Exclusion lists are reactive: your ads run somewhere, you review it, you remove it. An inclusion list turns that around. You name the sites your ads may appear on, and nothing else gets bought. People also call it an allowlist, or a whitelist in older material. Same idea.
It is the right tool less often than people expect, and Google supports it in fewer places every year. This guide covers where it still works, how to pick the sites, and how to keep the decisions defensible six months later.
Checked against Google's documentation in September 2026.
Where positive placement targeting still exists
| Campaign type | Can you name the sites? |
|---|---|
| Display campaigns | Yes. Managed placements let you target specific websites, and you can run a campaign against those alone |
| Video campaigns | Yes, for YouTube channels and videos |
| Performance Max | No. There is no website allowlist. You can exclude, and you can set content suitability, but you cannot restrict it to a list of sites |
| Demand Gen | Partly, and this is worth checking in your own account. Google's migration material says contextual targeting carries over, while its Demand Gen FAQ says topic and placement exclusions are not generally available at campaign or ad group level |
Two consequences follow. If most of your budget sits in Performance Max, an inclusion list is not available to you, and the practical lever is the account-level exclusion list. And because standalone Display campaigns are moving into Demand Gen, treat any inclusion list you build now as something to re-verify after your campaigns migrate.
The trade you are making
An inclusion list buys control and costs reach. A list of 200 sites is a fraction of the inventory Google would otherwise find, which usually means higher CPMs, slower learning and less volume. That is a reasonable trade for a regulated advertiser, a brand with a specific adjacency problem, or a campaign where one bad screenshot costs more than the incremental reach is worth. It is a poor trade for most direct-response accounts.
The middle path is usually better: run broad with a serious exclusion list, and keep a small inclusion-targeted campaign for the placements you actively want.
Where candidate sites come from
- Your own placement report. The sites that already perform are the least speculative candidates you have. Pull the report, sort by conversions rather than by spend, and start there.
- Your customers' media diet. Trade press, professional communities and the publications your buyers actually read.
- Publisher availability. A site you love is useless if it sells no inventory through the networks you buy on. Check that your ads can actually serve there before spending a review on it.
- Language and geography. A strong site in the wrong market is a candidate for a different campaign, not this list.
What to check before a site goes on the list
For each candidate, answer these. The scorer does the first three at volume; the rest need a person.
- Is it a real publication? A named author, an editorial process, a contact route, and content that existed before programmatic advertising did.
- What is the ad load? Count the slots above the fold. Heavy stacking depresses attention for everything on the page, including you.
- Does the content hold up? Open five articles. Scraped, spun or machine-generated filler is easy to recognise once you look.
- Is it suitable for your brand specifically? Safety is a floor everyone shares. Suitability is your judgement, and it changes by campaign and by week. Our suitability policy template is a way to write that down before you argue about it.
- What would this look like in a screenshot? If the answer makes you hesitate, that is the answer.
Record the decision, not just the domain
The reason inclusion lists rot is that nobody can remember why a site is on them. Six months later someone asks, and the list gets either defended blindly or thrown out. Record the evidence at the moment you decide.
| Domain | Category | Why it is a candidate | Reviewed by | Reviewed on | Decision | Next review |
|---|---|---|---|---|---|---|
| example-publisher.com | Trade press | Audience matches our buyer, editorial team named, 2 ad slots above the fold | 2026-09-20 | Include | 2026-12-20 | |
| example-review-site.com | Reviews | Genuine reviews, heavy affiliate load, check landing experience | 2026-09-20 | Test | 2026-10-20 |
Download the worksheet as CSV. It opens in any spreadsheet, and nothing you type goes anywhere near us.
Test it before you commit the budget
Run the inclusion-targeted campaign alongside your existing one rather than replacing it. Give it enough budget to leave the learning phase, and compare cost per conversion and revenue per visit rather than clicks, because the list will cut volume by design.
One caution on reading the result. In our own comparison of a curated list against an uncurated one, the curated campaign produced 52% fewer visits and 7x more revenue per visit. The two campaigns had different optimisation goals, so that is an observed difference and not proof that curation alone caused it. Measure your own.
Keep it alive
Quarterly is a realistic cadence. Publishers change owners, ad loads creep up, and a site that was right in March can be wrong in September. Put the next review date in the sheet, and re-check anything that has been on the list for a year without anyone looking at it.
If you would rather have this done for you, a managed analysis includes building the list and documenting the evidence behind each decision. If you only need the sites you already run on reviewed, we audit your top 100 placements for free.