SEO · Keyword Discovery Automation

Your tools show you
their index. Not yours.

Keyword Discovery Automation treats every query your site appeared for as a tracked keyword. There is deliberately no minimum-impression filter anywhere in the system, so you see the whole long tail Google actually served you, including the thousands of queries nobody thought to put on a watchlist.

Book a walkthroughAll of SEO
Keyword Discovery

What do we actually rank for?

34,062 queries, first-party
No minimum-impression filter
Long tail included, not trimmed
Piper, SEO Lead
Piper · SEO Lead
34,062 queries on our own site. Every one Google served us, not a sample.
No thresholds anywhere. A query with three impressions is still a query you rank for.
Brand fit, scored 0 to 100. Which of those rankings actually serve the business.
Two of our top queries were junk. "mercor data breach" and "kiro outage". Real traffic, no commercial relevance.
RunningStatus today
PiperAgent

Part of the SEO module. Switched on and off per account, so you never pay for a module sitting idle.

This automationRunning
How It Works

Four moves. Every one of them yours to stop.

1
It reads your own Search Console, not an index

Third-party tools can only show you keywords inside their own crawl. This reads what Google actually served your domain.

2
Nothing is filtered out

No minimum impressions, no minimum clicks. The full long tail is the point, because that is where the queries nobody guessed live.

3
Each query is scored for brand fit

An AI score, 0 to 100, of whether ranking for it actually serves your business. This is what stops vanity traffic reaching the work list.

4
What survives becomes the work list

Discovery feeds rank tracking and on-page work. Finding the query is step one, not the deliverable.

What You Get

The market, and what changes.

0US searches a month for keyword research tool
34,062queries tracked on our own site, with no threshold filter
0 minimumsa query with three impressions is still a query you rank for

Search volume: Google Keyword Planner, United States, pulled 31 Aug 2026. Every other number is measured on our own site.

The System Behind It

Keyword tracker

Self-hosted, reading your own Search Console. On Ruh AI it tracks 34,062 distinct queries with no threshold filter, and its brand-fit score is the reason two high-traffic news queries never reached the work list.

Running on live databuilt for Ruh AI firsthuman approval on every piece
See the numbers it produced ↗
Keyword tracker
Your dataYour brand packWhat worked last month
Piper
Piper · Keyword tracker
Held for your clickrender · reviewapprove · or rejectthen it ships
Honest Limits

What this does not do yet.

Single site today. One business per install, and running many sites at once is not yet proven.
The sync is manual. Ours went stale for over two weeks in August, and that is exactly the kind of thing this page should tell you.
It reads what Google reports. Queries Google withholds are invisible to it, the same as to every other tool.
Discovery is not ranking. Finding the query is the cheap half.
Questions

Asked before every walkthrough

How is this different from Ahrefs or Semrush?

Those show you keywords in their own index. This reads your own Search Console, so it sees the full long tail Google actually served you, including queries no crawler would guess.

Why no minimum-impression filter?

Because the filter is where the interesting queries get thrown away. The full first-party long tail is the differentiator.

Does it work for more than one site?

Not yet. Single site per install today, and we will say so rather than imply multi-tenancy that is unproven.

Invite Only

Switch on Keyword Discovery Automation.
Leave the rest off.

Book a walkthroughSee the Ruh AI numbers

Also in SEO: Rank Tracking Automation · On-Page Automation