Most brand safety programs start at the end of the process: buy broadly, then block the words, sites and categories that might cause trouble. That approach catches some risk, but it also blocks perfectly good content and still lets questionable inventory through. This post explains where blocklists fall short, how the quality of your supply changes the problem, and how to build a layered approach that protects the brand without starving campaigns of reach.
What keyword blocklists actually do
A keyword blocklist tells your buying platform or verification vendor to avoid pages containing certain words. It’s easy to set up and easy to explain to stakeholders, which is why almost every brand has one.
The trouble is that words are a poor stand-in for context. A blocklist can’t tell the difference between:
- A news report about a shooting and a review of a camera that “shoots” in low light.
- A health article about cancer screening and harmful misinformation.
- A recipe that mentions “killer” flavor and genuinely violent content.
Lists also tend to grow and rarely shrink. Each incident adds a few more terms, and over time the list blocks large parts of legitimate journalism, sports and entertainment. Advertisers lose scale on quality content, and news publishers, who depend on advertising to fund reporting, lose revenue for doing their jobs.
The risk blocklists don’t see
Overblocking is the visible cost. The less visible problem is what blocklists miss, because many brand safety failures have nothing to do with the words on a page.
Consider where problems typically come from:
- Unknown or misrepresented sites. An impression can be labeled as one domain and actually run on another. A page with no flagged keywords is not the same as a trustworthy page.
- Made-for-advertising sites. Pages built mainly to arbitrage ad spend, with thin content, heavy ad density and traffic bought cheaply. The text may look harmless.
- Invalid traffic. Bots and fraudulent schemes that generate impressions no human sees. Clean keywords don’t mean real people.
- Visual and audio content. Video, images and user-generated content that text analysis can’t fully read.
- Long reseller chains. When an impression passes through many intermediaries, it gets harder to know what you are actually buying.
In each case, the question isn’t “does this page contain a bad word?” It’s “do I know and trust where this ad is going?” That’s a supply question.
Start with curated supply
Curation means deciding up front which publishers and apps are eligible, rather than filtering the open exchange after the fact. A curated marketplace vets its sources before inventory is ever offered to buyers, so the starting pool is already known.
This changes the math of brand safety. When you buy from a smaller set of reviewed, identifiable publishers, you need fewer exclusions to feel comfortable, and your blocklist can go back to being a narrow tool instead of a blunt one.
Vortex is built around this idea: a curated, premium marketplace with no long-tail and no unknown inventory. Supply is checked with multiple layers of fraud detection, and supply paths are transparent, so buyers can see where their ads run.
When evaluating any source of supply, ask practical questions:
- Who reviews publishers before they are admitted, and what gets someone rejected?
- Can I see a list of sites and apps, or at least the domains where my ads ran?
- Are the ads.txt, app-ads.txt and sellers.json entries consistent with what I’m being sold?
- What fraud checks run before the bid, and what happens after?
Safety vs suitability
Brand safety and brand suitability are related but different, and mixing them up leads to bad settings.
Brand safety is the floor: content no reputable advertiser wants to appear next to, such as illegal activity, hate speech or graphic violence. These exclusions are broadly shared across the industry.
Brand suitability is specific to your brand. A spirits brand, a children’s toy company and a financial services firm have different comfort levels with the same content. A hard-hitting investigative piece might be fine for one and wrong for another.
Separating the two helps in practice:
- Keep safety exclusions short, universal and non-negotiable.
- Define suitability in terms of topics and tone, not just keywords, and write it down so planners and partners apply it the same way.
- Revisit suitability settings by campaign. A brand awareness push and a sensitive product launch may need different rules.
- Consider allowing high-quality news by default, with specific exclusions, rather than excluding news as a category.
Creative filtering protects the other side of the page
Brand safety is usually framed from the advertiser’s point of view, but publishers face the mirror problem. A bad ad, such as a deceptive offer, an auto-redirect or malware, damages the publisher’s audience trust and, indirectly, every legitimate brand that appears on the same site.
Creative filtering checks ads before they serve: blocking known bad actors, flagging disallowed categories and catching behavior like forced redirects. Vortex applies creative filtering as part of being brand-safe by design.
Why should advertisers care? Because the environment matters. Sites that protect users from bad ads keep their audiences engaged and loyal, which is exactly the kind of place a brand wants to be seen. A clean page is a better context than one surrounded by misleading ads, even if both pass a keyword check.
Building a layered approach
No single tool solves brand safety. A practical setup stacks several layers, each doing the job it’s good at:
- Supply selection. Start from curated, verified inventory with transparent paths.
- Fraud prevention. Filter invalid traffic before bidding, and monitor after.
- Safety floor. Apply a short, shared list of universally unacceptable content.
- Suitability rules. Layer brand-specific, context-aware settings on top.
- Narrow blocklists. Keep keyword and domain lists for specific, known issues, and prune them regularly.
- Reporting. Review where ads actually ran and adjust based on evidence, not anxiety.
Key takeaways
- Keyword blocklists overblock legitimate content and miss risks that aren’t about words on a page.
- Many brand safety failures come from unknown supply, made-for-advertising sites and invalid traffic.
- Curated supply shrinks the problem before any filter runs, so blocklists can stay narrow.
- Treat safety as a shared floor and suitability as a brand-specific, context-based decision.
- Creative filtering on the publisher side protects the environment your ads appear in.
If you’d like to see what buying from a curated, transparent supply pool looks like in practice, explore Vortex for advertisers.
