Discover New Demands From Outbound Links

Keyword tools show you demand people already search for. But some of the best product intelligence hides somewhere else: in the outbound links of high-traffic sites. Every new website that wants users must go out and leave links across the internet — and those links are a public trail leading straight to the new products currently being promoted.

(Figures in this article are as mentioned in the original source and have not been independently verified.)

1. The core idea: promotion leaves a trail

A new product cannot promote itself in silence. It posts on communities, submits to directories, buys mentions, answers questions — everywhere it goes, it leaves a link. High-traffic sites accumulate these outbound links at scale, which makes their outbound-domain list a radar screen: scan it, and you see which new sites and new products are actively pushing for attention right now.

The original traces this method to a concrete trigger: an article about a Russian photo app that hit 4.8 million visits in 20 days mentioned that two sites, pikabu.ru and vc.ru, were driving it serious traffic. Checking vc.ru’s outbound domains in Semrush surfaced many new sites — the method proved itself on first use.

2. The manual way, and why it hurts

Done by hand, the method works like this: take an outbound domain, check its whois record for registration time, check Similarweb for its traffic, and judge whether it is a genuinely new, already-moving product. Repeat for every domain on the list.

The pain is obvious — copying domains one by one and querying each data source separately is slow, and the list is long. The value is real, but the workflow does not scale. That is exactly why the original proposes turning it into a pipeline.

3. The pipeline: five steps, batched

The proposed automation chains five data sources in order:

1. Semrush — pull the high-traffic site’s outbound domains and save them in bulk.
2. Whois — batch-query registration times to find genuinely young domains.
3. Similarweb — pull traffic data to see which ones already have volume.
4. Archive.org — check indexed history to confirm the site is real and established enough to have snapshots.
5. Google Search API — pull additional related information for context.

Then filter the combined table with conditions like “domains registered in July, indexed by Google, archived, traffic above X,” and sort by traffic or by registration time. What comes out is a ranked list of products that are just launched, actively promoting, and already getting traffic — found in bulk, not one by one.

4. What the method is good for

Three uses, in increasing order of leverage:

Demand research. See what other builders are doing and which demands already carry traffic. A new site with fast-growing visits is a demand signal with a timestamp — someone is betting on this need right now.

Competitor analysis. Discover competitors you never knew existed. If a rival product is being promoted through the same channels, its links will show up in the same outbound lists — an early-warning system for your niche.

A product idea in itself. A community member in the original discussion (damo) took it one step further: the whole pipeline could become a service, sold to every developer who needs promotion. The method for finding demand becomes a product serving demand — a neat recursion worth remembering.

5. The toolchain, summarized

The full loop in one line: Semrush outbound domains → whois registration time → Similarweb traffic → Archive.org history, filtered and sorted across registration time, indexing status, and traffic volume.

Build it as a spreadsheet first, automate it second, and productize it only if the manual version keeps paying. Like every method in this series, its power is not in the tools — it is in the habit of reading the internet’s public exhaust for demand that is already moving.

Series: Build Overseas Websites with GeFei

This article is based on GeFei’s shares, reorganized in our own words. Source: https://github.com/qiayue/opengefei