Brave Search doesn't see RektRadar yet -- here's why and the fix

Most SEO advice ignores Brave Search. That’s a mistake for a crypto product, where a meaningful slice of the audience has already opted out of Google by default.

We just plugged the Brave Search API into our SEO cockpit and the result was blunt: rektradar.io is invisible there. This post explains what Brave Search is, why it matters more for us than for a generic SaaS, what we found when we queried it, and how the IndexNow protocol gets us into the crawl queue.

What Brave Search actually is

Brave Search is the default search engine inside the Brave browser, which has tens of millions of monthly active users skewing privacy-conscious and tech-savvy — the exact profile of a person who runs their own wallet, reads contract source on Etherscan, and is suspicious of opaque ranking systems.

The important detail: Brave Search runs an independent web index. It is not a wrapper around Bing or Google results. The crawler is Brave’s own, the ranking signals are Brave’s own, and the index decisions don’t propagate from any other engine. That cuts both ways. If you rank well on Google, that does nothing for your Brave Search visibility. If Bing indexes you, Brave doesn’t automatically follow.

For a crypto-native product this matters more than for a typical site. A non-trivial fraction of our visitors block Google trackers, use Brave or Firefox by default, and would rather find us through an engine that doesn’t profile them. If we don’t show up there, we miss them entirely.

The reality check: 0 results

We queried the Brave Search API for the obvious things. The results were unambiguous:

site:rektradar.io — 0 results.
ethereum scam detector — top 10 dominated by ethscamcheck.io, ethereum.org, and academic papers from ACM and ScienceDirect. RektRadar nowhere in the top 20.
rug pull detector, honeypot checker, ethereum token analyzer, crypto scam checker — same picture. Zero visibility.

Compare this to Google, where in the same week we hit position 12 for “ethereum scam detector” and have hundreds of indexed pages. The asymmetry is huge, and it’s structural, not editorial.

Why we’re invisible (and it’s not about quality)

Three things compound against a young domain on Brave Search:

Domain age. rektradar.io is months old, not years. Brave’s crawler heuristics weight discovery toward established hubs and propagate outward through links. New domains sit in a longer queue.
Backlink scarcity. Brave’s index is smaller than Google’s, so it relies more heavily on inbound links from already-indexed sites to discover new ones. We have a handful of mentions on Reddit and Twitter; that’s not enough signal.
No proactive ping. Google has Search Console, Bing has Webmaster Tools, and we use both. Brave Search has no equivalent submission UI. Without a push mechanism, we’re entirely at the mercy of crawl scheduling.

The good news is that the third problem has a solution that also helps with the first two.

The IndexNow protocol

IndexNow is a simple protocol where you POST a list of changed URLs to a single endpoint, and member search engines all get notified at once. Microsoft Bing and Yandex started it. Naver and Seznam joined. Brave Search joined as well — they wrote about it on their own blog, search.brave.com/blog/indexnow.

The mechanics are trivial:

Generate a key (random string), host it at https://rektradar.io/<key>.txt so the engine can verify ownership.
POST { "host": "rektradar.io", "key": "<key>", "urlList": ["https://rektradar.io/scam/PEPE", ...] } to https://api.indexnow.org/indexnow (or the Brave-specific endpoint).
The receiving engines queue the URLs for crawl. No waiting for spider passes, no submission UI to babysit.

A 200 response means the ping was accepted. It does not guarantee indexing — the engine still has to fetch, render, and decide. But it puts us in the queue with a fresh signal, which is the part we couldn’t do before.

How we wired it in

Our SEO cockpit already monitors Google Search Console and Bing Webmaster Tools through dedicated skills (the existing /bing skill is the reference for how we shape these integrations). Adding IndexNow followed the same pattern:

A small daemon watches our Postgres for newly created hub pages (any new contract scan that produces a /scam/<symbol>/<address> URL, plus blog posts and ticker hubs).
On each new URL it batches into a 60-second window, then POSTs the batch to api.indexnow.org.
The response, the URL list, and any errors get logged to OpenSearch so we can audit which submissions went through.
A daily check re-queries the Brave Search API with site:rektradar.io and a handful of priority queries, alerting if the indexed-page count doesn’t move.

The whole thing is maybe 200 lines of TypeScript. The hard part wasn’t the integration — it was figuring out which engines participate (the protocol page lists them but is not always up to date) and confirming Brave actually consumes the same endpoint as Bing rather than running their own. Their blog post settles it: they accept submissions through the standard IndexNow endpoint.

What to expect

Realistically, indexing will not flip overnight. Based on what others have reported, the timeline looks like this:

Days 1-3: First crawler hits show up in our nginx access logs from Brave’s user agent.
Week 1-2: A handful of pages start appearing for site:rektradar.io. Long-tail queries (specific contract addresses, our hub URLs) get hits before broad queries do.
Month 1+: Broader queries like “ethereum scam detector” become competitive only once we accumulate backlinks Brave’s index already trusts. IndexNow gets us into the corpus; ranking still has to be earned.

We’ll track all of it through the same cockpit and write a follow-up once we have real numbers.

Why monitoring multiple engines matters

If you’re building anything in crypto, the search-engine landscape your users actually use is wider than the typical post-mortem suggests. Privacy-skewed audiences split across Brave Search, DuckDuckGo, Kagi, and old-school Google with hardened extensions. Each of these has different index behavior, different ranking signals, and different submission paths.

The cost of monitoring an extra engine is low: an API key, a few hundred lines of glue, a daily check. The cost of being invisible on one of them is the entire user segment that prefers it. For us that’s a target demographic, not a long tail. We’d rather know.

If you missed why we built RektRadar, it covers the original motivation: we got tired of watching the same scam repeat itself, and we built a scanner that reads the chain instead of the marketing. Distribution is the next problem after detection. Brave Search is one piece of it.