> For the complete documentation index, see [llms.txt](https://oodlz-ai-studio.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://oodlz-ai-studio.gitbook.io/docs/outbound-agents/awareness/1-topics/competitor-discovery.md).

# Competitor Discovery

Live web research surfaces 12–18 actual content competitors per run — with seed-list fallback if discovery underperforms.

> **Real competitors found by live research — not a static list maintained by hand.**

AI tools that suggest blog topics from training data have no idea who the brand actually competes with for content right now. The Oodlz Topics producer rediscovers the brand's true content competitors at the start of every run via live web research. The output feeds the keyword extraction pipeline directly — every topic that leaves the system traces back to a competitor that is actually ranking.

## How It Works

The discovery step takes the brand profile — name, description, market, audience portrait — and runs a live web search for current content competitors. The result is a freshly-discovered list of 12–18 domains per run, scoped to the brand's actual competitive landscape today, not whatever the seed list captured at onboarding.

If discovery underperforms — fewer than 8 competitors found — the producer falls back to a hardcoded seed list maintained per brand. The fallback never triggers silently; the run log records when it fired and why.

The discovered competitor list is split into 2 sets — one for each topic loop downstream. Both loops query DataForSEO against these competitors for ranked keywords.

## What You Get

* 12–18 competitor domains rediscovered per run — never a stale list.
* Seed-list fallback for resilience when live discovery underperforms.
* Real ranked keywords behind every downstream topic — not AI imagination.
* Run log records every discovery, every fallback, every count.