Workflows

Overview

Obsurfable isn't a bag of disconnected tools — it's a workflow for improving how AI answer engines find, cite, and recommend your brand.

Every feature lives in one of four phases:

These map directly to the sections on your Dashboard and to deeper documentation sections (Setup, Monitor, Act, Discover) in this docs site.

Think of it as a loop: Setup → Monitor → Act → Discover → Monitor again.

Who it's for

Workflows help any Obsurfable user build a repeatable AI visibility program:

• Solo marketers who need a clear weekly rhythm.
• Cross-functional teams where Setup is owned by product marketing, Monitor by SEO, Act by content, and Discover by community.
• Agencies running the same playbook across client accounts.

Prerequisites

A signed-in Obsurfable account. For full workflow coverage (especially Act and advanced Monitor), a Pro plan is recommended. Free users can complete Setup and limited Monitor (prompt retrieve).

The four workflows in detail

Setup — brand context

Goal: Give Obsurfable accurate, rich context so prompts, insights, and generated content reflect your real positioning.

When to use it: At onboarding, after a rebrand, when entering a new market, or whenever AI answers drift from reality.

Key features:

• Company — Name, URL, overview, goals, and audience.
• Keywords — Category terms buyers use in AI queries.
• Use cases — Outcome-oriented scenarios ("reduce churn", "automate invoicing").
• Competitors — Alternatives AI might recommend instead of you.
• Sitemaps — URL inventory for coverage and site analysis.
• llms.txt — Machine-readable intent for AI crawlers.

Outcome: Higher setup health score, better auto-generated prompts, and more accurate retrieval results.

Typical cadence: Review monthly or after major positioning changes.

Monitor — observe and measure

Goal: See how AI answers questions in your category today — and where your site falls short.

When to use it: Weekly or on a schedule (Pro automations) to track visibility over time.

Key features:

• Visibility score — Roll-up of mentions, citations, and competitive gaps.
• Prompts — Run buyer questions through AI retrieval; check mentions and citations per prompt.
• Queries (Pro) — Build query fan-out trees; map sub-questions to URL coverage with gap analysis.
• Site analysis (Pro) — Crawl your sitemap/llms.txt URLs; report on structure, thin clusters, and retrieval readiness.
• Settings / Automations (Pro) — Schedule prompt re-runs and monitoring jobs.

Outcome: A prioritized list of gaps — missing citations, weak coverage nodes, structural issues — feeding Act and Discover.

Typical cadence: Weekly visibility check; monthly deep dive with Query trees and Site analysis on Pro.

Act — improve visibility

Goal: Turn monitoring insights into published artifacts that shift how AI retrieves and cites you.

When to use it: After Monitor surfaces specific gaps — missing comparison page, thin FAQ cluster, inaccurate entity description.

Key features:

• Insights (Pro) — AEO/GEO recommendations from prompt runs: meta, content, page, comparison, messaging, and more.
• Content (Pro) — Generate blog posts from insights; export Markdown/HTML or publish on your subdomain.
• Generate response (Pro) — Draft on-brand replies for social threads where your category is discussed.

Outcome: New or improved pages, posts, and replies that answer buyer questions in an AI-retrievable format.

Typical cadence: Sprint-based — batch content after a Monitor review; publish, then re-run prompts to measure lift.

Discover — external context

Goal: Find conversations, sources, and research on the open web that influence how AI systems understand your category.

When to use it: When Monitor shows competitor citations from third-party sites, or when you want community presence beyond owned content.

Key features:

• Conversations (Pro) — Search Reddit, Quora, LinkedIn, and YouTube for threads matching your keywords and use cases.
• Scrapbook — Save URLs and notes from anywhere on the web.
• Annotations — Central hub for notes attached to prompts, insights, queries, conversations, and more.

Outcome: A research library and engagement targets — places to participate, cite, or learn what narratives AI may be absorbing.

Typical cadence: Ongoing research; spike activity when launching new positioning or competitive campaigns.

Step-by-step: run the full workflow loop

Week 1 — Setup and baseline

1. Sign up free and complete onboarding.
2. Fill Setup: keywords, use cases, competitors, sitemap, llms.txt.
3. Run all company prompts once (Free) to establish baseline visibility.

Week 2 — Monitor and prioritize

1. Review visibility score and citation gaps on the dashboard.
2. Upgrade to Pro if you need insights, query trees, or re-runs.
3. Build 1–2 query fan-out trees for your highest-value topics.
4. Run site analysis to find structural issues.

Weeks 3–4 — Act

1. Open Insights from your best prompt runs.
2. Generate content for top granular recommendations.
3. Publish on your site or Obsurfable subdomain blog.
4. Draft social responses for high-intent threads (Generate response).

Ongoing — Discover and re-measure

1. Search Conversations for category discussions weekly.
2. Save useful sources to Scrapbook; annotate surprising findings.
3. Re-run prompts (Pro) after each meaningful publish.
4. Track visibility score trend month over month.

How workflows connect

flowchart LR
  Setup --> Monitor
  Monitor --> Act
  Monitor --> Discover
  Act --> Monitor
  Discover --> Act

• Setup feeds Monitor — without keywords and competitors, prompts miss the mark.
• Monitor feeds Act — insights and gaps become content tasks.
• Monitor feeds Discover — competitor citations point to external sources worth studying.
• Discover feeds Act — conversation research informs replies and content angles.
• Act feeds Monitor — new pages change retrieval results on the next run.

Tips

• Don't skip Setup after onboarding — Sitemap and llms.txt dramatically improve Site analysis and query URL matching.
• Monitor before you Act — Publishing without a gap list wastes effort; let prompts and insights prioritize.
• One workflow per session — Monday Monitor, Wednesday Act, Friday Discover beats context-switching across all four.
• Use Annotations as team memory — Link notes back to prompts and insights so context isn't lost between sprints.
• Close the loop — Every Act sprint should end with a prompt re-run on Pro.

FAQ

Do I have to follow all four workflows?

No — but each phase addresses a different failure mode. Setup-only gives accurate context but no measurement. Monitor-only shows gaps but doesn't fix them. Most successful teams run the full loop at least monthly.

Which workflow should I start with?

Always Setup, then Monitor with your first prompt retrieval. See Quick start.

Is the Visibility Director agent part of a workflow?

The Visibility Director agent (Pro) orchestrates across workflows — monitoring, gap analysis, content drafting, and publishing in one conversation. It's documented in the Agent section of these docs and complements the four manual workflows.

Where do Insights fit — Monitor or Act?

Insights are generated from Monitor data (prompt runs) but exist to drive Act. You'll often bounce between Monitor results and Insight recommendations in the same working session.