment and exclude these at the campaign level.
• Out-of-stock products: Ensure your feed is refreshed daily and out-of-stock items are excluded. PMax will continue spending on products that can't convert.
• Placement exclusions: Exclude irrelevant placements in Display — parked domains, low-quality apps, children's content. Go to Placements → Exclusion list in your account.

8. Reading PMax Reports

PMax reporting is less granular than Standard campaigns, but there's more data available than most advertisers use.

• Insights tab: Shows audience segments, search themes, and top-performing asset combinations. Check weekly. This is your primary view into what the algorithm is doing.
• Asset performance ratings: Each asset gets a Low / Good / Best rating. Replace Low-rated assets every 3–4 weeks. Don't delete — add new ones and the algorithm will deprioritise underperformers naturally.
• Search Themes: Add up to 25 search themes per asset group. These act as soft keyword signals — not exact match targeting, but directional guidance. Use your best-performing search terms from Standard Search campaigns here.
• Auction Insights: Compare your impression share against competitors. Useful for understanding where PMax is winning and losing.
• Campaign-level conversion segmentation: Break down conversions by "New customer" vs "Returning customer" — available in the Segments menu. Critical for measuring new customer acquisition cost separately from repeat purchase ROAS.

9. India D2C Performance Max Benchmarks — 2026

These benchmarks are based on ENZO Digital client data and industry aggregates. ROAS varies significantly based on AOV, product margins, brand recognition, and feed quality. A brand with ₹500 AOV will have different ROAS dynamics from one with ₹2,500 AOV even in the same category.

Saksham Mehra

Founder & CEO — ENZO Digital

Saksham manages performance marketing campaigns for D2C brands across India, the UAE, UK, and Australia — specialising in Google Ads, Meta Ads, and full-funnel acquisition systems. He has run PMax campaigns for D2C brands across fashion, beauty, health, and home decor categories.

LinkedIn Instagram X / Twitter