How should buyers judge new websites for advertising fairly?
Part of Lookalike websites
The short answer
Judge a new website on what it already shows, not on the history it has not had time to build. Check its editorial transparency, ad labeling, authorized sellers and page layout on day one, record missing traffic data as missing, watch its publishing over several weeks, then test it in its own line with a small, capped budget before scaling.
Key takeaways
- Thin history is a gap in the evidence, not a mark against the site; the checks that need no history still work on the first day.
- Editorial transparency, ad labels, ads.txt and sellers.json, and the page layout can all be judged before a new site has any traffic record.
- Public popularity datasets and forecasts leave out small and new sites by design, so their absence is recorded as missing.
- A test belongs in its own ad group or line item with placement targeting and a capped budget, so its results can be read site by site.
- New and established publishers are reviewed against the same written criteria, with only the history requirement set aside.
How should a buyer evaluate a new website?
A lookalike search or a discovery trail sometimes turns up a site that launched this year. It reads well, matches the seed and has no campaign history with the agency. The temptation is to treat it like an established site with a poor record. That gets the evidence wrong: a new site has no record, which is different from a bad one. The lookalike websites hub covers how such sites surface; this guide covers what to do once one does.
Definition
- New publisher
- A website whose editorial record, traffic data and buying history are too short to judge by track record, whatever the age of its domain. A new outlet on an old domain and a new domain from an established team both count.
| Evidence | Visible on day one | Builds over weeks | How to treat its absence |
|---|---|---|---|
| Editorial transparency | About page, named editors, bylines and dates | A visible publishing rhythm | Missing transparency is a real finding, not a timing issue |
| Ad labeling and layout | How sponsored content is labeled and how ads sit on the page | Whether ad load changes as traffic grows | Check it now and again before scaling |
| Selling path | ads.txt lines and matching sellers.json entries | Additional exchanges added over time | Hold until the file exists if the plan requires authorized sellers |
| Audience data | Absent until the site passes a dataset's threshold | Public popularity data once a threshold is passed | Record as missing; never score as zero |
| Campaign history | None | Test results from a small, separate buy | Run a test rather than assuming the result |
Which checks still work when a site is new?
Four kinds of check depend on what the site publishes, not on how long it has been publishing. Google's news transparency guidance (opens in a new tab) asks news sources for "Clear dates and bylines" and information about the authors, publication and publisher. The FTC's native advertising guide (opens in a new tab) says an advertisement "shouldn't suggest or imply to consumers that it's anything other than an ad". The ads.txt specification (opens in a new tab) defines who may sell the space.
Day one checks for a new publisher
- The about page names the people or company behind the site, and recent articles carry bylines and dates.
- Sponsored articles carry a label such as Ad, Advertisement, Paid Advertisement or Sponsored Advertising Content, terms the FTC guide lists as likely to be understood, placed in front of or above the headline.
- ads.txt is at the root domain, and its DIRECT lines resolve to the site's owner in the exchanges' sellers.json files.
- Five recent articles are original work on the site's stated subject, not rewrites that a quoted sentence search finds elsewhere.
- Ads leave room for the article on a phone and on a desktop, without auto refreshing slots or pop ups over the text.
- The site's subject still matches the seed it was found from, across the sampled articles.
The ads.txt check reads two account types. In the specification's words, "A value of 'DIRECT' indicates that the Publisher (content owner) directly controls the account", while RESELLER means the publisher has authorized another entity to control the account and resell its space. A new site sold only through a reseller is not suspect for that reason; it is a site that has not yet set up its own accounts. Domain registration dates are a separate clue, covered in the domain age guide.
How do you avoid penalizing new independent publishers?
Two planning datasets leave new sites out by design. The Chrome UX Report (opens in a new tab) includes an origin only when it is publicly discoverable and sufficiently popular, and says "An exact number is not disclosed." Display & Video 360's Inventory Availability report (opens in a new tab) "may not display data for targeting combinations that match fewer than 3,000 unique cookies." A rule that drops sites missing from these sources drops new and small publishers first.
The fairness case is practical as well as principled. A new specialist site that matches the seed closely is the kind of find lookalike work exists for, and it is also a site that can be missing from popularity data. The small websites guide and the fair publisher planning hub cover how to present such sites to a client.
How long should a new site be watched before scale?
A fixed number of months would penalize good sites without catching bad ones. The useful wait is for specific evidence: a steady publishing rhythm, an ad layout that holds as traffic grows, sellers that stay consistent, and a test with enough delivery to read. Google gives one reference point for the test itself: when an experiment shows no clear winner, it recommends "allowing your experiment to run for 2 to 3 weeks to gather data."
Google's experiments page (opens in a new tab) adds that if results are still undecided, "you may need to increase your budget or allow it to run for longer". For a new publisher that means an inconclusive test is a reason to continue or enlarge it, not a verdict on the site.
How do you test a new publisher with a small budget?
Five steps for a small, readable test
- Give the new sites their own ad group or line item. Use placement targeting so the budget goes to the chosen sites, and keep the established list in a separate line.
- Cap the budget knowing how caps work. Google Ads' average daily budget (opens in a new tab) allows a daily spending limit of "two times your average daily budget for most campaigns" and a monthly limit of "30.4 times your average daily budget for most campaigns".
- Use an experiment where it fits. Google's custom experiments (opens in a new tab) are available for Search, Display, Video and Hotel campaigns, and Display experiments always use cookie splits.
- Read results by site. In Display & Video 360 reports, the App/URL dimension (opens in a new tab) is "The name of a site, or mobile app", so each new site's delivery can be read on its own.
- Review against the written criteria. Apply the same viewability, suitability and quality standards used for established sites, and record the decision.
Keeping the test apart matters for both sides. Mixed into the main line, a new site's delivery is hard to read and can be crowded out by larger sites. Kept apart, the result shows whether this publisher works for this brief. The adjacent audiences guide describes the same separation for testing nearby topics.
How does PlacementGraph treat sites that are new or missing data?
What does a fair new publisher test look like on a real brief?
Fictional example
A fictional sneaker shop in Alaska
An unnamed Anchorage sneaker and streetwear shop asks for lookalikes of a sneaker culture blog it likes, for a spring release campaign reaching readers in Anchorage, Fairbanks and Juneau. The search returns established sneaker sites and three that launched in the past year: a release calendar site, a sneaker restoration blog and a streetwear news site.
The day one checks split them. The restoration blog has named writers, dated original articles and DIRECT ads.txt lines that match the exchanges' records. The release calendar site has no about page and labels paid posts with a logo only, which the FTC guide says is generally not enough, so it is left out with that reason. The streetwear site passes the checks but has no ads.txt file yet, so it is held.
The restoration blog and the established sites go into separate line items, the new site's line with placement targeting and a capped budget. Its audience shows as no data yet in the plan. After the test, the planner reads its delivery by site and reviews it against the same criteria as the established sites.
Frequently asked questions
Does a new website need an ads.txt file before it can be tested?
It needs one before a buyer can confirm who is authorized to sell its space. A plan that requires authorized sellers should hold a new site until the file appears at the root domain and its DIRECT lines match the exchanges' sellers.json entries. A missing file is a reason to wait or to ask the publisher, not a judgment on the content.
Can a new publisher join a private marketplace deal?
Yes, if it sells through the exchange that hosts the deal and passes the plan's checks. A planner can test a new site on its own line first so its delivery can be read separately, then add it to the deal's site list once the test review is written down.
Should the client be told which sites on a plan are new?
Yes. Label new publishers in the plan with the checks they passed, the data that is missing and the test that is under way. A client who sees why a new site is there, and how it is being tested, can judge it on that evidence while early reporting is thin.
What if a new site's traffic jumps suddenly?
Look at where the visits come from before scaling. The ANA's 2023 MFA definition lists a high percentage of paid traffic sourcing among the traits of made for advertising sites. A jump can follow a story that spread; a jump driven by sponsored link units pointing to the site deserves a closer look.
Sources
- Google News policies: transparency (opens in a new tab)Google Publisher Center Help
- Native Advertising: A Guide for Businesses (opens in a new tab)Federal Trade Commission
- ads.txt Specification Version 1.1 (opens in a new tab)IAB Tech Lab
- CrUX methodology (opens in a new tab)Google Chrome for Developers
- Inventory Availability report (opens in a new tab)Display & Video 360 Help
- Monitor your experiments (opens in a new tab)Google Ads Help
- About average daily budgets (opens in a new tab)Google Ads Help
- Set up a custom experiment (opens in a new tab)Google Ads Help
- Dimensions in reports (opens in a new tab)Display & Video 360 Help
- Industry leaders define made for advertising inventory (opens in a new tab)ANA
Researched and written by PlacementGraph Research. Read how we research, source and update these pages. Corrections: diana@placementgraph.com.