When should eco DTC use Demand Gen vs Performance Max?
Quick answer
Use Demand Gen to create visual demand on YouTube, Discover, Gmail, and Display. Use Performance Max to capture purchases once Shopping is stable, including Search and Shopping inventory Demand Gen does not own. Keep branded Search separate. Do not fund Demand Gen by starving a working PMax, and do not let Demand Gen auto-combine planet imagery with unscoped “eco-friendly” lines. PMax go-live lives in Performance Max. Feed hygiene lives in Google Shopping. Organic proof videos live in YouTube consideration.
Vs other Google guides
| Guide | Job |
|---|---|
| Performance Max | When to turn PMax on after Shopping: feed, asset groups, brand controls, TACoS |
| Google Shopping | Merchant Center titles, attributes, disapprovals |
| Branded search defense | Protect name queries; not a discovery campaign |
| YouTube consideration | Organic 6-video proof series. Demand Gen can amplify it; it does not replace it |
| Meta vs Google budget | Channel mix. This guide is which Google campaign type |
| This guide | Demand Gen vs PMax (and leftover Display) for eco DTC |
Pick the job, then the campaign
| Job | Run | Do not |
|---|---|---|
| People are not searching you yet; you need visual reach | Demand Gen | Dump this budget into PMax and hope YouTube “figures it out” |
| People search the category or SKU; you need purchases | PMax (after Shopping is clean) | Use Demand Gen as a Shopping substitute |
| People search your brand name | Branded Search | Let PMax eat brand; Demand Gen cannot defend Search |
| You need longer proof (materials, compare, care) | Organic YouTube series, then Demand Gen on the proof cuts | A 15-second slogan montage with “planet-positive” on the card |
| Old Google Display campaigns | Rebuild in Demand Gen with approved assets | Migrate vague GDN banners and call it a strategy |
Write one sentence on the brief: “Demand Gen creates demand. PMax captures it. Search defends the name.”
When Demand Gen is the go
- A launch or thin-search category where Meta/TikTok cannot be the only discovery surface
- Claim-safe video and images: product truth in the first seconds, scoped proof, no bare eco slogans
- PMax or Shopping already converts enough to fund creation without stealing capture
- You can name the audience (customers, visitors, viewers, lookalikes), not “everyone who likes the planet”
- You will preview asset pairs and freeze anything stronger than the PDP
Demand Gen can attach a Merchant Center feed for visual product ads. That is still not PMax. Feed titles must stay as honest as Shopping. See Shopping hygiene.
When to skip Demand Gen
- Tracking is messy, or you only have stock forests and “sustainable” wordmarks
- PMax is still no-go (feed, claims, or brand Search unfinished). Fix PMax gates first
- The only budget is branded Search or a profitable Shopping/PMax line
- EU creative would fail a Meta claims test. Video is still a claim. See EU green claims
Claim freeze for Demand Gen assets
- Same kit as PDP and PMax. Headlines and descriptions must already live in the substantiation folder.
- Preview pairs. Google will mix images, videos, and text. Reject any pair that reads as carbon-neutral, plastic-free, or “#1 eco brand” unless that line is approved.
- Audience is not a claim. Targeting “environmentally interested” people does not let you say the product is environmentally proven.
- YouTube is spoken copy. Captions, cards, and end screens must match the script. Organic series rules in the consideration guide still apply when you pay to amplify.
- Feed titles. If Demand Gen shows products, “eco” in the title is the same risk as Shopping.
Budget fence
| Line | Role |
|---|---|
| Branded Search | Untouchable fence. Never fund Demand Gen from here |
| PMax / Shopping | Capture. Keep it at a budget that exits limited learning |
| Demand Gen | Named creation line. Size it as a test (often a minority of Google until branded search lift shows up) |
Rebalance Google vs Meta with the channel mix guide, not last week’s Demand Gen ROAS.
Go / no-go and how to measure
| Call | When |
|---|---|
| No-go on Demand Gen | No tracking, no proof assets, or the money would come from brand Search / a still-learning PMax |
| Go on Demand Gen | Capture is stable, claim freeze exists, visuals are product-true, branded Search is funded, you will read lift not last-click |
| Partial go | Retargeting and customer-match only; or US-only until EU assets are scoped |
| PMax instead | The job is purchases from existing intent and Shopping is ready |
Score Demand Gen on branded query lift, later new-customer share, contribution, and claim incidents. Do not bake it off against PMax last-click. Use the ROAS calculator for capture; use incrementality before you scale either.
Creation and capture are not a bake-off
If Demand Gen “loses” to PMax on last-click, that is usually the model working as designed. Kill Demand Gen when it creates claim risk or no branded lift, not when it fails to look like Shopping.
Frequently asked questions
When should eco DTC use Demand Gen vs Performance Max?
Demand Gen for visual demand creation on YouTube, Discover, Gmail, and Display. PMax for purchase capture once Shopping is stable, including Search and Shopping. Keep branded Search separate.
Can Demand Gen replace Performance Max?
No. Demand Gen does not run the Search and Shopping inventory PMax is built for. Replacing a working PMax with Demand Gen is a capture cut, not an upgrade.
Should we migrate Google Display campaigns into Demand Gen?
Rebuild them. Move only approved, product-true assets. Do not lift-and-shift old GDN banners with unscoped eco lines.
How do we keep Demand Gen claim-safe?
Feed it the same approved lines as PDP and PMax, preview asset combinations, and treat audience names as targeting, not product claims.
How is this different from the YouTube consideration guide?
That guide is the organic proof series. This guide is the paid Google campaign-type choice. Amplify the series with Demand Gen; do not skip the series.